WO2012098544A2

WO2012098544A2 - Improved data entry systems

Info

Publication number: WO2012098544A2
Application number: PCT/IL2012/000029
Authority: WO
Inventors: Benjamin Firooz Ghassabian
Original assignee: Keyless Systems, Ltd.
Priority date: 2011-01-19
Filing date: 2012-01-19
Publication date: 2012-07-26
Also published as: WO2012098544A3

Abstract

A data entry system having a plurality of keys to which substantially all of the letters of an alphabet are assigned based on their common characteristics in their shapes. The letters are divided in a plurality of groups of letters based on the characteristics. In order to enter a word the user interacts with the appropriate keys. A word predictive system predicts the desired word based on the key interactions.

Description

IMPROVED DATA ENTRY SYSTEMS

RELATED APPLICATIONS

The present application claims priority from Israeli patent applications:

210,749 filed 19-Jan-2011

210,791 filed 20-Jan-2011

210,894 filed 26-Jan-2011

210,981 filed 31-Jan-2011

211,102 filed 06-Feb-2011

211,338 filed 21-Feb-2011

211,798 filed 17-Mar-2011

211,887 filed 23-Mar-2011

211,962 filed 27-Mar-2011

212,281 filed 12-Apr-2011

212,430 filed 17-Apr-2011

212,478 filed 26-Apr-2011

212,617 filed 01-May-2011

212,661 filed 03-May-2011

212,681 filed 04-May-2011

212,831 filed ll-May-2011

213,021 filed 19-May-2011

213,046 filed 22-May-2011

213,333 filed 02-Jun-2011

213,641 filed 19-Jun-2011

213,914 filed 03-Jul-2011

214,104 filed 14-M-2011

214,471 filed 04-Aug-2011

214,494 filed 07-Aug-2011

214,581 filed 10-Aug-2011

214,737 filed 18-Aug-2011

215,087 filed ll-Sep-2011

215,223 filed 18-Sep-2011

215,446 filed 27-Sep-2011

215,605 filed 06-Oct-2011 215,718 filed ll-Oct-2011

216,020 filed 27-Oct-2011

216,045 filed 30-Oct-2011

216,170 filed 07-Nov-2011

and claims the benefit, under 35 USC 119(e), of US provisional application

USSN 61556389 filed 07-Nov-2011

USSN 61560883 filed 17-Nov-2011

USSN 61562691 filed 22-Nov-2011

USSN 61577041 filed 18-Dec-2011

USSN 61582427 filed 02-Jan-2012

USSN 61584376 filed 09-Jan-2012

the disclosures of which are all incorporated herein by reference in their entirety.

This application is also related to various prior PCT publication, including WOO 1/131788, titled "Integrated Keypad System", WO03/007288 titled "Features to Enhance Data Entry Through a Small Data Entry Unit", WO04/095414 titled "System to Enhance Data Entry in Mobile and fixed Environment", WO05/122401 titled "System to Enhance Data Entry in Mobile and fixed Environment", WO07/114833 titled "Data Entry System", WO08/114086 titled "Combined Data Entry Systems", WO09/027817 titled "Improved Data Entry System", WO10/018577 titled "Data Entry System", WO10/089740 titled "Data Entry System", and WO11/073992 titled "l-eaturcs of a data cntr> s em^". the disclosures of all of which are also incorporated herein, by reference, in their entirety.

The titles used in this application and in the related applications may preferably not been considered as part of the descriptions.

FIELD OF THE INVENTION

The present invention relates to data input devices and particularly to devices and methods for entering letters and other symbols.

BACKGROUND OF THE INVENTION

Mobile devices including cellular phones, personal digital aids and miniature computers are widely used as they are easily carried around and can perform many tasks. One problem of the mobile devices is text entry, which is problematic due to their small size. One solution for text entry in mobile devices is using a limited key set in which at least some of the keys are ambiguously associated with a plurality of letters. Upon receiving a sequence of key strokes, the system a word of a dictionary/database. A word predictive system based on ambiguous key presses is used on mobile devices. Another problem of mobile devices is entering symbols other than the alphanumeric characters

Many mobile devices receive user input through a touch screen. Touch screens are also used on non-portable devices.

According to another method, in order to enter one of the symbols (such as a letter) among the group of symbols assigned to a key, the user may provide a predefined interaction with the key such as a pressing action on the key and provide a speech information corresponding to the symbol (such as speaking the appellation of the letter) for selecting the letter among the letters that are assigned to the key, and wherein the speech information may be detected and analyzed based on at least one of, the user's voice and/or the user's lip movements. According to one method symbols such as letters having ambiguously resembling predefined speech and that are hard to distinguish from each other through their corresponding speech may be separated from each other being assigned to different keys.

The data entry systems providing precise characters may be combined with the principles of a word predictive system based on key presses alone, to provide a highly accurate data/text entry system using few keys. The principles of word predictive systems based on key presses alone (e.g. T9) are known by people skilled in the art.

Methods for text entry using word predictive systems in which at least one letter is input unambiguously and other letters can be input ambiguously, are known. WG 2009/027817 (to Ghassabian) describes a data entry system that uses a system having two keypads. The first keypad has all the letters of an alphabet distributively assigned to a small number of keys (e.g. four) such that at least two of the letters are assigned to at least one of said keys of the first keypad keys. The second keypad (for example a telephone-type keypad) generally also includes all of the letters of the alphabet of the language, distributively assigned to at least some of its keys such that at least two of the letters are assigned to at least one of the keys of the second keypad set of keys. Typically, the letters are distributed such that an ambiguous key of the first keypad and an ambiguous key of the second keypad have preferably at most one common letter/character. For entering a letter precisely, the user may first press on the key of the first keypad to which said character is ambiguously assigned. The system may predict an undesired character. The user then may press on the key of the second keypad to which the desired character is also ambiguously assigned. As described, said pair of keys may have at most one common character which in this case is the desired character. The system thus provides said the desired character. For entering a word the user uses/taps-on the corresponding keys of the first keypad and the system predicts a corresponding word (if the word is in the dictionary), or a chain of characters (if the word is not in the dictionary). A correction procedure may be performed by the user when the system offers (predicts) an undesired word, in which the first keypad is used for character entry and the second keypad is used to replace one (or more) ambiguous character of the predicted word by one (or more) precise character (to correct a wrongly predicted set of characters). By using key-presses of two keypads, the data entry system may provide precise character identification, with the unique character common to the two keys being unambiguously selected thereby. By considering the ambiguous characters and the precise character(s) provides as described, the system may propose another word to the user.

Word predictive systems based on key press information use at least a dictionary of words that may also include or use a database the corresponding key presses for each word.

Data may be input through means other or additional to key presses. For example, a user may perform gliding actions upon a touch sensitive screen to duplicate or imitate key interactions.

Most mobile devices having touch screen use an on-screen QWERTY type keyboard to enter text and data. Such keyboard is not adapted to be on a screen. The screen of a device, is intended to display the output. Having a full keyboard with many keys on the screen covers a significant portion on the screen. This hinders using many current application and limits creation of many other applications.

SUMMARY OF THE INVENTION

In order to resolve the problem, a system freeing the screen from the input interface must be created. Said system must be intuitive, accurate, fast, and easy to understand and use.

An aspect of some embodiments of the invention relates to an input interface adapted to identify user interactions and to associate at least some of the user interactions with characters. A small number of the user interactions are associated ambiguously with all the letters of the Latin alphabet or another alphabet. The letters are assigned to the user interactions according to their shapes, in a manner which allows the user to quickly translate the shape of a letter to the interaction with which it is associated. In some embodiments, one of the user interactions is associated with all the letters meeting a specific shape feature.

Optionally, the letters are divided into several groups (e.g. four groups) based on a common characteristic in their shapes and wherein each group is assigned to a different user's interaction or to a different key of a keypad.

In some embodiments, the letters are assigned to the user interactions/keys according to w cu t.i im v a c ose c rc e, s an ng on one p n , s an ng n wc g n a large base.

An aspect of some embodiments relates to, one or more of the user interactions/keys is associated with letters having a specific shape characteristic in common. Said shape characteristics may be in their uppercase shape and/or in the lowercase shape. For example, in one embodiment, one of the user interactions or key may be associated with letters having a substantially closed circle in their (e.g. uppercase) shape, i.e., with some or all of B, D, O, P, Q, and R. Alternatively or additionally, one of the user interactions may be associated with letters standing on one point, i.e., with some or all of F, I, J, T, V, and Y. Alternatively or additionally, one of the user interactions may be associated with letters standing on two points, i.e., with some or all of A, H, K, N, M, W, and X. Alternatively or additionally, one of the user interactions may be associated with letters standing on a large base, i.e., with some or all of C, E, G, L, S, U, Z.

Optionally in some of the shape characteristics may have priority over the others. For example, the closed circle characteristic may have priority over the other shapes. As an example, the letters R, and P, may preferably assigned to the group of letters having a closed circle characteristic in their shape, even though said letters R, and P, may also being assigned, respectively, to the group of letters standing on two points, and standing on one point.

Optionally, a single letter may be assigned to two different groups. As an example, the letter "R", has a substantially closed circle in its shape and also stands on two points may be assigned to the two corresponding groups. Also as an example, the letter "P" may be assigned to the group of letters having a substantially closed in their shape and may be also assigned to the group of letters that stand on one point.

In accordance with some embodiments, the user interactions are each assigned a shape rule and each user interaction is assigned the letters meeting its rule. Optionally, the rules are given an order of precedence, such that if a letter meets two shape rules it is assigned to the user interaction having the rule with a higher precedence.

In some embodiments, four user interactions or keys are associated with all the letters of an alphabet such as the Latin alphabet. A first user interaction/key is assigned closed two leg letters with highest precedence. A second user interaction is assigned large arc letters with next to highest precedence. A third user interaction is assigned letters with a top or bottom horizontal bar and a lowest precedence interaction is assigned letters with an odd number of contact points with the bottom line.

In some embodiments, a small number of variations relative to the shape rules are allowed in order to allow for other considerations. For example, variations from the shape rules may be used to avoid letters assigned to a same user interaction in a second assignment of letters to user interactions, e.g., in accordance with the telephone keypad assignment, from being assigned to the same user interaction.

In some embodiments, the letters of the Latin alphabet in lowercase are assigned to four user interactions or to four keys in the following groups: filrtvy, ahkmnwx, bdgopq, cejsuz, wherein the first group includes the letters that stand on one point, the second group includes the letters that stand on two points, the third group includes the letters that stand on a large base, and the fourth group include the letters that include a substantially closed circle in their shape. Note that in this embodiment, the closed circle characteristic preferably has higher priority over other shape characteristics. Therefore, as an example, the letters "p" and "q" belong to the group fourth group rather than to the first group. It must be noted, that the letter "a" is exceptionally may be included in the second group. .

An aspect of some embodiments relates to a method of defining a location of a keyboard on a touch surface of an input interface of a device. The method includes identifying a simple sweeping gesture on the touch surface, that is not identified by the operating system of the device and positioning the keyboard around the location in which the simple sweeping gesture is provided. Using a gesture not identified by the operating system avoids the simple gesture from being interpreted as performing a different task than the keyboard calibration.

In an exemplary embodiment, the gesture comprises performing substantially the same sweeping gesture concurrently at two different locations. Optionally, the keyboard has two different parts and each of the sweeping gestures indicates the location of a different part. In some embodiments, the simple sweeping gesture comprises a downward movement along the touch surface. In another embodiment, the gesture includes a press-and-holding by a user's finger action provided between the two portions of the split keypad of the invention and providing a gesture until a desired location on the touch sensitive surface. Each portion of the split keypad may then be located on a different side of the user's finger on the sensitive surface.

An aspect of some embodiments relates to an input interface with a touch surface defining a plurality of letter zones which are each associated with a plurality of letters. User sweeping gestures from the letter zones in predetermined directions, regardless of the starting location of the sweeping gesture in the letter zone are interpreted as input of a specific one of the letters of the letter zone, according to the direction.

In some embodiments, the letter zones are soft keys which when pressed indicate ambiguous input of all the letters associated with the soft key letter zone. An aspect of some embodiments relates to an input interface with a touch surface defining a plurality of soft keys each associated ambiguously with a plurality of symbols. When touching a soft key, the symbols corresponding to the touched key are displayed in a large form on a display of the device, possibly the touch surface being the display.

Alternatively or additionally, a special character menu associated with the soft key, which serves as a menu selection for those special keys, is displayed.

An aspect of some embodiments of the invention relates to an input interface including a touch surface on which a plurality of soft keys are defined. The input interface is adapted to interpret a pressing on a soft key as ambiguously entering a plurality of characters in the alternative, and to interpret a directed sweep movement on the touch surface beginning, ending and/or passing through the soft key, as an unambiguous selection of a specific character.

Thus, the user is provided with the flexibility of selecting between ambiguous entry of a plurality of symbols and precise entering of a single character on a position by position basis.

Optionally, the directed movements through the soft key (e.g., beginning, ending or passing) are interpreted as corresponding to specific ones of the plurality of characters associated ambiguously with pressing the key. This makes the task of the user in determining the characters associated with the key simpler.

An aspect of some embodiments of the invention relates to an input interface including a touch surface adapted to identify user sweep movements on the surface and associate the sweeps with functions according to their direction and a current menu. The current menu is determined according to a received user input provided separately from the sweep movement. In some embodiments of the invention, the received user input determining the current menu, is received before the sweep movement. Alternatively or additionally, the user input determining the current menu is received concurrently with the sweep movement.

The received user input determining the current menu may be, for example, pressing a key or providing a directed sweep movement on the touch surface. Possibly, the menus are nested in accordance with a tree model to at least three, four or even more levels.

Optionally, the input interface defines at least four menus or even at least six menus. Optionally, each menu provides different interpretations to up to 8 or 9 sweep directions, although menus with more or fewer different interpretations may be used.

In some embodiments, at least some of the menus assign entrance of a specific symbol to one or more directions and assign functions, such as edit functions (e.g., copy, paste, cut), menu selection and/or function keys, to one or more other directions of the same menu.

An aspect of some embodiments of the invention relates to an input interface including a touch surface adapted to interpret directed sweep movements according to their direction, as compared to a corresponding menu. The menu is selected responsive to a time or length attribute of the sweep movement. In some embodiments, different menus are assigned to sweep movements of different lengths. Alternatively or additionally, different menus are assigned to sweep movements having different entry times or different durations of interaction of the user with the touch surface before beginning the sweep movement and/or after completing the sweep movement. For example, if a user presses on the touch surface and immediately begins the sweep movement along the surface, the sweep movement is interpreted according to a first menu, while if the user waits at least a predetermined duration while contacting the surface and only then begins the sweep movement, the sweep movement is interpreted as relating to a different menu.

Optionally, the corresponding menu also depends on the location on the touch surface from which the sweep movement starts.

An aspect of some embodiments of the invention relates to an input interface adapted to receive user interactions to which a plurality of symbols are assigned ambiguously. The input interface is adapted to display reminders of the symbols ambiguously associated with the user interactions, in a two dimensional array of the symbols. Optionally, the symbols are arranged in the two dimensional array in accordance with well known arrangement of symbols, such as the telephone pad arrangement. Alternatively or additionally, the symbols are arranged in the displayed reminders of a first set of user interactions according to their assignment arrangement in a second set of user interactions. For example, an input interface may receive the letters of the alphabet through two different sets of user interfaces: a first set in which four keys are assigned all the letters of the alphabet and a second set in which the keys are assigned to a telephone keypad arrangement. In such an example, the display reminders of the first set are optionally arranged in a two dimensional array in which the letters are arranged in their positions in the keypad arrangement.

In some embodiments of the invention, the two dimensional arrays have at least a 3 rows and 3 columns. Optionally, the letter arrangement may include empty spaces between displayed letters, such that the symbols correspond to desired positions, even when some positions do not have corresponding symbols.

Optionally, the user interactions comprise actuations of soft and/or hard keys. The display reminders may be displayed on the keys, adjacent the keys and/or at a different location. The display reminders may be displayed continuously or may be displayed when needed, for example when contact of a user with the soft key is identified. An aspect of some embodiments of the invention relates to an input interface adapted to receive two different sets (e.g., a first set and a second set) of user interactions corresponding to a same set of symbols, for example at least the letters of an alphabet. The symbols corresponding to the user interactions of the first set are displayed with at least one feature indicative of the arrangement of the symbols in the second set. The indicative feature may aid users to more quickly identify a second-set user interaction corresponding to a specific symbol and/or when the user is familiar with an arrangement of the second set to more quickly identify the first-set user indication corresponding to the symbol.

In some embodiments, the indicative feature comprises a color. Optionally, the symbol display for each of the first set user interactions is assigned a different color and the symbols displayed for the second set user interactions are displayed color coded with the color of the user interaction of the first set that includes the symbol.

An aspect of some embodiments of the invention relates to an input interface including a touch screen. A plurality of symbol zones (e.g., soft keys) assigned to one or more symbols are defined on the touch screen. Sweep gestures related to a symbol zone, for example beginning and/or ending in the letter zone, are assigned inputs according to their related symbol zone, while sweep gestures not related to any of the symbol zones are given different interpretations.

In some embodiments of the invention, sweep gestures related to a symbol zone are interpreted as providing a new symbol, while sweep gestures not related to a symbol zone are interpreted as correcting a previously entered symbol.

An aspect of some embodiments of the invention relates to a device including an input interface having a touch screen adapted to give different interpretations to different sweep gestures on the screen. Optionally, sweep gestures beginning from an edge of the touch screen is given a different interpretation than sweep gestures beginning at points within the touch screen. For example, sweep gestures beginning at a right and/or left edge of the screen may be interpreted as adjusting locations of soft keys on the touch screen, while sweep gestures on other locations are interpreted as inputting data, controlling a mouse and/or performing other tasks.

In some embodiments of the invention, a thin window is defined along one or more edges of the touch screen such that sweep gestures beginning at the edge of the touch screen are events directed by an operating system of the device to be handled by a process associated with the thin window and not by a general process which handles sweep gestures not assigned to any specific window. Optionally, the thin window has a width of less than 50 pixels, less than 20 pixels or even less than 10 or 5 pixels. In some embodiments of the invention, is barely viewable by the user or even not viewable at all. An aspect of some embodiments of the invention relates to an input interface having a touch pad, which identifies sweep gestures on the touch pad and interprets them as corresponding to various tasks and/or inputs according to their parameters, such as current menu, start location, direction, length and/or duration. The input interface is adapted to identify continued contact with the touch pad after completing the gesture and interpret such continued contact as an input or command additional to that of the sweep gesture or prolonging the input or command of the sweep gesture.

In some embodiments of the invention, a sweep gesture interpreted as a "shift" or other menu selection command takes effect as long as the contact with the touch pad after completing the gesture continues.

In other embodiments, a sweep gesture interpreted as entering one or more symbols (e.g., letters, digits, characters) is interpreted as being repeated every predetermined period as long as the contact with the touch pad at the end of the sweep gesture continues.

Optionally, when a task and/or input is indicated by a user contact with the touch pad at a different location, in addition to the sweep gesture, the interpretation of the continued contact does not require continued contact at the different location.

An aspect of some embodiments of the invention relates to an input interface including a touch surface adapted to interpret user interactions with the touch surface as corresponding to various characters. After the input of some characters, automatically or upon a command from the user, the input interface displays to the user a completion menu, not continuously provided, of completions and/or additions to the input character or to a current input word and the user may select the completion from the menu.

In some embodiments of the invention, the selection of an input from the completion menu is performed by a sweep gesture in a direction of the desired input in the displayed menu.

Optionally, the display of the completion menu is selected by the user by touching the touch surface at a location not assigned to a specific soft key. In some embodiments of the invention, the touching of the touch surface to select the completion menu is performed before the contact with the touch surface for entrance of the completed character is discontinued. The selection from the completion menu is optionally performed by a sweep gesture from the point of touching the touch surface, without discontinuing the contact with the touch surface between the contact with the touch surface and the selection from the completion menu. In some embodiments of the invention, one of the options of the completion menu may be display of additional options. Alternatively or additionally, after selecting from the completion menu, the user is allowed to request display of an additional completion menu, or the additional completion menu is displayed automatically.

The completion menu may provide word suffixes selected from a dictionary for the inputted current word, and/or commonly used completion sequences, such as sequences used in Internet addresses (e.g., com, net after entering a period) and/or the current date or time.

In some embodiments, the completion menu may be used to complete specific symbols, for example for selection of Pinyan symbols for Chinese characters, accent symbols for Latin characters and/or pronunciation marks for Hebrew letters.

An aspect of some embodiments of the invention relates to an input interface including a touch surface adapted to identify user finger swipes on the touch surface and to give different interpretations to swipes in different directions. At least some of the swipes are interpreted as control functions, such as copy, paste, or shift.

An aspect of some embodiments of the invention relates to an input interface including a screen, adapted to display entered text including letters and other characters and symbols in a first word processing window. One or more most recently entered words are optionally displayed in a second window, before their display in the first window or in parallel to their display in the first window, allowing separate control of the display of the current words not under constraints imposed by the word processing window.

In some embodiments, the second window is positioned in a fixed location on the screen. Alternatively, the second window is moved along the first window according to the location of the current words in the first window.

An aspect of some embodiments of the invention relates to a data entry unit which receives user selections of keys of the second keypad of the invention in form of at least the well known 12 key telephone keypad in the form of gliding actions (e.g. vectors) which indicate respective positions of keys in at least the 12 key telephone pad. For example, the vectors may be interpreted as relating to the key to which the vector points if its start point is at the middle 5 button of the telephone keypad. According to one embodiment of the invention, said second keypad may be a predefined imaginary keypad model which may not be shown to the user on a screen and said press/gliding actions on the screen duplicates interactions with the keys of said imaginary keypad.

The first and second keypads of the invention may have any number of keys and any configuration of letters on the keys as long as they comply with the principles of distribution of characters (e.g. letters) as described in detail (e.g. any of the (letter) keys of the first keypad and any of the (letter) keys of the second keypad preferably have at most one common character/letter). There is therefore provided in accordance with an embodiment of the invention, a data entry system, comprising an input interface adapted to identify a plurality of user interactions, a processor adapted to associate user interactions identified by the input interface with a plurality of letters of the alphabet, such that all the letters of the alphabet are assigned to at most eight user interactions, and most of the letters are assigned to specific user interactions according to one or more features of their shape in their printed form, such that most of the letters assigned to each user interaction have the same shape feature.

Optionally, the shape feature of at least one of the user interactions comprises the number of legs the letter has in its printed form. Optionally, the shape feature of at least one of the user interactions comprises whether the letter has an uneven number of legs.

Optionally, the shape feature of at least one of the user interactions comprises whether the letter has a large or a small arc. Optionally, the shape feature of at least one of the user interactions comprises whether the letter has a large arc. Optionally, the shape feature of at least one of the user interactions comprises whether the letter has an upper or lower horizontal line.

Optionally, the shape feature of at least one of the user interactions comprises whether the letter has a lower horizontal line. Optionally, at least one of the user interactions is associated only with the letters having a specific feature shape.

Optionally, one of the user interactions is associated only with letters having two legs.

Optionally, at least one of the user interactions is associated with all the letters having a specific feature.

Optionally, one of the user interactions is associated all the letters having a large arc.

Optionally, the user interactions with which the letters are associated are interpreted as ambiguously entering all the letters with which they are associated.

Optionally, the user interactions with which the letters are associated are interpreted as a first step of selecting unambiguously one of the letters associated with the user interactions.

Optionally, the user interactions comprise pressing of keys.

Optionally, the user interactions comprise sweeping gestures on a touch surface in different directions. Optionally, none of the user interactions are assigned a plurality of letters assigned to a same key in the telephone keypad.

There is further provided in accordance with an embodiment of the invention, a data entry system, comprising a touch screen, a processor, an operating system running on the processor adapted to interpret sweeping gestures on the touch screen and perform tasks responsive to the sweeping gestures and an input module running on the processor adapted to identify a specific sweeping gesture not identified by the operating system and to define a , _i . _{* *}

lo ttuuii υι a jvcyuoar on t e touc screen responsive to t e i enti ie sweeping gesmre.

Optionally, the input module is adapted to identify two gestures provided concurrently at different locations on the screen. Optionally, the input module is adapted to define a first portion of the keyboard at a location of a first one of the two gestures and a second portion of the keyboard at a location of a second one of the two gestures. Optionally, each portion of the keyboard includes fewer than 5 keys. Optionally, the two gestures are the same in direction.

There is further provided in accordance with an embodiment of the invention, a data entry system, comprising a touch surface and a processor adapted to define a plurality of letter zones, each assigned a plurality of letters, on the touch surface, to identify sweeping gestures on the touch surface beginning from the touch surface and to associate the identified sweeping gestures with one of the letters associated with the letter zone at which it begins, according to the direction of the gesture, without relation to the starting point of the sweeping gesture in the letter zone.

Optionally, the processor is adapted to define the letter zones as soft keys which when actuated are interpreted as input of all the letters assigned to the actuated zone, ambiguously. Optionally, the processor is adapted to define four letter zones associated with all the letters of the alphabet.

There is further provided in accordance with an embodiment of the invention, a data entry system, comprising an input interface adapted to receive a plurality of different user interactions, a processor adapted to assign symbols to the user interactions, at least some of the user interactions being assigned a plurality of letters and markings for a plurality of the user interactions indicating the assignment of letters to the user interaction, the markings including a two dimensional array of letters.

Optionally, the markings for the plurality of the user interactions include a 3x3 array of letter positions. Optionally, the markings for at least one of the user interactions include an empty array position between two letters.

Optionally, letters in the markings for the user interactions are arranged according to the arrangements of the letters in the telephone keypad.

Optionally, the user interactions comprise key actuations and the markings are placed on the keys. Alternatively, the user interactions comprise key actuations and the markings are placed not on the keys.

There is further provided in accordance with an embodiment of the invention, a method of receiving user data entry, comprising receiving a first user interaction, by a data entry system; inputting one or more characters corresponding to the first user interaction, receiving a second user interaction, by the data entry system, immediately after receiving the first user interaction, the second user interaction being a sweep gesture on a touch pad of the system, in a specific direction; and inputting an auxiliary input to the one or more characters, responsive to the direction of the second user interaction and to the one or more characters.

Optionally, the second user interaction is received before the first user interaction is completed. Optionally, the second user interaction is received before contact with an input interface of the system for entering the first user interaction is terminated. Optionally, the first user interaction comprises actuation of a soft key. Optionally, the second user interaction comprises a sweep gesture entered after actuating the soft key without releasing the actuation before beginning the sweep gesture.

Optionally, the first user interaction comprises a sweep gesture on a touch pad of the system. Optionally, the first and second user interactions are received through a touch pad of the data entry system, wherein the second user interaction comprises a sweep gesture beginning at a different location on the touch pad, than the ending point of the first user interaction.

Optionally, the second user interaction is started before the contact with the touch pad for the first user interaction is terminated. Optionally, the auxiliary input comprises an accent marking for the one or more characters. Optionally, the auxiliary input comprises a word completion for a current word including the one or more characters.

Optionally, the method includes receiving a third user interaction, by the data entry system, immediately after receiving the second user interaction, the third user interaction being a sweep gesture on a touch pad of the system, in a specific direction and inputting a further auxiliary input to the auxiliary input, responsive to the direction of the third user interaction and to the auxiliary input.

There is further provided in accordance with an embodiment of the invention, a data entry system, comprising a touch screen and a processor adapted to provide on the touch screen a keyboard of soft keys including one or more mode keys, wherein responsive to a mode key being pressed, the processor displays on the touch screen an arrangement of special characters not included in the keyboard and wherein the processor is adapted to identify sweep gestures on the touch screen and to interpret the sweep gestures according to a most recent actuated mode key.

Optionally, the keyboard of soft keys includes at least letter keys arranged as in a QWERTY keyboard.

Optionally, the keyboard of soft keys includes at least letter keys arranged as in a telephone keypad. Optionally, the one or more mode keys comprises a punctuation mode key corresponding to punctuation marks.

Optionally, the arrangement of special characters displayed responsive to the mode key comprises at least six special characters. Optionally, the arrangement of special characters displayed responsive to the mode key comprises not more than twelve special characters.

There is further provided in accordance with an embodiment of the invention, a data entry system, comprising a touch pad, an input interface adapted to receive user interactions; and a processor adapted to identify different user interactions received by the input interface and associate them with respective groups of characters, and to identify letters corresponding to complex sweep gestures on the touch pad based on handwriting recognition and to apply the identified letter to an input position selected responsive to a location of the complex sweep gesture on the touch pad.

Optionally, the input interface is adapted to receive the user interactions through the touch pad. Optionally, the input interface is adapted to receive the user interactions through soft keys.

Optionally, the processor is adapted to handle characters determined from complex sweep gestures on a most recently actuated soft key differently than from gestures not on the most recently actuated soft key. Optionally, the processor is adapted to handle characters determined from complex sweep gestures on a most recently actuated soft key as relating to a currently inserted character and to complex gestures not on any soft key as correcting a previously entered letter.

An aspect of some embodiments relates to using database(s) (i.e. an N-Gram database, wherein each N-Gram is assigned a frequency value, wherein N>0 (e.g. preferably, N>1) wherein N may preferably be a sequence of characters such as, for example, a word, a number, a special character, a sequence of special characters, a punctuation mark, a sequence of punctuation marks, or any combination of said constituents that may include alphabet characters) is consulted to guess the word intended by the user. According to a preferred method of search, a 2-gram word database if preferably generally is used by the system.

An aspect of some embodiments relates to using database(s) having compound words in its entries. A compound word is a sequence of words being connected to each other by a connector. A compound word is considered by the system/user as a single word/unit. A connector may be a predefined character (e.g. or a code) such as a special character or a space character. The connector generally represents a space character between the words of a compound word. Generally when a user enters a space character at the end of entry of a word predicted by the system, said space character is considered by the system as an end-of-the-word signal which confirms the acceptance of the word predicted by the system. In such case, according to one method, either the connector character is to be a character other than a space character, or the system may be designed in a way that the space character is considered to be a part-of-a- compound-word character, and not an end-of-a-word signal.

If the connector is a special character, then, when the system predicts a compound word, it may replace the connector(s) by a space character(s) before it presents the compound word to the user.

An aspect of some embodiments relates to a data entry entry system using an N-gram word database to predict words.

Optionally, when a predicted word is presented to the user, the user may accept or not accept it (e.g. hereafter to referred to as "reject". If the user rejects the predicted word, the system may propose another words. Additionally, the system may enter the sytem into the "Precise Character Mode" and show a corresponding second keypad for each letter key wherein the keys of each of said second keypads includes an identified character/letter to permit a

Correction procedure.

An aspect of some embodiments of the invention relates to entering the system from a Letter Mode to another mode such as Special Character Mode.

An aspect of some embodiments of the invention relates to using different N-gram, word databases by the word predictive system of the invention, wherein N>0.

An aspect of some embodiments of the invention relates to using a 1-gram word database by the word predictive system of the invention.

An aspect of some embodiments of the invention relates to using a 2-gram word database by the word predictive system of the invention.

An aspect of some embodiments of the invention relates to using a single word database by the word predictive system of the invention.

An aspect of some embodiments of the invention relates to a word predictive system predicting words that uses letters and special character(s) to predict a word including said characters.

An aspect of an embodiments of the invention relates to a grid covering a surface to detect the location of the impact of a user's finger on said surface by, both, the data entry system of the invention and by the operating system.

An aspect of some embodiments of the invention relates to a word predictive system wherein a predicted word is either accepted or rejected by the user.

BRIEF DESCRIPTION OF FIGURES

Exemplary non-limiting embodiments of the invention will be described with reference to the following description of embodiments in conjunction with the figures. Identical structures, elements or parts which appear in more than one figure are preferably labeled with a same or similar number in all the figures in which they appear, in which:

Fig. 1 is a schematic illustration of a mobile phone, in accordance with an exemplary embodiment of the invention;

Fig. 3A is a schematic illustration of a mobile device, in accordance with another exemplary embodiment of the invention;

Fig. 3B is a schematic illustration of a data entry device and an auxiliary data entry unit, in accordance with an embodiment of the invention;

Fig. 3C is a schematic illustration of a data entry device, in accordance with another exemplary embodiment of the invention;

Fig. 5 is a schematic illustration of entering data to a mobile device using sweeping acts, in accordance with an exemplary embodiment of the invention;

Fig. 6 is a schematic illustration of a key arrangement, in accordance with an exemplary embodiment of the invention;

Fig. 7 is a schematic illustration of a display responsive to a user touching a screen, in accordance with an exemplary embodiment of the invention;

Fig. 8 is a schematic illustration of correction of a sweeping gesture in progress, in accordance with an exemplary embodiment of the invention;

Fig. 10 shows a key arrangement for a mobile device, in accordance with an exemplary embodiment of the invention;

Figs. 101A-D are schematic illustration entry of first and second keypad in different mode instances, in accordance with some exemplary embodiments of the invention

Fig. 301 is a schematic illustration of a first and a second keypad and portions of the word database and a device using those features;

Figs. 302A-302F are schematic illustration of different group of symbols in different mode instances, in accordance with some exemplary embodiments of the invention;

Figs. 303-303A are schematic illustration entry of words including special characters , in accordance with some exemplary embodiments of the invention;

Figs. 304-304A are schematic illustration letters of the first keypad in a QWERTY arrangement, in accordance with some exemplary embodiments of the invention;

Figs. 305-305A are schematic illustration of using gliding actions from the keys to enter/insert precise characters, in accordance with some exemplary embodiments of the invention;

Figs. 306A-306C are schematic illustration of correcting and inserting characters of a word, in accordance with an exemplary embodiment of the invention;

Figs. 307 is a schematic illustration of entering ambiguous and precise special characters, in accordance with an exemplary embodiment of the invention;

Figs. 317 is a schematic illustration of using short and/or long gliding actions to enter different characters, in accordance with some exemplary embodiments of the invention;

Figs. 319-319A are schematic illustration of using functions, in accordance with some exemplary embodiments of the invention;

Figs. 320A-320B are schematic illustration of entering accented letters, in accordance with some exemplary embodiments of the invention;

Figs. 321A-321B are schematic illustration of entering accented letters, in accordance with some exemplary embodiments of the invention;

Fig. 322A is a schematic illustration of entering derivatives relating to a characters, in accordance with some exemplary embodiments of the invention;

Figs. 322B-322D, are schematic illustration of methods of word completion, in accordance with some exemplary embodiments of the invention;

Figs. 322E-322F, are schematic illustration of using features of the system with one hand, in accordance with some exemplary embodiments of the invention;

Fig. 324 is a schematic illustration of the first keypad of the invention having alphabetical letter layout, in accordance with some exemplary embodiments of the invention;

Figs. 327A-327E are schematic illustratios of a method of displaying the characters and functions on the second keypad, in accordance with some exemplary embodiments of the invention;

Figs. 331A-331D are schematic illustration of simplified user interface, in accordance with an exemplary embodiment of the invention;

Figs. 336-338D are schematic illustrations of shape-based letter arrangement on the keys and different methods of insertion and correction, in accordance with some exemplary embodiments of the invention;

Figs. 336-338D are schematic illustrations of shape-based letter arrangement on the keys of the first keypad, in accordance with some exemplary embodiments of the invention. Fig. 339 is schematic illustrations of shape-based letter arrangement on the keys of the first keypad, in accordance with some exemplary embodiments of the invention.

Figs. 340 is schematic illustrations of shape-based letter arrangement in Korean language on the keys of the first keypad, in accordance with some exemplary embodiments of the invention.

Figs. 341 is schematic illustrations of the first keypad and a second keypad of the invention, in accordance with some exemplary embodiments of the invention.

Figs. 342A-342B are schematic illustrations of shape-based interactions with a touch sensitive surface, in accordance with some exemplary embodiments of the invention.

Figs. 343-343A are schematic illustrations of shape-based letter arrangement on the keys of the first keypad, in accordance with some exemplary embodiments of the invention.

Fig. 344 is schematic illustrations of a social entertainment interface using the invisible input interface of the system, in accordance with some exemplary embodiments of the invention.

Figs. 345A-345B are schematic illustrations of shape-based letter arrangement on the keys of the first keypad and a second keypad in form a QWERTY keyboard, in accordance with some exemplary embodiments of the invention.

Fig. 346 is schematic illustrations of shape-based letter arrangement on the keys of the first keypad and a corresponding second keypad , in accordance with some exemplary embodiments of the invention.

Fig. 347 is schematic illustrations of different examples of predicting words of a dictionary, in accordance with some exemplary embodiments of the invention.

Figs. 349A-349B are schematic illustrations of different shape-based face pictures, in accordance with some exemplary embodiments of the invention.

Figs. 350A-350F are schematic illustrations of behavior of a dynamic keypad, in accordance with some exemplary embodiments of the invention.

Fig. 351 A is schematic illustrations of the word predictive system using connectors, in accordance with some exemplary embodiments of the invention.

Figs. 352A-352C are schematic illustrations of selecting functions, in accordance with some exemplary embodiments of the invention.

Figs. 353A-354 are schematic illustrations of correction procedures, in accordance with some exemplary embodiments of the invention.

Figs. 355A-355C are schematic illustrations of correction procedures relating to accented characters, in accordance with some exemplary embodiments of the invention.

Figs. 356A-358E are schematic illustrations of functions of several second keypads, in accordance with some exemplary embodiments of the invention.

Figs. 359A-359B are schematic illustrations of presentation of candidate words, in accordance with some exemplary embodiments of the invention.

Figs. 360A-360F are schematic illustrations of correction procedures, in accordance with some exemplary embodiments of the invention.

Figs. 361A-362B are schematic illustrations of correction and insertion procedures, in accordance with some exemplary embodiments of the invention.

Fig. 363 is schematic illustrations of functions of a key of the keypad, in accordance with some exemplary embodiments of the invention.

Figs. 364 is schematic illustrations of the system in the Precise Letter Mode instance, in accordance with some exemplary embodiments of the invention.

Figs. 365A-366 are schematic illustrations of the system used as the input interface of GPS, in accordance with some exemplary embodiments of the invention.

Figs. 367-368 are schematic illustrations of the several second keypads in Precise Mode instance, in accordance with some exemplary embodiments of the invention.

Figs. 369 is schematic illustrations of presenting shapes of the letters of the predicted word, in accordance with some exemplary embodiments of the invention.

Figs. 370-371B are schematic presenting word candidates, in accordance with some exemplary embodiments of the invention.

Figs. 372A-372B are schematic presenting shapes of the predicted word, in accordance with some exemplary embodiments of the invention.

Figs. 373 is schematic illustrations an exemplary a key of the keypad in a round-shaped form, in accordance with some exemplary embodiments of the invention.

Figs. 374 is schematic illustrations an exemplary keypad, in accordance with some exemplary embodiments of the invention.

Fig. 375 is schematic illustrations arrangement of letters on six keys based on their common shape characteristics, in accordance with some exemplary embodiments of the invention.

Fig. 376 is schematic illustrations of prediction of word candidates, in accordance with some exemplary embodiments of the invention.

Fig. 377 is schematic illustrations of entering URLs, in accordance with some exemplary embodiments of the invention.

Figs. 378A-378B are schematic illustrations of invisible keys of the keypad, in accordance with some exemplary embodiments of the invention. Figs. 379A-379B are schematic illustrations of interaction with the keys of the first keypad using gestures and/or tapping actions, in accordance with some exemplary embodiments of the invention.

Figs. 380A-385 are schematic illustrations of flow charts relating to the system using N- gram word database(s), in accordance with some exemplary embodiments of the invention.

Figs. 386A-386C are schematic illustrations of a simplified method of text entry, in accordance with some exemplary embodiments of the invention.

Figs. 387A-387D are schematic illustrations of text entry method designed for entering words including special character, in accordance with some exemplary embodiments of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS

According to a preferred embodiment of the invention, the system may include a first set of input signals provided for example through a first set of few keys (e.g. the first keypad) wherein at least the letters of one (e.g., or more) language are disributively and ambiguously assigned to at least some (e.g. preferably four) of said few keys such that to each of said at least some of said few keys more than one of said letters are ambiguously assigned (e.g. ambiguous keys). Preferably, said letters are distributed on said keys such that when a user presses the keys corresponding to the characters (e.g. letters) of a desired word (e.g. first input information) mostly the desired word is proposed to the user by the system based on said key presses (e.g. said first input information). It is understood that in some cases the desired word may not be proposed as a first choice to the user (e.g. the desired word may not have the highest priority among the words that correspond to the same key presses) and the system may propose another word (e.g. a non desired word that has a higher priority).

According to a preferred embodiment, the system may include at least a second set of input signals provided for example through a second set of keys (e.g. the second keypad) such as a telephone-type keypad keys wherein to at least some of the keys of said second set of keys at least the letters of the same one (e.g., or more) language are distributively assigned such that to each of said at least some of said keys of the second set of keys more than one of said letters are ambiguously assigned.

According to one embodiment of the invention, the letters on said few (e.g. four ambiguous) keys of said first set of keys and the letters on said additional/second set of keys (e.g. telephone-type keypad) may be distributed such that a (e.g. any) key of said ambiguous (letter) keys of said first keypad and a (e.g. any) key of said ambiguous (letter) keys of said second keypad may have at most one common character (e.g. letter).

Thus, by pressing a key of a first set of keys and a key of the second set of keys for a specific character/letter or for a specific character location within a word (e.g. a word may have one or more characters), the user indicates a specific letter/character (e.g. the common letter/character) unambiguously. In other words, in order to enter a precise character/letter the user may first press the key of the first keypad corresponding to said character/letter and then press the key of the second keypad also corresponding to said desired character/letter, or vice- versus. By considering said key presses and the fact that there is at most one common character on the two keys, the system precisely selects/enters the desired (e.g. the common) character/letter.

By using key-presses of two keys, the data entry system of the invention thus provides precise character identification, with the unique, character common to the two keys being unambiguously selected thereby. Although having utility on its own, the system as described hereinabove may usefully be used together with a software based word predictive system wherein ambiguous key selections are compared with words within a database including words and corresponding key press information as known, to predict or confirm the user's intention. The principles of such word predictive systems are known to those skilled in the art.

As an example, in order to enter a desired word, according to one method the user may first provide a first input information by pressing the (ambiguous) keys of the first keypad corresponding to the characters of the desired word. The system may propose/predict a word from the database corresponding to the first input information. A word being entered or being edited is hereafter referred to as the "current word". According to one method, one of the characters, preferably the first character, of a current word may be selected by the system. Optionally, the user may also be able to select a character of the current word. Preferably, the selected character may be highlighted. In accordance with one method, the word suggested is the word having the highest priority among the words corresponding to the input information. If the proposed word is not the desired word, the user may proceed with what is referred to herein as a correction procedure providing additional input information by pressing a key of the second keypad corresponding to one of the characters, eventually, the first character, of the desired word. By relating the key presses of the first and the second keypad corresponding to the character, and the fact that there is at most one character common to both selected keys, the character is precisely and unambiguously selected, and the system precisely recognizes the character. By considering the combined input information, the system proposes another word from the database corresponding to the combined information. If the new word proposed by the system is not the desired word, the user may provide additional input information, e.g. by pressing the key of the second keypad corresponding to another character preferably the next character of the desired word, and so on, until the system proposes the desired word.

According to another method, the user may first provide precisely at least one character of the desired word as described above, and additionally press the ambiguous keys of the first keypad that (non-uniquely) correspond to the remaining characters of the desired word. By precisely recognizing at least one character of the word and by also considering the ambiguous key presses corresponding to the remaining characters of the word, the system is able to more accurately predict the desired word and proposes it to the user. If the proposed word is not the desired word, the user may provide additional pressing actions on the keys of the second keypad corresponding to the additional characters of the desired word until the system predicts/proposes the desired word.

If the desired word is not in the dictionary, then, in addition to the first input information provided through the first keypad, generally, the user presses the keys of the second keypad corresponding to all of the letters of the desired word.

Preferably, the first keypad of the invention has generally less (letter) keys (e.g. preferably four letter keys) than the second keypad (e.g. a telephone type keypad having eight letter keys). This is because the main portion of a text may be entered through the first keypad without the need of using the second keypad. Obviously, having/using less keys makes the text entry easier and faster.

The data entry system and correction thereof as described hereinabove is a very fast and easy system. Briefly stated, according to a preferred method, the user types the desired word through the first keypad. Generally, the system proposes the right (e.g. desired) word. If not the user begins to retype the word using the second keypad. If the word is in the dictionary, most of the time, the system proposes the word as soon as retyping the word through the second keypad commences. If the word is not in the dictionary, the user generally retypes all of the letters/characters of said word. With such word predictive data/text entry system, the user does not have to worry about if the word is or is not in the dictionary database. He knows that at some point during typing/using the first keypad (e.g. the system may also include a word completion method) or during the retyping procedure using the second keypad the system will propose the desired word.

According to one aspect, the system may include an In- Word characters selection means to navigate on the characters of the word and select one of them for for example further correction. Said navigation may be made automatically by the system or arbitrarily by the user.

System overview

Fig. 1 is a schematic illustration of a mobile phone 100, in accordance with an exemplary embodiment of the invention. Mobile phone 100 comprises a plurality of input keys and an output screen 140. Each input key may be used by a user to provide one or more input signals, which represent letters, numbers, other symbols and/or control functions. In order to simplify the data entry task while limiting the number of keys, at least some of the input signals are ambiguously associated with a plurality of characters. A processor 130 receives indications of the input signals actuated by the user and selects one character for each received input signal, optionally using an internal word dictionary 132, which lists words with respective ratings of frequency of use.

In some embodiments of the invention, the keys of mobile phone 100 are included in two main groups, a first group 150 and a second group 160, which each has keys associated with all the letters of the alphabet. The letters are optionally distributed between the keys in each group, such that each pair of a key from the first group 150 and a key from the second group 160 has only a single letter in common. Thus, by pressing a first group key and a second group key for a specific character location, the user indicates a specific letter unambiguously. While generally processor 130 guesses the words intended by the user based on the first keys only, when this does not occur the user can easily use the second keys to indicate the precise letters desired. In other embodiments, one or more pair of first and second keys may have more than one letter in common, in order to reduce the number of keys required. Optionally, the letters in common are ones which are generally not interchangeable in words (e.g., Q and V), such that processor 130 will nearly always guess which letter is intended, using dictionary 132. In these embodiments, a mode key may be used to enter a mode in which specific letters can be entered unambiguously.

In an exemplary embodiment of the invention, first group 150 includes four keys, 101,

102, 103 and 104, assigned all the letters of the Latin alphabet. For example, in one embodiment, key 101 is assigned the letters HEBLXQ, key 102 is assigned the letters FARMZT, key 103 is assigned the letters GOVSYCK and key 104 is assigned the letters JUINWDP. Other letter arrangements may be used on the four keys 101-104 and/or the letters may be arranged on a different number of first group keys. Optionally, the letters are arranged in a manner which minimizes the chances of a sequence of key strokes corresponding to more than one frequently used word, for example by placing each of the vowels A, E, I and O on a different key. Keys 105 and 106 are optionally used for entering a space and a backspace, respectively. In some embodiments of the invention, second group 160 includes twelve keys arranged in the standard numeric keypad arrangement used in many mobile telephones, with three columns of four rows. Optionally, the digits are assigned to keys 111-119 and 121 and the letters of the Latin alphabet are assigned to eight keys, keys 112-119. It is noted that the first group (150) and second group (160) may include any suitable number of keys, and the letters of the alphabet may be distributed in each group between any suitable number of keys.

In another exemplary embodiment, the letters are assigned to the first group 150 and/or the second group 160 according to their shapes, in a manner which allows the user to quickly translate the shape of a letter to the key it belongs. Optionally, for a set of four keys 101-104 to which all the letters of the alphabet are assigned, a first key 101 is assigned round letters, e.g., CGJOQU, a second key 102 is assigned two leg letters, e.g., AHKNRWX, a third key 103 is assigned letters having a top or bottom horizontal line, e.g., BDEFLSZ, and a fourth key 104 is assigned letters having an odd number of legs, e.g., IMPTVY. In an alternative embodiment, the shapes of the lower case letters are considered in distributing the letters between the keys. For example, a first key 101 may be assigned letters going below the line, e.g., gjpqy, a second key 102, the remaining letters extending above the upper line, e.g., bdfhiklt, a third key 103 letters having a round lower part, e.g., ceorsu, and a fourth key 104 the remaining letters, e.g., amnuvwxz.

In some embodiments of the invention, the letters are assigned to the first group keys and/or to the second group keys in an order which is familiar to many users, such as the order of the alphabet and/or the QWERTY keyboard order. For example, for six keys corresponding to all the letters of the alphabet, each key may be assigned the letters of half a row of the QWERTY keyboard. Possibly, small variations from the user familiar key order are made, for example when such variations are expected to enhance the correct guessing of intended words using dictionary 132.

While it is best to use the exact QWERTY keyboard arrangement or the exact arrangement of some other familiar keyboard, in some cases small variations from the exact familiar keyboard arrangement may be used in order to avoid having a pair of a first group letter key and a second group letter key having more than one letter in common.

In order to allow quick access to at least some special characters, each of keys 101-104 may be assigned one or more special characters, completing the number of symbols assigned to each key to eight or to nine. Alternatively, each key 101-104 is assigned a single special character. The special characters on keys 101-104 with the letters are optionally those most used while typing words, for example "," (comma), "_" (underscore, "-" (hyphen) and "&" (ampersand). Alternatively, period may be used instead of one of the other symbols.

In a text entering mode, a current character position, referred to herein also as a cursor position, is defined on display 140. Optionally, in entering a word, the user presses a sequence of first group keys 101-104. When the user presses a first group key 101-104 for the first position of the word, the cursor position is associated with all the letters associated with the pressed key. One of the associated letters is displayed in the cursor position on display 140, for example a letter which in itself is a word or a letter which is most commonly used. The cursor position is then moved to an adjacent character position. When the user presses another first group key, the now current cursor position is associated with all the letters associated with the pressed key and one of the letters is displayed in the cursor position on display 140. The letter displayed for the cursor position is optionally selected based on all the keys pressed for the current word and not only the key pressed for the current cursor position. In addition, the letters displayed for the other character positions of the current word are optionally updated according to all the keys pressed for the current word.

After completing entering a word with the first group keys, processor 130 generally guesses the word intended by the user. In those cases in which the word was not guessed correctly, the user can enter additional information using the second group keys. In some embodiments of the invention, when the cursor is positioned at the end of a word, a press on a second group key refers to the first character position of the word associated with a plurality of letters (referred to herein as not disambiguated). Following the pressing of the second group key, the character position is associated only with the letter in common to the letters of the first group key and the second group key pressed for the character position. This letter is displayed on display 140 for the character position. Optionally, the letters displayed for the entire word are updated accordingly, based on the contents of word dictionary 132. The user may then press additional second group keys for subsequent character positions of the word, when processor 130 was still not able to guess the intended word. This process may be repeated, if necessary, until the entire word is entered again, this time using the second group keys.

In other embodiments of the invention, when a second group key is pressed it refers to the last character position of the word that was not disambiguated, rather than to the first letter of the word. It is noted that in some embodiments, the user may press second group keys before entering an entire word, for example when the user knows in advance that the word will not be guessed correctly by processor 130. It is noted that the processor 130 is configured to apply a second input signal provided by a second group key to a different position within the current word than a first input signal provided by a first group key, from a same cursor location. When a second group key is pressed while the cursor is in the middle of a word, with further letters of the word already entered on its right, the second group key optionally applies to the character position immediately to its left. Alternatively, as at the end of the already entered letters of the word, also when in the middle of a word, the second group keys pressed apply to the first not disambiguated character position of the word.

Backspace

In some embodiments of the invention, two different functions may be applied to a backspace key. A first interaction such as a taping action removes one character and a second type of interaction such as a for example, a predefined gliding action on said key removes the group of characters..

The second backspace function may optionally be applied several times in sequence until the entire current word is moved back to its original ambiguous form. In some embodiments of the invention, an undo button or input signal, which cancels the most recent input regardless of whether it was from a first group key or second group key is provided in addition to or instead of the backspace key.

Cursor position

Optionally, a user can select a cursor position using a mouse control or by positioning the cursor near the position to be selected. The selected cursor position may optionally be anywhere on the screen not necessarily in the most recently entered word. Pressing a first key for the selected cursor position optionally replaces the data of the current cursor position with the letters of the first key. Pressing a second group key for the selected position optionally replaces any disambiguation information associated with the cursor position, with the disambiguation by the pressed second key, leaving the first key information of the cursor position unchanged. Alternatively, second group keys take no affect in single letter character positions. Further alternatively, the letters of the second group key replace the current letter or letters of the character position.

Optionally, after pressing a first key for the position, the cursor position of mobile phone 100 remains in the same location, in case the user wants to perform disambiguation using a second group key. In contrast, when a second group key is pressed, the current cursor optionally moves to the next character position, to the next character position including a letter or to the next not disambiguated character position.

In some embodiments of the invention, the user can control the character position affected by the second group keys, separately from the cursor location which affects the character position that will be affected by a pressed first group key. Optionally, a correction point indicator points to a character position within the word, which will be corrected by the second group keys. In some embodiments of the invention, the correction point indicator is displayed on display 140, for example as a dot above or below the letter in the character position to which the indicator points. Optionally, the user can move the correction point indicator within the current word, for example by pressing a first sequence of two first group keys to move the indicator to the right and a second sequence to move the indicator to the left.

Fixation

Optionally, a predefined interaction with a key of the keypad is assigned to a fixation act which tells processor 130 that the selected/highlighted letter/character of the current word is correct. After fixing said character, the system may consider said precise letter and propose another word. The fixation procedure may be repeated for several characters, and each timed, the system may predict another word. Another predefined interaction with the same key or a predefined interaction with another key indicates to processor 130 that all of the the letters/characters of the current word up to the current location are accurate and these locations are disambiguated accordingly. The system may consider said precise character or characters

First - second options

In some embodiments of the invention, the first group 150 keys are used to enter letters and the second group keys are used for disambiguation. Optionally, when a second group key is pressed for a character position not associated with a group of letters entered by a first key, the character position is associated with a digit associated with the pressed second key or with a symbol, function or other key meaning associated with the second key.

In other embodiments of the invention, the second group keys are used to enter letters and the first keys are used only for disambiguation. In still other embodiments, the user may initially enter letters through either the first or second keys as desired by the user, possibly switching between the first and second keys even during entering of a specific single word. In some embodiments of the invention, the operation mode of processor 130 as to whether letters are initially entered through the first keys, the second keys or both, is user adjustable.

Mode keys

Optionally, one or more of the first group keys and/or the second group keys serve as mode keys which change the function of one or more other first group keys and/or second group keys. One or more mode keys optionally change the case of the letters entered by the first group keys and/or the second group keys to upper case. In some embodiments of the invention, the mode keys change the functions of the second group keys to various symbol modes for entering symbols other than letters. In an exemplary embodiment of the invention, one or more dedicated mode keys whose sole function is to move between modes, are used. Alternatively or additionally, one or more keys have two for more different functions depending on how they are actuated. For example, a short press on a key may have a first function, such as inputting letters or other symbols, while a longer press on the key has a different function, such as a mode key function. Alternatively or additionally, a key has a first function when pressed once and a different function when pressed twice within a short period. Further alternatively or additionally, a key has a different function when pressed together with another key. For example, in some embodiments, when a first group key is pressed and a second group key is subsequently pressed before the first group key is released, the pressing of the first group key is interpreted as a different function than if pressed without the subsequent pressing of the second key.

In some embodiments, the first group keys have two functions: a regular function for entering letters, as discussed above, and a mode key function which controls the functions of the second group keys. Possibly, also one or more second group keys have a mode key function, for example when there are more modes than first group keys. In other embodiments, the second group keys serve as mode keys which change the functions of the first group keys. In still other embodiments, the mode keys change the functions of both first and second group keys.

Optionally, when a mode key function is actuated, processor 130 moves into a symbol mode in which all actuations of the second keys are interpreted according to the symbol mode instead of their regular function, until a different symbol mode is entered or the symbol mode is otherwise released, for example by actuating the mode key function again or by pressing a first group key such as the space key 105. In some embodiments of the invention, processor 130 leaves the symbol mode when a key other than one belonging to the symbol mode is pressed. Alternatively, processor 130 leaves the symbol mode after a single actuation of a second group key. Further alternatively, the symbol mode is in effect only when the mode key is held pressed. In some embodiments, the user may instruct processor 130 to leave the symbol mode even if none of the second group keys were actuated during the symbol mode, for example because the user changed his mind.

In an exemplary embodiment of the invention, the six first group keys 101-106 define six different symbol modes which carry the various symbols and/or functions available on standard QWERTY computer keyboards. The symbols are optionally distributed between groups according to a common theme to allow them to be easily found by users. In one specific possible arrangement, the six mode groups include:

a text/punctuation-mark symbol group including symbols such as _ '" ; : & ? ! @ , an arithmetic symbol group including the symbols such as + # - / \ * | = % a digits symbol group including the digits 0-9

a rare symbol group including the symbols < > [ ] { } ' ~

a function group including the function keys F1-F12; and

a computer function group including esc, prtsc, Pause, PgUp, Ctrl, Alt keys.

It is noted that the above arrangement is just an example and other arrangements may be used. More mode groups may be defined for other symbols, for example, for letters of mathematical symbols provided by mathematical editors, foreign alphabets or even for the Latin alphabet letters for users interested in single letter keys. Furthermore, one or more groups may be assigned to commonly used sequences of letters, such as "tion", "ture" and "tice".

The second keys may also correspond to functions not generally included on keyboard keys, such as functions included in menus (e.g., copy, paste, exit).

In some embodiments, one or more keys may be included in a plurality of groups to allow them to be easily found by users, for example the parenthesis may be find in both text and arithmetic groups. Other particular symbols which may be included in several groups include, for example, "space", "backspace" and period which are commonly used.

In some embodiments of the invention, when entering a specific mode, processor 130 displays on screen 140 the layout of the symbols of the keys in the specific mode for the user's convenience. The layout is optionally displayed on a side of screen 140 farther from the mode key causing processor 130 to enter the specific mode. The layout display is optionally removed when processor 130 leaves the specific mode.

Optionally, the user may set whether the layout is displayed or not, possibly controlling the display for each mode separately.

Some symbols and/or sequences which are used frequently may be assigned to shortcuts, for example, to a special press on one of the keys, to a pressing combination on a plurality of keys or to additional keys of mobile phone 100. In some embodiments of the invention, such shortcuts may be associated with a sequence of ambiguous letters.

Key markings

The letter arrangements of the keys may be displayed at a predetermined location on screen.

Optionally, the letters corresponding to first group keys are displayed for each key in their order in the alphabet, so as to allow the user to find a desired letter quickly. Alternatively, the letters of each keys are displayed grouped together in words or syllables which are easier to remember that meaningless letter arrangements. Alternatively or additionally to displaying the layout of the first group keys 101-104 on display 140, the letters and/or symbols associated with the second group keys are displayed. The arrangement of the second group keypad may be displayed continuously or may be displayed only at specific times, for example when the user did not press a key for a predetermined time and/or when the backspace was used. In some embodiments, the second group keypad arrangement is displayed responsive to a user instruction. Optionally, the arrangement displayed is the current arrangement according to the current mode which interprets the second keys.

Display differentiation of ambiguous character positions

As mentioned above, in some embodiments, during entrance of letters using ambiguous keys, processor 130 presents in each character position on display 140 a single letter, selected from the group of letters with which the character position is associated. In some embodiments of the invention, the letters are displayed in a manner which differentiates between character positions associated with a plurality of letters, for which the displayed letter is only a guess and between character positions associated with a single letter. For example, different colors or different font sizes may be used to differentiate between single letter and multi-letter character positions. Further possibilities include surrounding letters with a frame, shading, blinking or color background. In some embodiments of the invention, a marking such as a star or line are displayed above, below or otherwise adjacent letters of multi-letter character positions. Alternatively or additionally, markings are displayed adjacent letters of single letter character positions.

Alternatively to marking all multi-letter character positions, entire words that include one or more multi-letter characters are marked using any of the above methods. Further alternatively, only the first letter of the word that corresponds to a multi-letter character position is marked.

In some embodiments of the invention, the markings used for multi-letter character positions are associated with the keys generating the multi-letter group for the character position. For example, character positions for which the user pressed key 101 may display their guess letter in a first color, character positions for which the user pressed the key 102 display a letter in a second color, etc. Optionally, positions associated with a single letter, for example because the user performed for them disambiguation with a second group key, are displayed in a default color, e.g., black. Alternatively or additionally, each key is assigned a different symbol which is displayed next to the letters representing character positions associated with the letters of the key.

Alternatively or additionally, processor 130 displays for recently entered character positions corresponding to a plurality of letters, some or all the letters to which the character position corresponds. Optionally, at the character position, processor 130 displays the letter currently guessed for the character position, based on the portion of the word entered so far. Beneath or above the character position, in the main window in which the selected and disambiguated letters are displayed, processor 130 optionally displays a letter row or column including the other letters associated with the character position. Thus, the user can easily determine whether a mistake was made in pressing the wrong key or the processor does not have enough information to guess the intended word. The letters in the letter row or column are optionally presented differently from the letter in the character position, for example being in a different color, a different font and/or a different size (e.g., smaller). Optionally, when a plurality of letter columns are displayed next to each other they are displayed in a manner such that they are not read together, for example in different colors, different fonts, different sizes and/or with gaps between the columns.

In some embodiments of the invention, the letter column of a position does not include the selected letter appearing in the character position as it is already shown. In other embodiments, the selected letter appears also in the letter column to allow easier identification by the user, particularly when the letters of the column are in a different size from the selected letter.

The order of the letters in the letter column (or row) is optionally selected without relation to the letters of other character positions of the same word, and generally the letters in the different columns do not usually form words. Each letter in the letter column is optionally included only once. In an exemplary embodiment of the invention, the letters are organized in the columns according to their order in the alphabet. Optionally, the letters are organized in the columns in eight rows, each row corresponding to a key of the telephone-type keypad.

In some embodiments of the invention, the user can control whether the symbols and/or letter columns are displayed. Alternatively or additionally, the system displays the symbols and/or letter columns automatically when the user may be confused, for example when the user presses the backspace key. In some embodiments of the invention, the letter columns are displayed only responsive to a press of the user on a letter, to which press the system responds by displaying the column of the pressed letter or of all the letters of the word. Possibly, responsive to the press, the selected letter or the word of the selected letter is enlarged on the display. Alternatively, the user glides with his finger over a word for which the columns are desired.

Key arrangements The arrangement of keys shown in Fig. 1 is brought as an example and many other arrangements including keys of various shapes (e.g., round, square, half circle) in various arrangements (e.g., rows, columns, circle, ellipse) may be used. Optionally, the keys of the first group and/or of the second group are arranged such that a single finger or two fingers can touch all the keys of the group concurrently, so that the user can press all the keys without looking at the keypad.

In some embodiments of the invention, the displayed letters in the letter columns are presented as soft keys which the user may press to associate their respective character position with the selected letter. The letter column soft keys may be provided instead of, or in addition to, the second group keys. Optionally, in these embodiments, the letter columns of the current word are presented larger than the letter columns of previous words, in order to allow easy access to the soft keys of the letters of the columns. Optionally, the user does not need to use the soft keys of the letter columns but rather may use them if desired.

Soft keys

Fig. 3A is a schematic illustration of a mobile device 300, in accordance with an exemplary embodiment of the invention. Mobile device 300 includes hard keys 201-206 for the first group of keys and a touch screen 320. The second group keys are soft keys 301-312 displayed on touch screen 320. Thus, mobile device 300 may have very few hard keys, for example less than ten or even less than eight hard keys. In some embodiments of the invention, also the first group keys are soft keys displayed on touch screen 320, such that mobile device 300 may have less than four hard keys or even may not have hard keys at all. Keys 201-206 optionally have the same functions as keys 101-106 (Fig. 1), respectively, although they are arranged differently.

The second group keys 301-312 may be shown permanently on screen 320, at least whenever in a text entry mode, or may be displayed only when the user requests their display, for example by pressing a control key. In some embodiments of the invention, the second group keys 301-312 are displayed permanently in a small arrangement and are enlarged when their use is expected, for example when the user presses one of the small keys or when entering a special key mode. When a first group key is pressed, the second group keys 301-312 are optionally removed entirely from the display or are made small, until the next time they are required, so that they do not take up too much space on screen 320.

In some embodiments of the invention, second group keys 301-312 are displayed in the center of screen 320. Alternatively, second group keys 301-312 are displayed on one side of the screen or at the top and/or bottom of the screen 320. For example, the second group keys may be displayed in one or two horizontal rows on the top and/or bottom of screen 320. In some embodiments, the second group keys 301-312 are displayed at the opposite side from the key that instructed them to be displayed, e.g., a mode key. Optionally, keys 301-312 move along screen 320 as data is entered, so as not to interfere with seeing the displayed entered data.

In some embodiments of the invention, the first group keys are soft keys in addition to, or instead of the second group keys. The first group keys may all be positioned together adjacent each other on the touch screen 320 or may be separated into two different areas on the touch screen, for example on the right and left sides of screen 320. While the soft keys are shown as being presented on a large touch screen, in other embodiments the soft keys may be presented on a touch pad which is not used for display and/or may be provided on a plurality of small touch surfaces, each of which presents a single soft key or a plurality of soft keys. The touch surfaces may all present the same number of keys or different touch surfaces may present different numbers of keys. When a second-group key is pressed and not released for a long time, the function of the key is optionally repeated until it is released. The repeating of a function of a key when held pressed may be performed also for the first group keys in those embodiments in which a long press on the key does not have a different function than the short press. Alternatively, processor 130 may differentiate between three or more lengths of holding a key pressed. A short press on a first group key 201-204 enters a set of letters, a longer press serves as a mode key and an even longer press enters the set of letters repeatedly until the key is released. In an exemplary embodiment of the invention, a short press on a first group key enters the characters associated with the key, a longer press enters into a first symbol mode, for example a mode of numerical digits and an even longer press enters a different symbol mode, such as a function key mode.

The principle of different pressing durations may be used also for second group keys. For example, in one or more symbol modes, a short duration press on each key may be associated with a first function while a longer press is associated with a different function. For example, a mode associated with both digits and function keys F1-F12 may be provided. A short press on one of the second group keys enters the corresponding digit, while a long press enters the corresponding function key (e.g., F7).

In order to allow use of the entire screen 320 for display of data while it is used for soft keys, in some embodiments, screen 320 may not show the locations of the soft keys but rather the user is trusted to remember them. This is especially practical when there is only a limited number of soft keys, such as when using screen 320 for no more than six or four soft keys. Alternatively or additionally, the locations of the soft keys may be calibrated dynamically according to the user acts. For example, at the beginning of a text entry session the user may indicate the desired location of the keys, for example by entering an X shape or a single tap pressing on a specific point, such as the center of the desired keyboard. A short sweeping gesture in the direction of one of the four keys, from a center point at which the four keys meet, may be used both to indicate the layout of the keys and to actuate the key. The system optionally identifies the short sweeping gesture and according to its direction identifies the actuated soft key. In addition, the starting point of the sweeping gesture is identified as the meeting point of the four keys and accordingly their locations are fixed until a next calibration. The space and backspace soft keys are optionally fixed at the bottom left and right corners of the touch pad or at any other suitable location.

Optionally, the calibration method may be used at any time to move the keyboard to a different location.

In some embodiments of the invention, the calibration is performed by sweeping gestures beginning at edges of the touch pad. Optionally, in order to avoid mistakes, the user is required to providing sweeping gestures from right and left (or top and bottom) edges concurrently to indicate calibration of the location of the keys.

Optionally, in order to avoid the need to reprogram an operating system of mobile phone 100, thin bar shaped windows, possibly not seen by the user or barely seen by the user are defined on the edges of the touch screen 320. When the user begins a sweeping gesture at the edges of the touch screen the system initiates an event which is transferred to a process corresponding to the bar windows and performs the calibration.

In some embodiments of the invention, the soft keys of the first group 150 are provided in two sub-groups on opposite sides of the touch screen. Both sub-groups may be calibrated together or the user may calibrate the position of each sub-group separately, for example in order to see text underneath the sub-group of keys.

In an exemplary embodiment, short gestures are gestures of a length less than the radius or diameter of a key. Alternatively, for gestures starting from a specific key, short sweeping gestures are ones that do not exit the borders of the key in which they originated.

In some embodiments of the invention, a compromise is provided between complete marking of the locations of the soft keys and no marking at all, in the form of small hints to the locations of the keys.

Fig. 3B is a schematic illustration of a data entry device 360, in accordance with an embodiment of the invention. Instead of including hard keys, device 360 includes (invisible) soft keys 362 on the right and left of the touch screen 320. In this example, each of the two groups of keys of the split (invisible) keypad, has three keys forming a column of keys. The first group has two letter keys and a backspace key, and the second group has According to one method, two letter keys and a space key.

Fig. 3B also illustrates the possibility of using an auxiliary small data entry unit 366 which communicates with device 360. Device 360 may be a computer which is heavy and auxiliary data entry unit 366 is used by the user for data entry. In other embodiments, data entry unit 366 may be used with a desk top computer, a notebook computer or any other device. Data entry unit 366 may communicate with the main device through wires or wirelessly, using any method known in the art.

Unit 366 optionally comprises a touch pad 369 and a processor 367. Processor 367 optionally differentiates between sweeping gestures, described herein below in detail, and pressing acts on touch pad 369. Touch pad 369 is optionally divided into six key areas, corresponding to keys 201-206 of Fig. 3A, and processor 367 identifies for each user-press the actuated key. Optionally, two dot markings 368 are provided on the touch surface to help the user orient fingers in pressing on the keys. Alternatively, two bulges are used so the user can feel them.

In some embodiments of the invention, the guessing of intended characters based on the input information from the user is optionally performed by processor 367 within unit 366. In other embodiments, the guessing is performed by a processor of device 360.

Auxiliary unit 366 may be very simple and very inexpensive to produce. Optionally, touch pad 369 is not a touch screen and unit 366 does not include any hard keys. Processor 367 is optionally dedicated only for the data reception and possibly the character guessing. In other embodiments, the auxiliary unit may include additionally hard keys and/or a touch screen.

In some embodiments of the invention, auxiliary unit 366 is battery operated. Optionally, unit 366 has a USB connection (or any similar connection type) through which the battery is charged. Alternatively or additionally, auxiliary unit 366 communicates with data entry device 360 through a USB connection. A long USB wire may be used to allow connection to the computer while auxiliary unit 366 is in the users hand.

In some embodiments of the invention, auxiliary data entry unit 366 is of a small size, for example having an area of less than 10 square centimeters or even less than 5 square centimeters.

In some embodiments, auxiliary data entry unit 366 is integrated into a USB memory stick. Thus a single pocket sized device serves both as a storage unit for important data and as a keyboard for entering information. Auxiliary data entry unit 366 can be connected to any computer through a USB slot and the user can then access the information stored in unit 366 and use the keys and/or touch screen of unit 366 to interact with the accessed data.

When the soft keys of the first group are divided between two different locations on screen 320, the user optionally provides a calibration signal for each portion of the keypad separately. Alternatively, the user provides a calibration signal for one portion of the first keypad and the location of the other portion is set accordingly, for example so that they are symmetrical relative to a middle axis of screen 320.

In some embodiments of the invention, processor 130 accepts calibration signals, such as a tap or sweeping move of a finger, only if it is provided by a specific finger, based on fingerprint comparison or finger width. For example, whenever a press with a thumb is identified it may be interpreted as a calibration signal, while pressing with other fingers are interpreted as regular data input. This prevents inadvertent calibration and/or mix-up between data entry and calibration. In addition, fingerprint identification may be used as a security measure to prevent use of mobile device 300 by unauthorized users. In other embodiments, calibration signals are received from any finger.

Identification of different fingers or different finger orientations may be used also in data entry. Instead of assigning the keys specific areas on screen 320, different finger presses anywhere on the touch screen are interpreted as different keys. For example, pressing with the tip of the finger may be interpreted as key 201, the wide face of the finger as key 202, with the narrow side of the finger as key 203 and pressing with the thumb as key 204.

In some embodiments of the invention, processor 130 determines which finger is touching screen 320 and accordingly controls which of the first and second keypads is being accessed by the user. Using a first finger or first finger orientation is interpreted that the user is pressing a key from the first group 150, while a second finger or finger orientation is used to indicate the user is pressing a key from the second group 160.

It is noted that the first and second groups of keys may have different numbers of keys than those described above, according to the specific needs of the users. Particularly, the second group of keys may have different numbers of keys in different symbol modes as required. For example, in a punctuation mode the second group may include 10 or 12 keys, while in a Greek alphabet mode, the second group may include a key for each Greek letter. In some embodiments, in one or more modes, the second group keys include a plurality of sets of twelve keys, each set organized in an array of three columns and four rows, even if some of the keys are not used. Optionally, the groups of 12 keys are separated from each other such that they are distinct. As the 12 keypad arrangement is very well known, organizing the keys in groups of 12 keys in a 3x4 arrangement may make it easier for the user to identify the keys needed.

In some embodiments of the invention, processor 130 is adapted to identify pressing on two different virtual keys on screen 320 concurrently and provide a different input than if only one virtual key is pressed. The user may optionally press on the two virtual keys with any two fingers. Alternatively, different pairs of two fingers pressing on screen 320 are given different interpretations.

In some embodiments of the invention, in addition to allowing the use of second group keys (and their equivalents) for disambiguation, when an ambiguous key is pressed the system may displays s soft key for each of the ambiguous letters/characters as an identified character.. Alternatively or additionally, soft keys for disambiguation are displayed, automatically or upon a user command, when a user selects a previous character position for disambiguation thereof. In some embodiments of the invention, the soft keys displayed correspond to the last actuated first group key. In other embodiments, the soft keys displayed are of the first non-disambiguated character position in the current word, regardless of the cursor position. It must be noted that soft keys may have different arrangements. For example, they may be arranged on a three-by- three matrix key arrangement or they may be arranged in a line.

Sweeping gestures

Instead of using soft keys on screen 320, other user gestures may indicate the various input signals. In some embodiments of the invention, processor 130 identifies straight line sweeping gestures on screen 320 and according to their direction, length and/or other attributes, associates each identified sweeping gesture with one or more symbols or functions.

In some embodiments of the invention, the sweeping gestures are used for the functions of the first group keys or the second group keys, instead of providing soft keys of the respective group on screen 320. The sweeping gestures are optionally in directions selected according to a known arrangement of the keys of the group, such that a user accustomed to a specific key arrangement can easily adapt to using the sweeping actions. For example, instead of providing a 12 key telephone keypad arrangement, processor 130 is adapted to associate sweeping acts according to their directions, with the keys of the known telephone keypad arrangement.

Use of the sweeping gestures allows the user to provide input anywhere on the screen, instead of confining the user input to a specific area on the screen.

Optionally, in embodiments in which processor 130 is adapted to identify sweeping gestures, soft keys are not provided on screen 320 in order to avoid sweeping gestures from being interpreted as pressing on soft keys. Alternatively, screen 320 is divided into a plurality of zones, a zone for sweeping gestures and one or more zones with soft keys. For example, the soft keys may be placed around the outskirts of screen 320, while the center of screen 320 is used for sweeping gestures. In another example, screen 320 is divided into right and left halves. One half is used for soft keys and the other for sweeping gestures.

Fig. 5 is a schematic illustration of entering data to mobile device 300 of fig. 3A using sweeping gestures, in accordance with an exemplary embodiment of the invention. Instead of pressing on soft keys 301-312 (Fig. 3A), the user provides straight line sweeping gestures, represented by arrows 381-388. Each sweeping gesture corresponds to the key in the direction from key 305 to the desired key on a standard telephone keypad or a variation thereof. For example, to enter the letters DEF of key 303, the user provides a sweeping motion on screen 320 in the direction of arrow 383, in a diagonal between the right and the upward directions. Similarly, a sweeping gesture in the direction of arrow 382 represents the letters ABC, in the direction of arrow 384 represents the letters GHI, in the downward direction of arrow 387 represents letters TUV and in the direction of arrow 388 represents the letters WXYZ. In some embodiments of the invention, to enter the letters JKL the user taps once or twice on the screen at any point, indicating the middle of the standard telephone keypad. Alternatively, the letter organization 370 shown in Fig. 6 is used, and the letters JKL are entered by a sweeping gesture in the direction of arrow 381. Alternatively, the sweeping gesture in the direction of up and to the left indicated by arrow 381, corresponding to the key of the digit "1" on the telephone keypad, is not associated with letters, but rather is interpreted as a backspace act canceling disambiguation of the previous character position.

Optionally, the sweeping gestures are required to start in a specific predetermined location, so that inadvertent sweeping gestures are not interpreted as input. In some embodiments of the invention, the location of the sweeping gestures on the screen may be calibrated by the user. In other embodiments, the sweeping gestures may be provided anywhere on screen 320, such that sweeping gestures following arrows 392, 395 and 396 have the same effect as those of arrows 382, 385 and 386, respectively. These embodiments allow the user more flexibility.

In other embodiments, the sweeping gestures are directed from the location of the desired key to the center of the keypad. In these embodiments, arrow 392, for example, corresponds to the letters TUV and arrow 395 corresponds to the letters GHI.

Letter organization 370 (Fig. 6), or any other letter organization used, may be displayed on screen 320, may be engraved or otherwise posted on the casing of mobile device 300 or may not be provided at all, based on the assumption that users know the organization by heart. In some embodiments of the invention, letter organization 370 is not shown on screen 320 regularly, but is shown when the user touches screen 320, as such touching indicates the user probably wants to enter data. Optionally, the letter organization 370 is shown at a predetermined location on the screen or at a location which is currently vacant. Alternatively, letter organization 370 is displayed at the location where the user touched the screen, allowing the user to sweep his finger toward the direction of the displayed desired letters, from the point where the finger originally touched the screen. The user may optionally control if and/or how letter organization 370 is displayed.

Alternatively or additionally to displaying letter organization 370 responsive to the user touching the screen, guiding lines are displayed showing the user the path on which the sweeping gestures are to be performed. Further alternatively or additionally, border lines between the sectors of the different sweeping gestures are displayed.

Fig. 7 is a schematic illustration of a display responsive to a user touching the screen, in accordance with an exemplary embodiment of the invention. Optionally, responsive to a user touching screen 320 at point 700, letters 702 are displayed around point 700, together with grid lines showing the sectors in which sweeping gestures will be interpreted as relating to the respective letters. The user can then easily perform, for example, sweeping gesture 706 within the grid lines of the letters DEF.

Instead of displaying all the letters in letter organization 370, processor 130 determines which letter will be used for each sweeping gesture for a letter selected by the system or by the user in a word currently being entered/edied (e.g. hereafter may be referred to as "current word"), given that it already is related to a plurality of letters from a first key actuation, and only these letters are displayed. Displaying only a single letter makes the task of identifying the desired sweeping gesture much easier. In other embodiments, all the letters are displayed, but the letter corresponding to the character position is highlighted. In still other embodiments, for each sweeping gesture direction, processor 130 displays the word that will be selected for the current word character position sequence, if the sweeping gesture is selected. In cases in which a single sweeping gesture is required to receive the desired word, this makes the selection of the sweeping gesture very simple as the user sees the desired word in the direction of the required sweeping gesture. In some embodiments of the invention, processor 130 displays both the letter and the guessed word, for the convenience of the user, in case more than one sweeping gesture is required to achieve the desired word.

In still other embodiments, the letter or letters are displayed not when the user touches the screen but when the sweeping gesture begins or ends, and only the letter or letters corresponding to the entered sweeping gesture are displayed. The user is optionally given a chance to easily correct his sweeping gesture if necessary, responsive to the display of the letter corresponding to the sweeping gesture. For example, in embodiments in which the gesture is determined based on the relative direction between the beginning and ending point, the user may change the direction of the sweeping gesture in the middle, as illustrated by Fig. 8, to change from an erroneous entering of the letter N to entering the letter D. Alternatively or additionally, when processor 130 identifies an abrupt slowdown in a sweeping gesture, the sweeping gesture is ignored, as the abrupt slowdown is generally indicative of a hesitation by the user. Similarly, in some embodiments of the invention, if an abrupt lifting of the finger is identified responsive to the display (e.g., within a predetermined short period from the display), the sweeping gesture is ignored. The speed threshold for differentiating between regular and aborted sweeping gestures may be predetermined or may be based on learning the operation habits of the user of the specific device.

Optionally, when the user begins a backspace sweeping gesture, processor 130 displays the current word that will be affected by the backspace act, possibly highlighting the letter in the character position which will be affected. A similar display may be performed for the backspace direction in embodiments in which letters are displayed in all directions or in some of the directions.

In an exemplary embodiment of the invention, when the user touches the screen, the key arrangement for all sweeping directions, is displayed. When the user begins the sweeping gesture, the letters for all directions are removed from the display, except for the direction of the sweeping gesture in progress.

In some embodiments of the invention, each sweeping gesture must begin from a state in which the user's finger does not touch the screen 320. In performing the sweeping gesture, the user presses his/her finger against the screen, glides the finger in the required direction and removes the finger from the screen. In other embodiments, the sweeping gesture does not require removing the finger from the screen when it is completed. Rather, a sequence of a plurality of sweeping gestures, referred to herein as a complex gesture, may be provided by the user without lifting the finger from the screen. Optionally, each time the user changes the direction of the movement of the finger on screen 320, processor 130 considers the turn point as the end of a previous gesture and the beginning of a new sweeping gesture, whose direction is evaluated relative to this point. Alternatively, the direction of each straight line segment of the complex sweeping gesture is evaluated relative to the starting point of the entire complex gesture.

A single complex gesture may be used to enter an entire word, or several complex gestures may be used to enter a single word in parts. Furthermore, a word may be entered by one or more complex gestures together with one or more single position sweeping gestures.

In some embodiments of the invention, when a sweeping gesture is a very long straight line it is considered as two separate sweeping gestures in the same direction. Alternatively, long sweeping gestures are considered the same as short sweeping gestures, as not always can users easily control the lengths of their sweeping gestures. Further alternatively, when the sweeping gestures are used to represent keys of a keypad of more than nine keys, processor 130 may differentiate between sweeping gestures representing keys in similar directions, such as keys 307 and 310 or keys 308 and 311 of Fig. 3A, based on the length of the sweeping gesture. In other embodiments, the keys are arranged such that each key has a unique direction which is not generally interchanged with the other directions.

Long sweeping gestures may be interpreted as long presses on their corresponding key. For example, in an embodiment in which a short press on a key provides a digit and a long press enters a function key, the user may indicate a short press by entering a short sweeping gesture and a long press by a long sweeping gesture. Alternatively, the gestures corresponding to short and long presses on the same key may be differentiated using other methods. For example, while short presses may be represented by simple straight lines, longer presses may be represented by wiggled lines, a loop at the end of the line or any other suitable marking.

Optionally, if the user's finger is identified as remaining touching screen 320 without moving after a sweeping gesture, the function of the gesture is repeated until the user's finger is removed from screen 320, in a manner similar to the press and hold function known in the art for keyboard keys. In some embodiments of the invention, the repetition is of the last function entered, even if a sequence of letters was entered in one complex sweeping gesture. Alternatively, the entire sequence of letters or functions entered by the last sweeping gesture is repeated until the user's finger is lifted. For example, if the user performed a complex gesture which operated as pressing keys 201, 202 and 204 in a sequence, if the user keeps his finger touching screen 320 at the end of the sequence, processor 130 acts as if keys 201, 202 and 204 are repeatedly pressed in sequence.

Alternatively to interpreting crooked sweeping gestures as formed of a plurality of straight lines, each representing a different key press, in some embodiments of the invention, crooked sweeping gestures are interpreted based on their beginning and ending points, as if they were a straight line between the beginning and ending point.

In some embodiments, crooked sweeping gestures are interpreted as separate sweeping gestures having a totally different meaning than the straight lines from which they are formed. Optionally, crooked sweeping gestures are used as shortcuts representing a plurality of keys.

In some embodiments of the invention, the sweeping gestures are used to represent only some of the functions of the second keys, such as the letter groups, while digits and/or symbols in symbol modes are provided using soft keys. In other embodiments, the sweeping gestures replace all the functions of the keys they represent. Optionally, the user may decide whether to use sweeping gestures or soft keys.

In some embodiments, processor 130 determines if sweeping gestures relate to mouse control or character entry based on whether the user is in the middle of entering a word. Optionally, when in a symbol mode or when in the middle of entering a word, sweeping gestures are interpreted as referring to second-group key functions, and otherwise the sweeping gestures are used to control a mouse cursor.

Optionally, sweeping gestures entered may be entered by a stylus or a finger

Separate zones could also be used instead of using mode keys. Sweeping gestures in a first zone may be interpreted as corresponding to keys of first group 150, while sweeping gestures of a second zone are interpreted as corresponding to keys of second group 160. Alternatively or additionally, sweeping gestures of one or more zones may be interpreted as corresponding to keys in a symbol mode (such as a caps-lock mode or a mathematical symbol mode), avoiding the need to press a mode key.

In an exemplary embodiment of the invention, one of the zones is assigned to sweeping acts which correspond to mode keys. Alternatively, specific sweeping acts are assigned to the mode key functions, such as sweeping acts which are longer than regular sweeping gestures. Different types of sweeping gestures may also be used to differentiate between other groups of keys. For example, in embodiments in which sweeping acts may be used for both first group and second group keys, the user may use relatively short sweeping gestures for the first group keys and longer sweeping acts for the second group keys, or vice versa. In other embodiments, the sweeping gestures differ in their shape. For example, straight and curved sweeping acts may be differentiated by processor 130, and each type of sweeping act may be assigned different groups of keys. In one embodiment, straight diagonal sweeping gestures are assigned to first group keys, while straight horizontal and vertical sweeping gestures arid curved sweeping gestures are used to indicate the keys of the second group.

The different zones could all be used for soft keys, could all be used for sweeping gestures or one or more could be used for sweeping gestures while one or more others are used for soft keys. In an exemplary embodiment of the invention, a first zone displays soft keys of the first keypad and the rest of the area of the screen 320 not occupied by the first keypad is considered a second zone in which sweeping gestures corresponding to second keypad keys are received.

Optionally, sweeping gestures extending between zones are ignored. Alternatively, sweeping gestures are interpreted according to their start location or end location. Further alternatively, sweeping gestures are associated with the zone including the majority of their length.

Another method to differentiate between sweeping gestures for different symbol modes is requiring use of different fingers, numbers of fingers or combinations of fingers. For example, sweeping gestures using a right thumb may relate to a first group of symbols, while sweeping gestures with three fingers relate to a second group of symbols.

Sweeping gestures for disambiguation

Alternatively or additionally to using sweeping gestures to indicate specific keys, sweeping gestures may be used to select a single letter, symbol or function from the multiple letters (or symbols or functions) associated with a soft key. Optionally, pressing on a soft key ambiguously enters the letters associated with the key, while entering a sweeping gesture beginning from the location of the soft key enters a single letter associated with that sweeping direction for that key. In some embodiments, the keys are first group keys 101-104 and the specific letters associated with the directions of the sweeping gestures starting from each key are assigned according to the key arrangement of the second group keys, as discussed above, for example with reference to Fig. 6. Thus, in addition to or instead of, the possibility of separately entering for a specific cursor position both a first group key press and a second group sweeping gesture, the user may enter the first group key press and the second group gesture at the same time.

Optionally, in order to indicate the center key 378 of model 370 of Fig. 6, the user presses the key 101-104 for a long duration which allows the system to differentiate between a regular press which is interpreted as corresponding ambiguously to the letters associated with the key and a long press corresponding to a specific single letter. In some embodiments, sweeping gestures from first keys to indicate specific letters are used in addition to sweeping gestures for indicating second group 160 keys. Any of the methods discussed above regarding differentiating between different interpretations of sweeping gestures may be used to differentiate between sweepings in this embodiment. For example, sweeping gestures starting from first group soft keys or an area assigned to the first group 150 soft keys are interpreted as precise letters while sweeping gestures starting in other areas of touch screen 320 are interpreted as referring to second group 160 keys. Sweeping gestures from pressed keys may be used also for second group keys or other keys of a data entry device. Particularly, sweeping gestures from keys to select particular symbols may be used also in number and/or symbol modes.

Extended character sets

While in the above description the first and second key groups had the Latin alphabet letters in common, the same principle may be used for other characters including other alphabets, digits and symbols. In one embodiment of the invention, each first group key includes also a single non-letter symbol in its group of ambiguous characters. In the second group keys, the symbols are optionally associated with key 301, allowing fast disambiguation when necessary.

Fig. 10 shows a key arrangement for a mobile device 950, in accordance with an exemplary embodiment of the invention. Keys 201-204 are assigned the symbols shown in symbol arrangement 980, with the symbols of box 961 corresponding to key 201, the symbols of box 962 corresponding to key 202, etc. In this example, each key corresponds ambiguously to a set of letters and to other symbols. For example, key 201 corresponds ambiguously to the digits 0 to 9 in addition to a group of letters, key 202 corresponds ambiguously to the arithmetic characters 972 (e.g. the arithmetic symbol group of characters discussed above in relation to the mode keys) in addition to the set of letters, key 203 corresponds ambiguously to the punctuation characters 973 (e.g. the text symbol group of characters) in addition to its set of letters, and key 204 corresponds ambiguously to the rare characters 974 (e.g. the "RARE" group of characters) in addition to its set of letters.

Mobile device 950 recognizes two second group keypads 951 and 952. Preferably, the characters on the keys 201-204 of the first keypad are distributed on the keys of the second virtual keypads 952 and 951, such that any key of the first keypad has at most one common character with any of the second group keys.

Predefined tapping or sweeping actions (e.g. 982) on the right side of the touch screen optionally correspond to interacting with the keys of the second group keypad 952, and predefined tapping or sweeping actions on the left side (e.g. 984) of the touch screen optionally correspond to interacting with the keys of the second group keypad 951. Optionally, in guessing which character to use for character positions for which a second group key was not actuated, preference is given to the letters. Alternatively, the letters and other symbols are given equal preference, depending on their rating in dictionary 132. In some embodiments of the invention, in order to select a letter for disambiguation, the user presses the screen on the right zone and responsive thereto the layout 952 is displayed. The user than performs a sweeping gesture corresponding to the desired letter. Similarly, when the user wants to enter a digit or symbol in the disambiguation, the user presses on the screen in the left zone and the key arrangement 951 is displayed to direct the sweeping gesture. Other display methods described herein may also be used as well as other display methods known in the art.

These embodiments are particularly useful when database 132 is adapted to learn new words frequently used by the user. Such words may be, for example, passwords, URL addresses, slang and unofficial SMS symbol talk, formulas, computer programming language expressions, etc. While in the first time these "words" are entered the user will in many cases need to enter the entire word twice, once with the first group keys and then with the second group keys (or their equivalents), the next time the same word is entered, there is a fair chance it will be guessed by processor 130.

The inclusion of all the digits on a single first group key makes it easier to remember which key is to be pressed for a digit. In other embodiments, however, the digits are distributed between the first group keys to increase the chances an intended number is identified.

When the embodiment of Fig. 10 is used with display of soft keys described above with reference to Fig. 3C, soft keys may be displayed for all the letters and symbols on the pressed first group key. Alternatively, soft keys are displayed only for the non-letter characters and the user performs disambiguation of letters by sweeping gestures.

Using first key combinations instead of second keys

Combinations of first group keys may be used as shortcuts for tasks which users may want to do in a simpler way then using mode keys and the second keys or their equivalents (e.g., sweeping gestures).

Optionally, pairs of first group keys pressed together are given meanings of tasks which are used relatively frequently. In some embodiments, pairs of first group keys have the functions of arrow keys which allow navigating the cursor around the display and/or navigating a correction point indicator around the display. In other embodiments, pairs of first group keys do not have any function.

In some embodiments of the invention, pairs of first group keys may be used to perform the functions hereinabove assigned to the second group keys. Accordingly, all the input tasks described above may be achieved with only six keys. Optionally, when a user presses one key and while it is held pressed he presses another key, the pressing of the two keys is considered a separate act than a short or long press on one of the keys. Accordingly, in these embodiments, processor 130 is adapted to identify three different patterns of actuations of first group keys:

1) a short press on a key - interpreted as a set of letters, a space or backspace

2) a long press on a key - interpreted as a mode change 3) pressing a key and while it is held pressed pressing a second key - interpreted as a second group key.

Using the four keys 201-204 provides 12 unique sequences of pressing one of the keys and then another of the keys while the previous is held pressed. These 12 unique sequences may be associated with the 12 keys of the second group keypad in any correlation. In some embodiments of the invention, a correlation which allows the user to intuitively correlate between the first key sequences and the second keys, is used.

In an exemplary embodiment of the invention, the correlation between the first key sequences and the second group keys is based on the direction from the first-group key that was pressed and held to the first-group key that was pressed subsequently. For example, pressing and holding key 201 (Fig. 3 A) and subsequently pressing key 202 defines an arrow pointing to the right. Pressing and holding key 203 and subsequently pressing key 202 defines an arrow pointing to the corner in the upper right. Each arrow is optionally interpreted as corresponding to the second-group key of key arrangement 370 (Fig. 6) to which an arrow of that direction points from the center 378. Accordingly, pressing and holding key 202 and subsequently pressing key 204 before key 202 is released, refers to the key carrying the letters TUV. If processor 130 is in a letter mode, it will take as its input the letters TUV. If it is in a symbol mode it will take as its input the symbol or symbols associated with that key.

Using this correlation limits the second group of keys to only eight keys instead of 12 keys which are possible, but has the advantage of being intuitive to the user.

It may be desired to be able to enter 12 second group keys using the first group keys. In some embodiments of the invention, instead of using long presses on the first keys 201-204 to enter various symbol modes, long presses on keys 201-204 are associated with the functions of keys 305, 310, 311 and 312 (Fig. 3 A), respectively, of the second key group.

In other embodiments, the six keys 201-206 (Fig. 3 A) are used to generate arrows corresponding to all 12 keys of the known telephone keypad arrangement. The sequences of keys (202,201), (204,203) and (206,205) generate a virtual left arrow which corresponds to key 304. The sequences (201,202), (203,204) and (205,206) correspond to key 306, the sequences (204, 202), (206, 204), (203,201) and (205,203) correspond to key 302 and the sequences (201,203), (203,205), (202,204) and (204,206) correspond to key 308. Sequences (201,204), (203, 206) correspond to key 309, sequences (203, 202) and (205, 204) correspond to key 303, sequences (202,203) and (204,205) correspond to key 307 and sequences (204,201) and (206,203) correspond to key 301. Sequence (201,206) corresponds to key 312, sequence (202,205) corresponds to key 310 and sequences (202,206) and (201,205) correspond to key 311. Optionally, sequences (205,201) and (206,201) which define a virtual long arrow upward correspond to key 305. The sequences (206,201) and (205,202) may be undefined or may be used for other tasks, such as entering symbol modes (instead of or in addition to long presses on the first keys), caps lock or a partial backspace which removes disambiguation signals.

The same principal of arrows defined by the order of pressing keys may be used with other arrangements of the first group keys. For example, with relation to keys 101-106 shown in Fig. 1, the key pair (104,101) optionally corresponds to key 301 and key pair (102,101) optionally corresponds to key 304. Although arrows in the directions of keys 310, 311 and 312 are not provided by this arrangement, other pairs may be used for these keys, such as (106,105) for key 311, as this is the longest arrow available, (106,103) for key 310 and (105,104) for key 312.

The above embodiments and methods may be combined with various other methods of data entry, such as any of the methods described in PCT publications WO01/131788, titled "Integrated Keypad System", WO03/007288 titled "Features to Enhance Data Entry Through a Small Data Entry Unit", WO04/095414 titled "System to Enhance Data Entry in Mobile and fixed Environment", WO05/122401 titled "System to Enhance Data Entry in Mobile and fixed Environment", WO07/114833 titled "Data Entry System", PCT application IB2007/004481, and PCT application IB2008/002263, the disclosures of all of which are incorporated herein by reference in their entirety.

As another example, when a user indicates a previously entered word, which was not completely disambiguated yet, the system may display for selection one or more words which may be used instead of the word guessed by processor 130.

In some embodiments of the invention, the number of keys provided included in a data entry device may be reduced by differentiating between different finger orientations when pressing keys. For example, the keys may be soft keys and in identifying pressing on the screen the area of the pressing finger is determined to identify whether the press was performed with the tip of the finger or the flat portion.

In some embodiments of the invention, a speech recognition system is added to the above described data entry systems or is used instead of the second group keys. For example, in order to enter a special character, the user may press (e.g. and-hold) the key corresponding to the group that includes the desired character and provide a predefined speech relating to the desired symbol such as speaking said symbol so the system recognizes said speech and selects the desired symbol among that group of symbols. Preferably, for better accuracy, symbols included in a single symbol group, do not have similar names which can be confused in speech recognition. Allowing the user to use either second group keys or speech signals, allows the user flexibility to accommodate to different environments (e.g. very noisy environments, in a meeting).

The data entry system of the invention may include a plurality of modes (e.g. instances) wherein in each mode the input interface of the system may change its appearance and/or may require a different type of interaction. Some of those modes are described hereafter:

1) The data entry system of the invention is a word predictive system therefore its main mode is the (e.g. Letter) Predictive Mode. Fig. 101A shows a device 101000 having a touch screen 101009, and an on-screen keypad of the invention having the keys 101001-101004 to which the letter of an alphabet are assigned, and two more keys, 101005 including the backspace function, and 101006 including the space character. By tapping on the appropriate keys relating to d desired word, the word predictive system of the invention as described earlier may predict a word from a corresponding word database.

2) According to one method, if a user touches a letter key and provides a gliding action from a touching point towards a letter printed on/from said key, said letter may be entered precisely. The key of the keypad in the instance of the system in which a gliding action is provided for entering a character precisely may be considered as a second keypad of the invention, because providing a gliding action in each different predefined direction may correspond to a different letter (e.g. printed) on it. According to one method, when a user provides a gliding action on a key, said key is enlarged. Preferably, when a user touches a key to provide a gliding action, a copy of said (e.g. original) key may be located under the user's finger wherein the center of said key is located at the user's touching point. By doing so, the direction of the gliding action from a touching point towards a predefined letter on a key remains the same regardless of the user's finger touching point on the (e.g. original) key. To enter precisely a character that is assigned to the center key/zone, a predefined interaction such as a short gliding action or a back-and-froth gliding action (from the touching point) may be provided.

During the entry of a word, providing one or more gliding actions (e.g. 101010) corresponding to one or more precise characters in addition to providing one or more tapping actions on the keys (e.g. ambiguous key presses) may help the system to better predict a word. Said gliding actions may be provided before all of the input information corresponding to the characters of a word is provided (i.e. "Insertion" procedure) and/or after all of the input information corresponding to the characters of a word is provided and the system predicted a non desired word (i.e. "Correction" procedure), so that the system may predict another word corresponding the input information and including the at least one precise character (e.g. "Combined" input information). It must be noted that in addition the input information corresponding to the letters, the combined input information may include the position of each letter within a word being entered,

3) According to one embodiment, if the user presses-and-holds one of the (e.g. letter) keys of the keypad, then, as shown as an example in fig. 101B, the system may enter into the

Special Character Mode. In this case, according to one method, at least the key being pressed- and-held may change its appearance and may display the special characters assigned to said key in that mode. According to one embodiment, providing a gliding action 101110 from the user's touching/pressing point on said key towards a special character on said key may enter said special character precisely.

4) According to one embodiment, a predefined interaction, such as providing a predefined gliding action on the backspace keys, may enter the system into the/a Precise Letter Mode, wherein a/each letter key of the keypad may be replaced by a keypad, wherein tapping on any of at least some of its keys may correspond to entering precisely a letter on said key. Each of said replacing keypads is described as a second keypad of the invention throughout this application. Fig. 101C shows as an example, four second keypads 101201-101204 of the invention in the Precise Letter Mode. According to one embodiment, in some cases which some of them will be described later in this application, the system may automatically enter in the Precise Letter Mode.

5) According to one embodiment, a predefined interaction, such as providing a predefined gliding action on the backspace keys, may enter the system into the/a Precise Special Character Mode, wherein a/each letter key of the keypad may be replaced by a keypad, wherein tapping on any of at least some of its keys may correspond to entering precisely a special character on said key. Each of said replacing keypads is described as a second keypad of the invention throughout this application. Fig. 10 ID shows as an example, four second keypads 101301-101304 of the invention in the Precise Letter Mode. According to one embodiment, in some cases which some of them will be described later in this application, the system may automatically enter into the Precise Special Character Mode.

Preferably, substantially all of the letters of a language are assigned to four of the keys (e.g. 101001-101004) of the keypad. As mentioned, the keypad may preferably also include a space key 101006, and a backspace key 101005. Preferably, the letter keys (e.g. 101001- 101004) are arranged such that to form two columns of letter keys. Preferably, the space key is positioned on the side of a first column of letter keys and the backspace key is positioned on the side of a second column of letter keys. Optionally, each of said space or the backspace keys may form a column of keys together a first and a second column of letter keys, respectively.

Fig. 301 shows as an example, an exemplary first keypad 301000 as described above wherein to four of its keys 301001-301004 all of the letters of the English language are distributively assigned. The keypad 301000 also includes two additional keys 301006, 301005 to respectively provide Space character and Backspace function.

Fig. 301 also shows an exemplary second keypad 301010 of the invention which in this example is a telephone-type keypad wherein all of the letters of the English language are also distributively assigned to eight of its keys.

As an example, by considering the portion of the database of words 301009 used by the system, in order to enter the word "day" the user may press on ambiguous keys 301004, 301002, 301003. The system may propose/predict the word "was" which corresponds to said key presses and has the highest priority among the words of the database corresponding to said key pressing actions. At this time the user may proceed to the correction procedure of the invention and press on the key 301013 of the second keypad 301010 to which the first character (e.g. "d") of the desired word id assigned. By considering the key pressing action corresponding to the first character of the desired word provided on the key 301001 of the first keypad 301000 and on the key 301013 of the second keypad 301010, the systems understands that the only common character (e.g. "d") on said keys is precisely the first character of the desired word. By considering said character and the key press information corresponding to the remaining characters of the desired provided through the first keypad the system proposes the word "day" which corresponds to the combined information provided by the user and has the highest priority among the words of the database corresponding to said combined information.

As described above, the interaction with the keys of a keypad may be duplicated of imitated by providing tapping, gliding or swiping actions preferably on a touch sensitive surface or in the air. For example, if the device relating to the data entry system of the invention has a touch sensitive surface and the second keypad of the invention is an imaginary telephone-type keypad model, according to one embodiment of the invention, providing a tapping action on any location on said surface may duplicate a taping action on the center key of said keypad to which the characters "JKL5" are generally assigned. In order to duplicate an interaction with another key of said keypad, the user may provide a gliding action on the touch sensitive wherein said gliding action duplicates a gliding action departing from the center key of the imaginary second keypad towards the desired key of said keypad.

As an example, fig. 301 shows a device 301030 having a touch screen 301038 and using the data entry system of the invention. Said device includes/uses the first keypad being split in two portions wherein the first portion includes the keys 301031, 301033, 301035 duplicating the keys 301001. 301003, 301005 of the keypad 301000, and the second portion includes the keys 301032, 301034, 301036 duplicating the keys 301002, 301004, 301006, of said keypad 301000. The device 301030 also uses an imaginary keypad model 301010 which is in the form of a telephone-type keypad and is used as the second keypad. In the example of the entry of the word "day", after the user presses the corresponding keys of the first keypad of the device, the system may propose the word "was", In order to duplicate/imitate the pressing action on the key 301013 of the imaginary second keypad, the user may provide the gliding action 301037 that duplicates a gliding action 301021 departing from the center letter key 301015 of the imaginary keypad model 301010 towards the desired key 301013.The system recognizes that the user interacts with the key 301013 of the imaginary keypad model, and based on the principles as described in detail, proposes the word "day".

Also as an additional example, the gliding action 301038, duplicates the imaginary gliding action 301022, which corresponds to interacting with the keys 301019 of the imaginary keypad model 301010.

Symbols such as special characters, commands, and functions, of PC keyboard and/or customized, may be grouped in different categories based on their characteristics (e.g. digits 0-9, punctuations, arithmetic characters, PC commands, etc.). Each group may be assigned to a different mode instance provided by providing a predefined type of interaction (e.g. a long, double, strong, pressing action) other than another the predefined type of interaction (e.g. a short, soft, tapping/pressing action) which is used/required with the keys of at least the first keypad for entering ambiguous characters such as letters. According to one embodiment, after the system enters into a corresponding mode, the system shows each of the symbols of the corresponding group on one of the keys of an on-screen keyboard for example in the form of a telephone-type keypad. The user then may provide a pressing action, or a gliding action on or corresponding to the desired key shown on the screen. Figs. 302A to 302F are shown to describe as an example the "mode" keys of the invention in detail.

Fig. 302A shows as an example, the first keypad 302000 similar to the one shown in Fig. 301. In this example, the special symbols used with the data entry system are grouped in six categories, "TEXT" (e.g. mostly punctuation mark characters), "NUMB" (e.g. mostly digits 0- 9), ARITH (e.g. mostly arithmetic characters), "RARE" (e.g. mostly other and/or rarely characters), "FUNC" (e.g. mostly PC and/or customized functions), and "F1-F12". Each of said groups is assigned, respectively, to one of the keys 302001-302006 of the first keypad 302000. The appellation of each group is printed on the corresponding key. As an example, if the user desires to enter a punctuation mark character "?" which belongs to "TEXT" group, according to one embodiment, he may press and hold the key 302001 for a predefined laps of time to enter the system in the mode instance corresponding to said group. The system may show each of the characters of said group on one of the keys of an on- screen (e.g. telephone-type) keypad 302100. In this example, the symbol "?" is shown on the key 302108. According to one method, in order to enter the symbol "?", if the screen is touch sensitive, then the user may press the key 302108 or he may provide another interaction such as a gliding action to duplicate said key interaction. If the screen is not touch sensitive, and the device uses a physical telephone-type keypad as the second keypad, then the user may press the corresponding key of said physical keypad.

Also as another example, in order to enter the symbol the user may press (e.g. and hold) the key 302001 and press then press the key 302101 if the screen is touch sensitive.

Figs. 302B to 302F show five more groups of symbols, corresponding to five more modes "NUMB", "ARITH", "RARE", ."COMMNAD", and "F1-F12", assigned to the keys 302002 to 302006, respectively For example, in order to enter arithmetic character "=", the user may press and hold the key 302003 corresponding to the "ARITH" mode of the first keypad 302000, and for example, after the system shows the second keypad 302300 as shown in fig. 302C showing the locations of the symbols of "ARITH" group on said second keypad, the user interacts with the corresponding key 302309 to enter his desired special character "=".

According to one embodiment of the invention, the words of the database of the system may include any symbol such as letters, numbers/digits, punctuation characters, etc. Obviously, in many cases such as when the first keypad of the invention has only few keys such as four keys, one second keypad of the invention may not be enough to include all of said ambiguous symbols/characters such that, preferably, any ambiguous key of the first keypad and any ambiguous key of the second keypad have at most one common symbol/character. For example, if the first keypad of the invention has four ambiguous keys only and the words of the database of the system include substantially any of the characters available on a PC keyboard then having at least an additional (e.g. at least one more) second keypad may become necessary. This is because each key of the first keypad may, preferably, include at most a number of characters which does not exceed the number of keys of the second keypad or vise-versus.

Using a device having a touch-sensitive surface wherein tapping/gliding actions on a first side of said touch-sensitive surface corresponds to interacting with a first second keypad of the invention, and tapping gliding actions on at least a second side of said touch-sensitive surface corresponds to interacting with at least an additional second keypad of the invention may permit the entry of any word having any (type of) character.

Fig. 303 shows as an example, a first (virtual) keypad of the invention 303000 having four ambiguous keys 303001-303004 as shown in many embodiments before. In this example, in addition to letters each key includes other ambiguous characters. For example, key 303001 includes the digits 0 to 9 (e.g. the "NUMB" group of characters 303011), key 303002 includes the arithmetic characters 303012 (e.g. the "ARITH" group of characters), key 303003 includes the text punctuation characters 303013 (e.g. the "TEXT" group of characters), and key 303004 includes the rare characters 303014 (e.g. the "RARE" group of characters). Preferably, the characters available on the keys of the first keypad are also distributed on the keys of two second keypads 303052, 303051, such that any pair of keys consisting of an ambiguous key of the first keypad and an ambiguous key of any of the second keypads, have at most one common character. In this example, in addition to the first keypad of the invention (e.g. split into two portions 303081, 303082 wherein its four ambiguous keys duplicate the ambiguous keys 303001-303004 of the virtual keypad 303000), the device also includes the two (e.g. virtual) second keypads 303051 and 303052 (e.g. which may be considered as two second keypad models). In this example, predefined tapping/gliding actions (e.g. 303087) on the right side of the touch screen of the device 303080 may correspond to interacting with the keys of the second keypad 303051 (e.g. the first second keypad), and predefined tapping gliding actions (e.g. 303088) on the left side of the touch screen of the device 303080 may correspond to interacting with the keys of the second keypad 303052 (e.g. the additional/second second keypad).

As an example of the above mentioned embodiment, fig. 303a shows a device 303100 having the data entry system of the invention that includes a (split) first keypad of the invention as shown before, wherein four of its keys 303101-303104 duplicate the four ambiguous keys 303001-303004 of the keypad model 303000. The system also includes two (e.g. virtual) second keypads 303111 and 303112, wherein in this example the first second keypad 303111 mostly includes the letters of the alphabet and the second (e.g. additional) second keypad 303112 includes many of the special symbols distributed on its keys as described above. By considering the portion 303190 of the database of the system, according to a first example, in order to enter the word "f+9$" that is included within the dictionary database, the user may first press the corresponding keys 303102, 303102, 303101, 303103, of the first keypad corresponding to the characters of the desired word. The system may propose the word "talk" corresponding to said pressing actions and having the highest priority. The user may provide a gliding action (not shown) on the right side of the screen to duplicate interacting with the key of the second keypad that corresponds to the first character (e.g. "f") of the desired word. In this example, the only word that corresponds to said combined information is the word "f+9$". The system proposes said word.

With continuous description of the embodiment, according to a second example, in order to enter the word " #m6$ " which does not exist in the dictionary, the user may first press the corresponding keys 303102, 303102, 303101, 303103, of the first keypad corresponding to the characters of the desired word. The system may propose the word "talk" corresponding to said pressing actions and having the highest probability. At this time the user begins to reenter said word through the keys of the second keypads. For this purpose, the user:

-first provides a gliding action 303141 on the left side of the screen that corresponds to interacting with the key 303181 of the corresponding second keypad 303112. The only common character between the key of the 303102 and the key 303181 is the character "#". The system enters said precise character.

-the user then provides a gliding action 303142 on the right side of the screen that corresponds to interacting with the key 303196 of the corresponding second keypad 303111. The only common character between the key of the 303102 and the key 303196 is the character "m". The system enters said precise character.

-the user then provides a gliding action 303143 on the left side of the screen that corresponds to interacting with the key 303186 of the corresponding second keypad 303112. The only common character between the key of the 303101 and the key 303186 is the character "6". The system enters said precise character.

-the user then provides a gliding action 303144 on the left side of the screen that corresponds to interacting with the key 303184 of the corresponding second keypad 303112. The only common character between the key of the 303103 and the key 303184 is the character "$". The system enters said precise character.

The desired word is being displayed on the screen and preferably added to the word database of the system so that next time that the user desired to enter said word, it can be entered faster.

QWERTY arrangement

According to one embodiment of the invention, the first keypad of the invention may have six letter keys arranges in two columns and three rows. The distribution and arrangement of the letters on said six keys may be such that to resemble to the arrangement of the letters on a PC keyboard such as a QWERTY keyboard. According to one method, the second keypad of the invention may be a keypad having ambiguous letter keys such as a telephone type keypad with .

As an example, fig. 304 shows the letter keys of a QWERTY keyboard 304010 wherein said letter keys are divided into two columns 304001-304002 (e.g. by a separator 304008) and three rows forming six group of letters 304011-304016. The letter keys are separated in two columns such that the keys on the left column are usually pressed by the left hand of a typist and the keys on the right column are usually pressed by the right hand of a typist. Fig. 304 also shows a first keypad of the invention 304020 having six ambiguous letter keys 304021-304026 wherein each of said keys representing ambiguously one of the corresponding letter groups 304001-304006, accordingly. As an example, a typist knows that the letter "D" is on the middle row of the letter keys of the QWERTY keyboard and that he usually enters it through his left hand, therefore in order to enter said letter ambiguously through the keypad 304020 he will press the middle left key 304023 of said keypad.

The keypad 304020 may be used as the first keypad of the invention used with the word predictive data entry system of the invention having a first and (at least) a second keypad. As shown in the current figure, in this example, the second keypad of the invention is a telephone- type keypad with slight modifications 304030 such that each of the keys of the first keypad and each of the keys of the second keypad have at most one common character/letter.

In the embodiment above, although the first keypad has more than four letter keys (e.g. 6 letter keys), because the arrangement of the letters on said six keys resemble to the arrangement of the letters of a PC keyboard such as the QWERTY keyboard, therefore the user can begin to type fast and easy immediately without the need of a learning period to get used to the locations of letters on the keys of the first keypad. On the other hand, the location of letters on the keys of the second keypad which preferably is a telephone-type keypad (with slight modifications) is also a familiar layout, therefore the overall system may become very user friendly. It is understood, that the first keypad af the invention may include additional keys such as two keys for entering Space and Backspace characters.

In the example above, which is only one example of such layout modification, in the keypad 304030 the location of the letters in each of the pair of letters "F" 304032 and "J" 304031, "X" 304036 and "K" 304035", "N" 304034 and "Q" 304033, are being swapped between them.

Alternatively, according to another embodiment of the invention, the first keypad of the invention may have six ambiguous letter keys wherein the letter distribution may remind a PC keyboard such as a QWERTY layout with slight modification, and the second keypad may be a telephone-type keypad with no letter arrangement modification. Fig. 304a, shows an example of such embodiment wherein the letter assignment to the keys of the first keypad 304040 resemble to the letter assignment to the keys of the first keypad 304020 of fig. 304 with slight modification. In this example, the location of the letters in each of the pair of letters "M" 304047 and "W" 304048, "X" 304046 and "K" 304045", "C" 304041 and "Q" 304042, "F" 304044 and "J" 304043, on the first keypad are being swapped between them, accordingly. In this example, the letters being swapped between them have resembling shape so that the user can easily get used to the layout change. Also, in this example, the second keypad is a telephone-type keypad 304050 with no modification.

Alternatively, according to another embodiment both the first keypad having six keys reminding the arrangement of the letters on a PC keyboard and the second keypad of the invention being a telephone type keypad may be slightly modified in their respective letter arrangements.

Entering (directly) Precise Letters

According to one embodiment of the invention, the data entry system of the invention

(e.g. using a first and a second keypad) may include a method for entering precise characters (e.g. directly) without entering into the correction procedure of the invention. If the keys of the first keypad of the invention are touch sensitive keys, then a gliding swiping action on and/or departing from an ambiguous key of the (first) keypad to which a desired letter/character is assigned, duplicating a gliding action departing from the center letter key (e.g. the key to which the digit "5" is assigned) of an imaginary second keypad such as a telephone-type keypad) towards a key to which also said desired letter is assigned may provide the desired letter. Because said key of the first keypad and said key of the imaginary keypad have at most one character/letter in common said character said desired character may be precisely entered by the system. In the current embodiment, when a gliding action is being provided, the touching point (e.g. the beginning point of the gliding action) on said key may be interpreted by the system as corresponding to the center letter key of the imaginary telephone keypad to which the digit "5" and/or the letters "JKL" are generally assigned, and based on that the gliding direction and the corresponding key of the second keypad is recognized. Obviously, the departing point may be anywhere on the corresponding key of the (first) keypad.

With continuous description of the current embodiment, fig. 305 shows as an example, a device 305000 having a touch sensitive surface including a first keypad of the invention as shown and described previously in this application, having four ambiguous letter keys 305001- 305004. The keypad also has two additional keys 305005, 305006 as described before (e.g. for Sp and BkSp, mode keys, etc.)

In this example, in order to enter the letter "a" precisely (and directly), the user may provide a gliding action 305011 on and/or departing from the ambiguous key 305002 to which the letter "a" is ambiguously assigned duplicating a gliding action departing from the center letter key of a telephone-type keypad towards the key of said telephone-type keypad to which the letter "a" is ambiguously assigned. Also for example, in order to enter precisely the letter "b", the user may provide a gliding action 305012 on/departing from the key 305001 upward, and in order to enter precisely the letter "n", the user may provide a gliding action 305013 on/departing from the key 305004 to the right direction (e.g. from the center key of the telephone keypad towards the key of said keypad having the digit "6" and the letters "mno"). In the example of fig. 305, some other gliding actions departing from different keys in several directions for entering the letters "E", "F", "S", "O", and "U" are also shown as examples.

With continuous description of the current embodiment, according to one method, in order to enter any on the letters "jkl" which are located on the center letter key of a telephone keypad, the user may provide a long pressing action on the corresponding key of the first keypad and then release it. For example, a long pressing action on the key 305001 and releasing said key may correspond to entering precisely (and directly) the letter "1", and a long pressing action on the key 305003 and releasing said key may correspond to entering precisely the letter "k", and a long pressing action on the key 305004 and releasing said key may correspond to entering precisely the letter "j".

According to another method, as described in this and previous patent applications filed by this inventor, the letters "jkl" may be assigned to the key of the telephone-type keypad (e.g. used as the second keypad of the invention) to which the digit "1" is assigned. In this case, in order to enter precisely any of the letters "JKL", the user may provide a gliding action on and/or departing from the corresponding key of the first keypad in a direction duplicating a gliding action departing from the center key of an imaginary telephone-type keypad towards the key to which the digit "1" is (e.g. and in this method, also the letters "JKL" are) assigned.

With continuous embodiment of the invention, as an example, fig. 305a shows the device 305000 having a touch-sensitive surface and a imaginary second keypad 305110 wherein the letters "JKL" are assigned to the key 305111. In this example, providing the gliding action 305107 departing from the key 305004, which its trajectory duplicates a gliding action departing from the key 305115 of the imaginary second keypad toward the key 305111of said keypad may provide precisely the letter "J". Also, providing the gliding action 305108 departing from the key 305003, which duplicates a gliding action departing from the key 305115 toward the key 305111 of the imaginary second keypad may provide precisely the letter "K". Also, providing the gliding action 305109 departing from the key 305001, which duplicates a gliding action departing from the key 305115 toward the key 305111 of the imaginary second keypad may provide precisely the letter "L".

Obviously, the method of entering (directly) precise characters as described above may be used with the word predictive data entry systems of the invention. As an example, during the entry of a word the user may enter precise characters without entering into a correction procedure for entering said precise characters.

According to one embodiment of the invention, during the entry of a word using the word predictive system of the invention using a first and (at least) a second keypad of the invention, providing gliding actions on and/or departing from the letter keys of the first keypad of the invention may provide corresponding precise characters, and providing gliding actions outside said letter keys may correspond to the correction procedure of the invention. As an example, Fig. 306a shows a device 306000 and features resembling to the device 305000 and the features shown in fig. 305. Fig. 306 also shows a portion of an exemplary dictionary 306008 of the invention. As an example, after the user provides two (fast) pressing/taping actions on the ambiguous key 306002, the system may propose the word "at" which corresponds to the input signals provided by the two taping actions (e.g. the first input information) on said key 306002. The first letter 306012 of the proposed word may automatically be selected/marked by the system. At this time, if the user provides a gliding action 306011 departing from the key 306003 wherein said gliding action duplicates a gliding action departing from the center letter key of a telephone type keypad towards the key to which the letters "PQRS" are assigned, then the system may relate said gliding action 306011 to providing the letter "s" and adds said information to the first input information provided previously. By considering/combining the first input information ambiguously corresponding to the two beginning characters of the desired word and the third character which is the precise letter "s" which is the third character of the desired word, then as shown in fig. 306b, the system may propose the word "mrs" which corresponds to said combined information and has the highest priority among the words of the dictionary corresponding said combined information. Preferably, the system may keep to select/mark the first ambiguous letter 306112 of the proposed word. At this time, if the user provides a gliding action departing from out of the letter keys of the first keypad such as the gliding action 306111 towards a desired direction (e.g. in this example, upward), then as shown in fig. 306c, the system predefinely may enter into the correction procedure of the invention and may relate said gliding action to the ambiguous letter selected/marked by the system and accordingly provide the letter "a". By considering the combined information provided by the user until then, the system may provide the word "ats" which corresponds to said combined information and has the highest priority among the words (and/or stems) corresponding to said combined information. At this time, preferably, the system selects/marks the next ambiguous letter 306212 of the proposed word. Obviously, if desired word is longer than the proposed word, the user may provide additional ambiguous key presses information, and/ or additional (direct, or through correction procedure) precise characters corresponding to his desired word.

According to one embodiment of the invention, the keys of the first keypad of the invention may also be used for entering (directly, as described above) special symbols such as special characters, commands, etc., by providing the corresponding gliding actions on and/or departing from said keys.

As an example, after the user enters the system into a desired mode instance by for example providing a long pressing action on the corresponding key of the first keypad, preferably without removing his finger from the screen, he may provide a gliding action corresponding to the desired symbol on and/or departing from said key in the corresponding direction.

According to another embodiment of the invention, a short gliding action on and/or departing from a key (of the first keypad) in a predefined direction may correspond to entering a corresponding precise letter, and a longer gliding action on and/or departing from a key (of the first keypad) in the same predefined direction may correspond to entering a corresponding special symbol in a mode instance assigned to said key.

After the system enters into a mode instance, the system may show the symbols of said mode instance and their corresponding location on the second keypad of the invention (e.g. a telephone keypad), on the screen of the device so that the user see his desired symbol and provide the corresponding gliding action.

According to one method, if the letters "JKL" are assigned to the center key of the second keypad of the invention (e.g. a telephone type keypad), then providing a long pressing action on a key of the first keypad and releasing it may correspond to the similar character which is also on said key of the first keypad. In this case the mode instance may not include a special symbol on the center key of the second keypad of the invention.

According to another method, if the letters "JKL" are not assigned to the center key of the second keypad of the invention (e.g. a telephone type keypad), then providing a long pressing action on a key of the first keypad and releasing it may correspond to a special symbol which is predefinely assigned to the center key of the second keypad of the invention in the corresponding mode instance.

According to one embodiment of the invention, the system may have two (e.g. identical) sets of first keypads, one used for entering ambiguous characters and one used for entering precise characters in the manner described above. The first keypad for entering ambiguous characters may include hard keys and the first keypad for entering precise characters may have touch-sensitive keys such as touch pad keys or on-screen keys (e.g. or vise versus). According to one design, the hard keys may be click buttons for entering ambiguous characters and the soft keys may be touch pad keys located on the top of (e.g. and around) said hard keys for entering precise characters as described.

The system of entering precise characters based on gliding actions on and/or departing from each of the keys of the first keypad of the invention as described above may be extremely beneficial for entering out-of-dictionary words. The user may enter a word (e.g. that he knows it is not included within dictionary) precise character by precise character, by a single swiping action (e.g. a long pressing action if the character is on the center key of the second keypad) provided for each of said characters. Said word then may automatically be added to the dictionary.

Dialing

Dialing by text has been described in different patent applications by this inventor. Obviously, searching a target such as a person (e.g. based on different text such as his name, address, etc.) may be done through the media search described earlier and/or by using the system for entering direct precise characters as described above. According to one method, based on said text information provided, the system implemented in a device may first search the target in the device's database and if it does not find it may send said text information to the remote center for the same purpose. The next steps has been described and been shown by examples before.

Entering Precise and Ambiguous Special Characters

Method 1 -Entering precise special characters which are not part of a word

As mentioned before, quick tapping/pressing actions on a (letter) key of the first keypad of the invention may be related to entering ambiguous letters/characters assigned to said key, and a long pressing action or a gliding action on or departing from said/a key, wherein said gliding action 1 preferably begins shortly after the user touches said key, may correspond to entering a precise character such as a letter assigned to said key. Also as mentioned before, according to one embodiment of the invention, a long pressing action on a key may enter the system into a mode instance corresponding to said key. Then, if the user continues his interaction (e.g. preferably without removing his finger from the key/touch-sensitive-surface) by providing a gliding action in a predefined direction, preferably when the user ends his gliding action, said gliding action may be related by the system to entering a corresponding precise symbol such as a special character assigned to said key.

With continuous description of the current embodiment, as an example, fig. 307 shows a device 307500 with a touch screen which includes the keys 307501-307506 of the first keypad of the invention. In this example, if the user provides quick pressing actions on the letter keys 307501-307504, the system relates said actions to entering a word ambiguously. For example, by considering the portion 306008 of the word database of the system as shown in fig. 306a, if the user provides quick pressing actions on the keys 307502, 307502, 307503, the system proposes the word "may" corresponding to said input information and having the highest priority among the words corresponding to said input information. On the other hand, according to an example of the current embodiment, if the user presses and holds the key 307501 for at least a predefined laps of time (e.g. longer than said quick pressing actions), then the system may enter into the mode instance (e.g. the A ITH mode) which in this example corresponds to said key. In this example, the second keypad of the invention is based on an imaginary telephone-type keypad model. The imaginary keypads 307110, 307210, 307310, 307410, show respectively the symbols (e.g. special characters) assigned to the keys 307501, 307502, 307503, 307504, in the corresponding mode instances (e.g. ARITH, NUMB, RARE, TEXT, respectively) of the imaginary keypad model. The symbols of the ARITH mode are shown on the keys of the keypad instance 307110. After the system enters into said mode, according to one example, if the user (e.g. continues his interaction and) provides a gliding action 307122 to duplicate interacting with the key 307102 of the keypad model in said ARITH instance, then the system enters the special character "-" assigned to said key. Accordingly, after the system enters into said ARITH mode, according to another example, if the user (e.g. continues his interaction and) provides a gliding action 307123 corresponding to the key 307103 of the keypad model in said instance, then the system enters the special character "+" assigned to said key. Also accordingly, as an additional example, after the system enters into the TEXT mode by pressing and holding the key 307504 of the first keypad, according to one example, if the user (e.g. continues his interaction and) provides a gliding action 307429 corresponding to the key 307409 of the keypad model 307410 in the TEXT instance, then the system enters the special character "@" assigned to said key. Fig. 307 also includes additional examples for entering some other special characters.

Preferably, the current embodiment may be used for entering precise special characters which are not part of the characters of a word. For example, at the end of the entry of the phrase "Good Luck" the user may enter the special character "!" 307508 by providing a predefined interaction such as press and holding the corresponding key 307504 for entering the system into the TEXT mode instance and provide the gliding action 307428 (e.g. corresponding to interacting with the key 307406 of the imaginary keypad in the TEXT mode instance) departing from said key 307504.

It must be noted that according to one method, the special character/symbol assigned to the center key of the second keypad in one or more of the instances may be entered by providing a different interaction such as for example, a back and forth gliding action ending on or closed to the departing point of the gliding action on or departing from said key, etc.

Method 2 -Enterim precise special characters which are part of a word

According to one embodiment of the invention, Method 1 described above may be slightly modified and be used for entering quickly/directly precise special characters during the entry of a word that includes special characters through the word predictive data entry system of the invention. According to one method, such method may be similar to said Method 1 for entering a precise special character, but with an additional interaction before/during/after each interaction used for entering a symbol/precise character through Method 1 (or vise versus). As an example, a predefined interaction such as press and holding action with/on a letter key of the first keypad, then press and holding another key of the first keypad may enter the system into the mode instance assigned to said another key wherein said mode instance is used for entering precise special characters of a word during the entry of said word through the word predictive system of the invention. The user then may provide a gliding action on or departing from said another key such that to duplicate a gliding action departing from the center key of the second keypad towards the key of the second keypad to which said symbol/special character is assigned. As an example, by considering fig. 307, during the entry of a word through the word predictive system of the invention, if said word includes the special character ")" which is assigned to TEXT mode instance and the user desires to enter said character precisely, according to one method, the user may first provide an initial interaction such as press and holding a key other than the key 307504 to which the TEXT mode instance is assigned. The user then may press and hold the key 307504 to enter the system into the TEXT mode instance for entering precise special characters of the desired word to be entered, and then provide the gliding action 307426 corresponding to the desired character ")" which is assigned to the key 307406 of the imaginary keypad model in the instance 307410. Said special character along with other input information such as ambiguous key presses provided for the entry of the desired word may be considered by the system to predict the desired word.

According to one embodiment, preferably, the Method 2 just described is preferably used for adding precisely a special character in a location where the caret/cursor is located during the entry of a word.

Other methods

Creating, entering and correction procedure for entering words including special characters through the word predictive system of the invention have been described before and shown through figs 303-303a. Said method requires interacting on different zones for entering different categories of characters. The method described and shown before in figs 303-303a is preferably used where the In- Word character selection means is located to for example replace an ambiguous character with a precise special character. If said system is used in conjunction with the Method 1 and/or Method 2 described above in the system, then, the departing point of interaction/gliding location in the method shown through the figs 303-303a may preferably be outside the zones of the keys of the first keypad of the invention.

According to an alternative method (e.g. hereafter called the Alternative Correction Method for Entering Precisely Special Characters) providing a predefined interaction on a key of the first keypad for entering the system into a desired mode instance and then providing a gliding action corresponding to a special character in said instance provided from outside the zones of the first keypad of the invention may be used by the correction procedure of the invention for replacing a selected ambiguous character (during the entry of a word) by said precise special character.

As a first example of the embodiments and methods for entering (precise characters) just described, by considering the portion 303190 of the dictionary of words of the invention of fig. 303a which includes words with special characters, according to a first example, in order to enter the word "f+9$", the user may first (short) press the keys 307502, 307502, 307501, 307503. The system may propose the word "talk" having the highest priority and automatically select the letter "t". At this time, for example, the user may navigate within the letters of the proposed word and select the ambiguous character "a". At this time, according to the Alternative Correction Method for Entering Precisely Special Characters as described above, the user may proceed to the correction procedure of the invention in order to replace the character "a" by the precise character "+" of the desired word. The user may enter the system into the corresponding mode instance by providing a long pressing action on the key 307501 which corresponds to ARITH mode, and provide a gliding action 307511 on the screen of the device departing from outside the zones of the first keypad of the invention, wherein said gliding action being provided in the direction corresponding to duplicating interaction (described in detail before) with the key 307103 of the second keypad of the invention 307110 in the ARITH mode to which the special character "+" is assigned. By considering the combined input information the system replaces the character "a" by character "+" and may propose the word "f+9$" which is the only word of the database that corresponds to the combined input information.

As a second example of the embodiments and methods just described, according to Method 2, in order to enter the word "f+9$", the user may first provide a (short) pressing action on the key 307502 to provide a first character ambiguously. He then press and hold a (letter) key of the first keypad (other than the key 307501) such as the key 307502, then provide a long press on the key 307501 to enter the system into the corresponding mode which in this example is the ARITH mode. At this time the user may continue his interaction by providing a gliding action 307123 on or departing from the key 307501 in the direction that duplicates a gliding action departing from the center key 307105 of the imaginary second keypad of the invention in ARITH mode instance towards the key 307103 to which the special character "+" is assigned. The system enters said character "+" precisely. The user then may (short) press on the keys 307501 and 307503 (e.g. corresponding to the two remaining ambiguous characters of the desired word). By considering the combined information provided by the user for entering the desired word the system predicts the word "f+9$" which is the only word of the dictionary that corresponds to said combined information.

It must be noted that the different methods described for entering precisely special characters can be swapped and assign to any of the interactions as described above or to other types of interactions designed by people skilled in the art.

According to one embodiment of the invention, during the entry of a word, if the user moves (manually) the In- Word selection means (moving a selection means inside a word has been described before in detail) arid selects a letter/character of a word, then providing a precise character/letter or a precise special character may (predefinely) replace said selected character by said precise character and the system may consider said replacement and may propose/predict another word based on the overall input information corresponding to said word provided until that moment.

The methods of entry of one or more precise characters at any moment and any (cursor/caret) position during the entry of an ambiguous word through a word predictive system may resolve the problem of fluctuation of non desired words/stems proposed by the system during the entry of a desired word. For example, by entering (easily and directly) one or more precise letters of a word, in most cases the system may be able to predict correctly other characters of the desired words during the entry of the desired word. This greatly overcomes the fundamental problem of the word predictive systems which is characters fluctuations during the entry of a desired word. In addition, the system permits to easily enter short/customized word created by the users of SMS, formulas, passwords, URLs, etc.

It is understood that the second keypad may be any familiar keyboard such as a QWERTY keyboard.

Long/short glide for Precise letter/sp char

According to another embodiment of the invention, a short gliding action on and/or departing from a key of the first keypad (and preferably ending outside said key) in a predefined direction may correspond to entering a corresponding precise letter, and a long gliding action on and/or departing from a key of the first keypad in the same predefined direction may correspond to entering a corresponding special symbol in a mode instance assigned to said key.

According to one method, a very short gliding action on or departing from a key may correspond to the corresponding letter on the center key of the second keypad of the invention. For example, if the second keypad is a telephone-type keypad, then said very short gliding' action on a key of the first keypad may correspond to the corresponding letter among the letters "JKL". In the current embodiment, if a key of the first keypad-has em of the letters of the center key of the second keypad, then preferably the second keypad in the mode instance corresponding to said key of the first keypad may not include a special symbol at its center key. On the other hand, if a key of the first keypad does not include any of the letters of the center key of the second keypad, then a mode instance (e.g. "NUMB" mode) having a special character such as a digit (e.g. digit "5") on the center key of the second keypad may be assigned said key. In this case providing said very short gliding action on said key may correspond to special character on said center key. The current embodiment may eliminate the need of a timer for calculating a long press for entering the system into a mode instance.

With continuous description of the current embodiment, as an example, fig. 317 shows a device 317026 resembling to the device 307500 of fig. 307, By considering a telephone-type keypad as the virtual/imaginary second keypad of the invention, an by also considering the mode instances of fig. 307, as an example, providing the short gliding action 317012 departing from the key 317001 may correspond to entering precisely the letter "b", and providing the long pressing action 317011 from said key in the same direction may enter the character In this example, providing the very short gliding action 317025 (e.g. in any direction) on/from said key 317001 which includes the letter "1" may correspond to entering precisely the letter "1" which is also included on the center key of the second keypad (e.g. in this example, a virtual telephone- type keypad). Also as an example, providing the very short gliding action 317035 (e.g. in any direction) on/from the key 317003 which includes the letter "k" may correspond to entering precisely the letter "k" which is also included on the center key of the second keypad (e.g. in this example, a virtual telephone-type keypad).

In the above-mentioned example, the mode instance "NUMB" (e.g. digits 0-9) may be assigned to the key 317002 of the first keypad which does not include any of the letters (e.g. "jkl") of the center key of the second keypad. In this case as an example, providing the short gliding action 317016 departing from the key 317002 may correspond to entering precisely the letter "a", and providing the long pressing action 317011 from said key in the same direction may enter the digit "2". In this example, providing the very short gliding action 317015 (e.g. in any direction) on/from said key 317002 which does not include any of the letters (e.g. "jkl") of the second keypad may correspond to entering the digit "5" which is also included on the center key of the second keypad (e.g. in this example, a virtual telephone-type keypad).

Gliding actions in different directions on/from the keys other than the letter keys may be assigned to other characters/functions preferably those used frequently. As an example, a short gliding action upward 317016 on/from the key 317006 may correspond to a dot ("."), and a short gliding action downward 317018 on/from the key 317006 may correspond to a comma (","). Also as an example, a long gliding action upward 317021, a long gliding action to the left 317020, a short gliding action to the left 317017, on/from said key 317006, a short gliding action downward 317019, and so on, may correspond to other special character and/or functions such as "CapsLock", Return", etc.

It must be noted, that although short gliding action was explained to be used for entering letters, and long gliding action was explained to be used for entering precise characters, obviously, the short gliding actions may be used for entering a first group of characters/functions and long gliding actions may be used for entering another group of characters/functions and/or vise versa.

It must be noted that according to one method, the letter/special character/symbol assigned to the center key of the second keypad in one or more of the instances may be entered by providing a different interaction such as for example: a back and forth gliding action ending on or closed to the departing point of the gliding action on or departing from said key, or;

a long pressing action on said key of the first keypad,

etc.

^"t-T 5 Short gliding action

^ According to one embodiment of the invention, a taping action on any of the ambiguous

(e.g. letter) keys of the first keypad may correspond to an ambiguous character of said key, and a (e.g. very) short gliding action on or departing (e.g. in any direction, or in a predefined direction such as upward) from said key may correspond to entering a precise (e.g. certain)

10 character assigned to said key wherein said character is also assigned to the center key of the second keypad (e.g. the center key if said second keypad is a three by three matrix of keys, or the key "5", if it is a telephone-type keypad) of the invention.)

Also, according to one embodiment, providing a (very) short gliding action on or departing from (e.g. in any direction, or in a predefined direction such as upward) the center key

15 of the second keypad (e.g. a telephone-type keypad) of the invention may correspond to interacting with said key. For symbols on the other keys the user may provide a corresponding longer (e.g. longer than a predefined length) gliding action.

As an example, during the entry of a symbol such as a special character or a function, after providing a predefined action on a key of the first keypad corresponding to a desired

20 mode/group, the user may touch/press the center key of the second keypad (e.g. if said second keypad is a fixed keypad, then the user may press on said key, but if the second keypad is a virtual keypad model then said pressing action may be provided anywhere on the screen. This matter has already been described in detail). At this time, the system may show the symbols corresponding to said mode on the keys of a second keypad. If the desired symbol is on the

25 center key of the second keypad, then in order to enter said symbol the user may provide a short gliding action (e.g. in any direction such as the upward direction) on/from said center key. For symbols on the other keys the user may provide a corresponding longer gliding action as described before. According to one method, after the user touches/presses the center key, if he removes his finger from the screen without providing any gliding action, then according to one

30 method, nothing may happen (i.e., the system may not respond to that interaction).

According to another example based on the current embodiment, during the correction procedure, if the user taps/presses on the center key of the second keypad without providing a gliding action, the system may not relate said interaction to entering a character and eventually nothing may happen. In this example, if the user desires to interact with the center key of the second keypad he may be required to provide a predefined interaction such as a (very) short gliding action on/from said key as described above, or alternatively, he may provided a back- and-forth glisding action. According to another embodiment a press-and-holding action on a key of the first keypad of the invention may enter said key into said key/the system into the precise character mode. If the user removes his finger from the key without providing a gliding action, the system may relate said interaction as ambiguously corresponding to a special character on said key of the first and/or the second keypad (e.g. they may have said special characters in common). Accordingly, a short pressing (e.g. tapping) action may ambiguously correspond to a letter/character assigned to said first key.

According to another embodiment of the invention, if the user provides only a long pressing action on a key of the first keypad and removes his finger from the touch sensitive surface, then the system does not relate said interaction to entering a symbol and eventually does nothing.

Common Special Characters and Commands

Some frequently used Special Characters and Commands may be assigned to interactions such as taping or gliding action in different (predefined) directions on or from the keys (preferably other than the letter keys) of preferably the first keypad.

Fig. 319 shows as an example, a device having the first keypad of the invention having four letter keys, and two additional keys 319005 and 319006, to which some frequently used characters and commands are assigned. As an example, a pressing action anywhere on the key 319005 may correspond to the backspace ("Bk") function. Accordingly, a gliding action downward 319022 on/departing anywhere from said key may correspond to the Return (— ' ) function, and a gliding (e.g. and holding) action upward 319021 departing anywhere from said key may correspond to the Shift function (†) function. Also as an example, a pressing action anywhere on the key 319006 may correspond to the space ("Sp") function, a gliding action 319033 departing anywhere from said key towards lower left side may correspond to the comma ( " , " ) character, a gliding (e.g. and holding) action upward 319032 departing anywhere from said key may correspond to Shift function (†), and a gliding action to the left 319031 departing anywhere from said key may correspond to deleting one word/stem to the left of the cursor, etc.

It is understood that although in this example (e.g. because the keys 319005, 310006, are close to the edge of the device) only five symbols on each key are considered to each being assigned to a gliding action in a different direction on/departing from the corresponding key, obviously, more symbols relating to more different gliding directions (e.g. such as those 319045 shown in enlarged exemplary key 319015) may be considered for each key. For better view, the key 319006 is also shown in larger dimension 319016.

In the left and the right tables below, an exemplary schema of the symbols respectively assigned to the keys 319005 and 319006, and the exemplary manner to enter them are described.

- BkSp Bk Press on Key - Sp Sp . Press on Key

- l ck Word LW Changes all ; current - Undo Swipe

characters of current word - Shift

to Certain i^; -tr Swipe and hold + press

the letter key(s)

- Shift - - Swipe and hold + press

the letter key(s)

- Word BkSp Swipe and remove your

finger to erases on word.

- Tab Swi e and remo e y ur

h— Swi e and hold to repeat

finger for one.

Swipe and hold to rep eat . erasing words. :

- Enter < _*J Swipe and remove your • Swipe for on one word

finger for one. Swipe and hold to repeat

Swipe and hold to rep e t Swipe for one

- Del Del Swipe and remove your Swipe and hold to repeat .

finger for one.

Swipe and hold to rep eat :

In the examples in the tables above, according to one method, if the user provides one gliding action and at the end of the gliding action he removes his finger from the touch sensitive surface, then the system provides one symbol such as one "return" or one "tab" action. On the other hand, if at the end of the gliding action the user does not remove his finger from the touch screen surface then the system may repeat said symbol every one predefined laps of time until the user removes his finger from the touch sensitive surface.

Procedures of repeating a symbol such as a character or a function (e.g. moving a cursor/caret over several characters or up/down over several lines, repeating a Tab function several times, etc.) have been described before, and are being described in some more detail later in this application. It must be noted that functions/special characters and their locations on the keys as shown and described above are an exemplary embodiment of the invention to describe the principles of the concept. Other functions/special characters, and/or other locations on keys may be considered to assign functions/characters to the same or other keys.

Shift

According to one method, a SHIFT action for capitalizing one or more letters may be performed by providing a gliding action (e.g. in a predefined direction such as 319021, or 319032) on preferably a key that does not include ambiguous letters such as the key 319005 or the key 319006 (respectively), and (e.g. simultaneously) pressing a desired letter key. As an example, providing a gliding and holding action 319032, and (e.g. simultaneously/then) providing a gliding action 319011 for entering the precise character "b" may result in capitalizing said letter to provide the letter "B".

According to one method, if the user provides a shift gliding action (e.g. 319032) and lifts his finger from the touch sensitive surface without providing a pressing action on a letter key, then the shifting actions cancels and the system does nothing.

With continuous description of the current embodiment, according to one method, if the user provides the gliding action corresponding to the shift function and ends the gliding action without lifting his finger from the corresponding key, he may then provide more than one pressing actions on and or gliding actions on or from one or more letter keys to provide several corresponding capitalized precise and ambiguous letters, respectively.

There is an advantage in assigning the Shift command to a gliding action rather than to a taping action on a key. When the user is typing fast, between entering two ambiguous characters by taping on the corresponding keys, he may for example provide a pressing action on the Space key to which the Shift function is also assigned. Providing said three consecutive pressing actions may be ambiguous for the system because in many cases the user may not lift his finger from the screen/surface before he provides a next taping action. The system may not recognize the user's intention whether he desires to provide a space character between the two letters or whether he intends to capitalize the last letter. By restricting the user to providing a different type of interaction such as a gliding action for the shift on the space or BkSp keys (and not a pressing action), the system may easily recognize the user's intention.

Special characters and commands

As described before in detail, symbols such as special characters and commands may be grouped in different groups and depending on their characteristics and each of the groups being assigned to (a predefined interaction with a key of preferably the first keypad. Fig. 319a shows as an example, most of the symbols of a PC keyboard and some customized functions being grouped in different groups and being assigned to the keys of the first keypad printed on the keys 319001-319006 the first keypad. In this example, each of the characters of one group is represented by one key of a telephone-type keypad. In this example, some keys of the telephone keypad are not assigned to a symbol. In the example of fig. 319a, the group of F1-F12 functions is not included. To include such group or additional new groups, said groups may be assigned to either another additional key, or to another type of interaction with a key, etc. For example, instead of two command keys, there may be only one command key such as 318005, and the Fl- F12 group may be assigned to the other command key, for example to 318006. According to one embodiment, when the system or a key of the first keypad enters into the special character mode, the system preferably shows a corresponding second keypad of the invention, for at least said key or for all of the keys of the first keypad (e.g. by replacing said key, or all of the keys of the first keypad).

How to enter derivative characters

Methods of entering derivatives or related symbols of a character such as accented characters relating a non-accented character, or combined/compound characters relating to one character have been described before. An additional method for the same purpose is described hereafter:

According to one embodiment of the invention, after the user provides a gliding action corresponding to providing an unambiguous (i.e. a certain) character, for example either through the correction procedure or through a method of the entry of a precise character such as for example through providing a gliding action through the first keypad alone, (preferably, without removing his finger from the screen) and preferably when/after the system provides/proposes said precise character, the user may provide a (e.g. an additional) touching action (e.g. preferably, simultaneously) at another location on the screen/touch-sensitive surface.

According to one method, if said character has one or more derivative characters such as accented characters or combined/compound characters, then at the new touching point or on another location relating to the new touching point, a new (e.g. an additional) second keypad such as a telephone-type keypad may appear on the screen. Said keypad may include at least some, preferably all, of the accented and/or combined/compound characters corresponding to said provided character. At this time, based on the gliding principles such as those of the invention, the user may provide a corresponding gliding action from said new touching point towards/relating to the desired derivative symbol on said new (additional) second keypad. The system enters said derivative symbol.

According another method, after the user provides a gliding action corresponding to the original letter/special character on/from an appropriate key, at the end of the gliding action (e.g. after a pause without removing the finger from the screen, or after removing the finger from the screen), the system may show a second keypad of the invention including the derivative characters of the original letter/special character, respectively, either unde the user's finger or on another location on the screen. The user then may provide an appropriate interaction on said second keypad or anywhere on the screen relating to a key of said keypad as described before, to enter one of said derivative characters assigned to a key of the second keypad. As an example, as shown in fig. 320a, in order to enter the letter "c" which is a derivative of the letter "c", the user may first provide a corresponding gliding action on the key 320003 of the first keypad. The system may proposes/enter the letter "c". In this example, after a pause the user removes his finger from the screen and the system showes on the screen a second a telephone-type keypad 320011 (e.g. having a 3X3 matrix of keys) having derivative (e.g. accented) characters of said shown/proposed character "c" on its keys may appear on the screen. At this time, as shown in fig. 320b,- the user may provide the corresponding gliding action 320107 on or relating to a key to which his desired character is assigned on the new keypad and the system outputs/enters said derivative character (e.g. "9").

With continuous description of the current embodiment, based on the principle just described, other types of derivative characters such as " qu' " and " j' "as described in detail before, may also be entered similarly.

It must be noted, that according to one method, if the user does not provide any gliding action on the additional second keypad which is formed under the touching point or relating to the touching action, then the system does nothing.

The method just described, may be applied for entering any other tree-form data such as looking for media search, entering precise characters, entering commands into devices, etc. For example, at a first touching point on the screen, a media search system may show a different type of media on each key of the second keypad. The user may select his desired type of media such as "music" through a corresponding gliding action. Then with his other finger, the user may touch anywhere on the surface, and the system may provide a different derivative such as "band", "1970s" 1980s" "type of music", etc on each key telephone keypad, and so on. According to another example, it can also be used for entering commands or functions within menu bars, etc. For example, word processor menu bars may be assigned to telephone-type keypad keys. The user may select one of the bars by providing a corresponding gliding action. When the system selects said bar, then (preferably before removing his finger from the touch sensitive pad), the user may touch on a location on the screen and the system may open another preferably similar keypad with contents of the bar. The user selects one of them and so on.

Based on the same principles, also precise characters may be entered. For example, the letters of a language may be grouped in several forms and each of them may be assigned to a key of a (e.g. telephone-type) keypad. Through the same method, first the user may select one of the groups, and then one of the letters of the selected group. If said letter has derivatives, then at the next stage, the user may select one of them.

Repeating According to one embodiment of the invention, at the end of providing a gliding action for entering a symbol such as a letter, special character, or a function such as the movement of a cursor (e.g. caret) in a text, etc, if the user stops said gliding and holds his finger in contact with the screen without removing it from the screen surface, then the system may repeat entering said symbol until the user removes/lifts his finger from the surface. If for providing a symbol a first and a second interaction on two different locations (e.g. on two different keypads) on the screen is being provided, then, according to one method, the user may remove his finger from the first location on the screen and the system may continue repeating said symbol until the user removes also his other finger from the screen.

As an example, by considering the device 319000 of Fig. 319a and the command group

CMDl symbols represented by the second keypad 319055 in said mode instance of Fig. 319a, if the user presses (e.g. and holds) the key 319005 and then provides a gliding action rightward on the screen, to move the cursor (caret) to the right, then different methods for moving said cursor in said direction may be considered. According to one method:

- At the end of the gliding action, if the user removes his finger immediately from the screen, the system may move the cursor to the right over one character;

At the end of the gliding action if the user stops the gliding action without removing his finger from the screen, the system may move the cursor to the right over characters until the user removes his finger from the screen.

According to another method, as described previously, the length of the gliding action may define the number of characters over which the cursor will be moved in said direction.

According to one embodiment of the invention, after a user provides a letter that has an accented derivative, if he desires to provide an accented derivative and for example touches the touch sensitive surface such as the touch screen, the system may show the second keypad of the invention wherein its keys include at least some of the accents corresponding to said letter. According to one method, the second keypad of the invention may include the accents that may be common for all of the characters having accented derivatives. This method may simplify the graphical user interface (e.g. GUI) because a single set of accents assigned to the keys of the second keypad may generally be used for entering accented characters of different characters. As an example, fig. 321a shows a device 321000 having a touch screen 321009 and the first keypad of the invention having six keys 321001-321006. After the user provides a gliding action 321011 relating to the Certain/Stable character "e", then as shown in fig. 321b, the user may touch the screen 321009 with for example another finger and the system shows the keys of the second keypad 321108 of the invention to which at least most of the accents of letters in for example different languages are assigned. At this time, in order to enter the letter "e" including a desired accented letter, the user may provide the appropriate gliding action 321112 corresponding to the desired accent "e" so that the system provide the desired letter "e". In this example, providing the gliding action 321111 may result in entering the accented character "e". For sake of clarity, according to another example, if the user enters the letter "a" and then (e.g. while the cursor is immediately after said letter) touches the screen (e.g. outside the first keypad) then the system shows the second keypad with accents 321131 as described above. At this time if the user, for example, provides the gliding action 321111 then the system provides the accented letter "a".

In some languages an accented derivative of a character may include several accent symbols. For example, an accented letter of the letter "e" such as the letter "έ" may be presented by more than one accent symbols. In this case, according to one method, either said accents may be assigned to one of the keys 321131 of the second keypad, or for example at the end of providing the gliding action corresponding to the letter "e", the user may hold said finger on the screen and provide consecutive gliding actions 321111, 321112, corresponding to the plurality of accent symbols corresponding to the desired derivative " 6 " of the letter "e".

It must be noted that the arrangement of the accented letters or the accents on the keys of the second keypad may be organized such that it would to be easy to remember them. For example, the letters "e", "e", and "e", may respectively be assigned to the keys 321121, 321122, and 321123 so that they respectively correspond to the direction of their

actions for entering them.

The second keypad (model) of the invention may be used for entering "derivatives" of any symbol. For example, after the user provides the interaction for entering a dot character (i.e. "."), if the user touches the screen of the device in another location, then the system may show the keys of the second keypad of the invention to which several chain of characters that for example may begin with a dot character, are assigned. Then the user may provide the desired gliding action. As an example, fig. 322a shows a device 322000 having a touch screen 322009 and the first keypad of the invention. In this example, in a manner described before, providing a gliding action 322014 by a user may relate to entering a period/dot. At this time, (e.g. without removing his finger from the screen) if the user touches the screen on a location preferably outside the first keypad, then the system may show the second keypad of the invention 322008 wherein to at least some of its keys some "derivations" of the character dot are assigned. At tis time, as an example, providing the gliding action 322015 may provide the chain of character ".com", and according to another example, providing the gliding action 322018 provides the chain of character ".00", and so on.

Note that in fig. 322a, the gliding action 322011 corresponds to the "enter" function, and a gliding action towards lower-left side on/from the key 322005 corresponds to entering the character "," 322012 (e.g. comma character).

Word Completion

According to one embodiment of the invention, the system may also include a method of word completion. During the entry of a word, the system may, automatically or based on user's demand, show to the user the words having at least the length of the portion of the current word being entered until that moment. Said words may be shown using the second keypad of the invention and being entered according to principles of the invention. As an example, fig. 322b shows a device 322100 having a touch screen 322209 and the first keypad of the invention having six keys including four letter keys as are often shown and used in different figures of different embodiments. After pressing the keys 322002, 322001, 322003, 322003, and 322001 of the first keypad, the system may propose the word "those" 322110. At this time, the user may provide an interaction such as providing a gliding action 322111 which enters the system into word-completion mode and informs the system to show to the user some of the words/stems of the dictionary which their beginning characters have the same key pressing actions, and wherein said words/stems and that have the highest priorities among the corresponding words of the dictionary. Said words may be shown on the keys of the second keypad of the invention which preferably will be shown to the user after the user provides said word completion action/demand (e.g. said gliding action 322111). Then, if the desired word is on one of the keys of the second keypad, the user may provide the corresponding gliding action with for example, his other finger on the screen. As an example, if the desired word is the word "revolution" 322227, then the user may provide the gliding action 322112.

With continuous description of the current embodiment, according to a first method, the completing word having the highest priority (e.g., "absolute" 322229) may be proposed to the user based on the gliding action 322111 alone on a specific location before the second keypad appears. If said word is not the desired word, then the user may proceed to touching the screen with, for example, his other finger so that the system shows the second keypad with additional words.

In the example of fig. 322b, the words "above" 322226 and "above-mentioned" 322228 are part of the list of words which are shown on the keys of the second keypad. According to a second method, the completing word proposed to the user based on the gliding action 322111 alone, may be a longer word having the same beginning characters as the proposed word. For example, if the user provides a correction procedure for replacing the word those by the word "above" and provides the gliding action 322111 for entering the system into the word completion instance, then the system may propose the word "above-mentioned" based on said gliding action 322111 alone.

According to one embodiment of the invention, a word proposed through the word completion system may be a word that may have additional different completing stems. For example, in the example above, after the user selects the word revolution 322227 by providing a gliding action 322112, preferably he may touch with his other finger on another location, preferably outside the first keypad, on the screen. At this time, as shown in fig. 322c, the system may open another second keypad of the invention 322308 showing different completing stems corresponding to the word "revolution" The user may then provide the desired gliding action. In this example, the user provides the gliding action 322311 corresponding to the stem "ize". The system then provides the word "revolutionize".

According to one embodiment of the invention, if several proposed words completing portions have similar beginning character/s (e.g. similar portion) but their remaining chains of characters being different, then said beginning characters may be shown as one choice on a second keypad key. After entering said portion by, for example, providing a corresponding gliding action, if the user touches on another location on the screen, then the system may show on different keys of another second keypad said different remaining chains of characters of said several word completing portions.

According tone embodiment of the invention, after entering the system into the word- completion and touching the screen, the system may show the second keypad with selected words/stems. If the desired word/stem is not on the keys of the said second keypad, the user may provide another touching action, preferably on another location on the screen, and the system may show another second keypad of the invention including additional corresponding words/stems (e.g. having less frequently use). And so on.

According to one embodiment of the invention, during the entry of a word, the system may including a means such as a predefined gliding action providing on a key such as a space or backspace key so that, preferably in addition the current predicted word, the system may show, preferably on a first second keypad of the invention (e.g. having a three by three matrix of keys as described before), the words corresponding to the input information provided by the user until that moment and having the length corresponding to said input information. In order to select one of the words proposed on the first second keypad, the system may provide an appropriate gliding action for example anywhere on the screen. Additionally, the system may have a system of word completion wherein the words completion system proposes, preferably on a second second keypad of the invention, the words corresponding to the input information provided by the user until that moment and having a longer length than those proposed on the first second keypad. According to one method, any of said first second keypad and the second second keypad may be shown to the user automatically by the system during the entry of a desired word.

Chain of characters, for example, commonly used/included within various words of a language (e.g. a portion of a word) and that preferably begin with a same letter can also be entered based on the procedures using the second keypad of the invention as described above. For example, separately or in addition to the accented letters (if any) of a letter, chains of characters such as prefix, suffix, etc., beginning with said letter may be defined and being entered based on the procedures described above. As an example, when a user provides a "Certain'Vprecise character (e.g. through a corresponding swipe-and-holding action), the frequently used chain of character within words beginning with said letter may be shown on the keys of the second keypad of the invention which will appears on the screen. At this time, the user may enter one of said chain of characters if he desires. Along with said chain of character, the accented characters or other characters related to said precise character may also been shown on the keys of the second keypad.

Suffix/prefix

With continuous description of the current embodiment, fig. 322d shows as an example, a device 322400 having a touch screen and the first keypad of the invention. As an example, after the user provides a gliding-and-holding action 322411 corresponding to the precise character "t" (e.g. in this application, the non-precise character may also be called "Unstable" character and the precise character may also be called "Stable" character). At this time, if the user touches on the screen (preferably outside the first keypad) preferably with another finger, then the system may show several chain of characters relating-to or beginning-with said letter "t" on the keys of the second keypad of the invention 322408. The user then may provide a corresponding gliding action (e.g. preferably, with said another finger) to enter or add his desired chain of characters to the current proposed word/stem. In this example, after the user enters the stem/word "fic", he may provide said glide-and-holding action 322411 corresponding to the letter "t", and then provide the 322412 for entering the portion-of-a-word (e.g. suffix) "tion". In some embodiments in which suffix selection is supported, the processor is adapted to select for an entered key sequence, shorter words having more suffixes, rather than longer words which do not allow for many completion suffixes, even when the longer word has a higher ranking in the dictionary. Optionally, the shorter word is given an extra score and the shorter word is selected if the difference between the score of the shorter and longer word are smaller than the extra score.

Features to permit using the system with one hand

Some of the features of the system requiring interactions with two fingers may be designed such that to require one finger interaction only. For example, the "shift" gliding action may be provided first and then the user may remove his finger from the screen and the next character being entered will be capitalized.

Generally, at least one of the gliding actions departing from any of the letter keys of the first keypad of the invention may not correspond to any precise letter. For example as shown in fig. 322e, the gliding actions 322511-322514 respectively departing from the keys 322501- 322504 may not correspond to any precise character on their respective keys. According to one embodiment of the invention, providing any of said gliding actions may correspond to entering the system into the mode instance assigned to its corresponding key. After providing such gliding action:

According to a first method, the user may remove his finger from the screen and the system may show him the corresponding second keypad of the invention including the characters of the corresponding mode. The user then may provide a gliding action anywhere on the screen to provide his desired special character.

According to a second method, if the user removes his finger from the screen nothing happens, but if he does not remove his finger from the glide ending point on the screen at least for a while, then the system may show him the corresponding second keypad of the invention including the characters of the corresponding mode. The user then may either provide a gliding action from said point (e.g. without removing his finger), or from anywhere on the screen to provide his desired special character.

As an example, if the user provides the gliding action 322513 the system enters to the "123..." mode (e.g. "NUMB" mode). According to one method, at the end of the gliding action the system may show the second keypad wherein its keys include the digits 0-9. At this time, as shown as an example in fig. 322f, providing the gliding action 322509 may correspond to entering the digit '6".

1^st KP alphabetical order According to one embodiment of the invention, the letters on the keys of the first keypad may be arranged in alphabetical order. Preferably, the letters on the second keypad may be arranged such that a letter key of the first keypad and a letter key of the second keypad have at most one common character/letter. An example of such arrangement is shown in fig. 324 wherein on the touch screen of the device 324400 the letters of an alphabet are arranged in alphabetical order on the keys 324401-324404 of the first keypad. Accordingly, according to an exemplary method, the letters of said alphabet are also arranged on the second keypad 324408 to correspond to the principles described. In this embodiment, some modifications in the order of arrangement of letters on the keys of the first and/or the second keypad may be considered to for example augment the accuracy of the word prediction by the system.

Calibration

Different calibration procedures regarding defining the location of the dynamic first keypad of the invention have been described before. According to one embodiment of the invention before beginning to enter text on a location on the screen, the user may provide a predefined word such as the word "theo" (using a keypad model resembling to that of fig. 101A) such that to touch four different locations on the screen. The system then may assign to base of said locations one of the (corresponding) keys of the keypad. If the user desires to enter said word then he continues typing if not according to one method he may provide a backspace. The system then may erase the world entered but will consider the locating of the four keys.

Showing characters corresponding to a key press

According to one embodiment of the invention, when a user touches a key of the first keypad, mainly/only the symbols (e.g. letters, special characters, functions, etc.), of that key in for example, normal and mode instances (e.g. similar to symbols assigned to Shifted and non- Shifted instance of the key of PC QWERTY keyboard), may be shown on the corresponding keys of the second keypad of the invention. As an example, fig 327 shows a device 327000 having the first keypad of the invention including the keys 327001-327006 wherein four of said keys 327001-327004 include letters. In this example the name of the mode instances are printed on the top of each key of the first keypad. After the user touches the key 327003 of the first keypad, the second keypad of the invention 327008 (e.g. also called "reference keypad") may change to mainly show the letters and the mode special characters (e.g. in this example, digits) of said key 327003 of the first keypad on the keys of the second keypad 327008. As shown in this example, each key of the second keypad 327008 shows at most one corresponding letter and/or one corresponding special character which in this example is a digit. For example, the center key 327015 has the letter "k" and the digit "5". This way the user can see his desired character on the second keypad before for example providing a desired gliding action. At this time preferably:

On one hand, if the user desires to enter a letter, as described earlier, he can provide a gliding action by his finger which is on the key of the first keypad. According to one method, immediately after staring the gliding action, knowing that the user intends to enter a Certain letter, the system provides a new second keypad of the invention having only the letters (e.g. preferably, in larger font) of that key of the first keypad of the invention under the finger of the user such that the center key of the second keypad positions under the fingertip of the user. This way the user can easily provide the appropriate gliding action in the correct direction without or with a minimum of mental effort. In the current example, as shown in fig. 327a, when the user begins to move his finger on the key 327003 to for example provide an appropriate gliding action to enter a precise character, the system removes the reference/second keypad 327008 of fig. 327 and replaces it by the new reference keypad 327108 having mainly/only the appropriate letters preferably with larger font. The center key 327115 of the reference keypad is preferably located under the user's fingertip. Now, for example, a beginner user of the system can easily provide his desired gliding action on the reference keypad 327108.

- On the other hand, if the user desires to enter a special character, (e.g. while touching the key of the first keypad), preferably, with another finger, he may touch a location on the screen outside said key of the first keypad. At this time the system may show a new second/reference keypad of the invention, having only/mainly symbols such as special characters of said key of the first keypad in mode instance (e.g. preferably, in larger font), under the finger of the user such that the center key of the second keypad positions under the fingertip of the user. This way the user can easily provide the appropriate gliding action in the correct direction. In the current example, as shown in fig. 327b, while holding the key 327003, the user may touch a location 327217 of the screen and the system may redefine the second keypad of the invention 327208 such that to include only the digits, preferably with larger font, belonging to the mode instance interaction with the key 327003, and locates said second keypad under the finger of the user such that the center key 327215 locates under his fingertip. Now, for example, a beginner user of the system can easily provide his desired gliding action on the reference keypad.

In both cases described above, preferably, when the system shows the new second keypad, it removes the former second keypad from the screen.

In this embodiment, at the beginning the user can see his desired character on a second keypad which is far from his finger, therefore, he may know in advance the direction of the gliding action for his desired symbol. Then, by locating the second keypad under the user's finger, the system helps the user to provide his interaction (e.g. swipe) with more precision and with a minimum mental requirement.

The current embodiment may be applied for the entry of any other symbol such as functions. In the current example, as shown in fig. 327c, taping on the keys 327005 and 327006 may respectively provide Backspace function and Space character. As for the letter keys, to different types of interactions with the keys 327005 and 327006 different group of symbols such as standard and customized functions may respectively be assigned. As an example, if the user touches the key 327005, the reference keypad 327008 in a first (e.g. predefined) location 327215 such as the center of the screen may change to show the characters and functions that are assigned to said key in both normal and mode instance (e.g. similar to characters assigned to Shifted and non-Shifted instance of a key of PC QWERTY keyboard). As described for the letter keys, and shown in fig. 327d:

- on one hand, if the user glides his finger on said key 327005, the system may show under the user's finger 327415 a second keypad of the invention 327418 having the symbols of the key 327005 of the of the first keypad in normal (e.g. non-mode) instance. The user then may provide the desired gliding action as described for the letter keys.

- on the other had, if the user (e.g. preferably with a second finger) touches a location 327425 on the screen outside the key 327005, then as described for the letter keys, the system may enter into the mode instance of the key 327005 and show on a new second keypad of the invention 327428 mainly/only the symbols that are assigned to said key 327005 in the mode instance. The center of said keypad 327428 preferably will be located under the fingertip of the user's second finger. The user then may provide the desired gliding action (e.g. with the second finger).

Fig. 327e shows the symbols such as characters and functions that are assigned to different types of interactions with the space key 327006. They may be entered as described for entering symbols of other keys of the first keypad. In this example, when the user touches the space keys 327006, the system shows a modified second keypad 327508 arranged in similar manner as described for other keys, wherein each key said keypad shows at most one symbol in non-mode instance and/or one symbol in mode instance related to the key 327006. For example, the key 327516 of the reference keypad 327508 relates to "Tab" function 327518 if the key 327006 is interacted in a non-mode manner (e.g. the user provides a corresponding gliding action departing from the key 327006), and said key 327516 relates to the In- Word function 327515 if the key 327006 is interacted in a mode instance manner (e.g. the user provides a corresponding gliding action departing from outside of the key 327006).

According to one embodiment of the invention, instead of the gliding actions on the screen outside the keys of the first keypad, one virtual or hard key (e.g. an existing key such as the space key, or an additional key, may be assigned for the departure of the gliding actions relating the correction procedure. This way the system more easily may distinguish between the correction and mouse interactions on the screen.

According to one embodiment of the invention, a key or a predefined interaction such as a gliding action for example on or from a key may enter at least all of the letter keys of the first keypad into their mode instances. In this case, each gliding action departing from a (e.g. letter) key of the first keypad may correspond to its corresponding symbol in the corresponding mode of said key. According to one method, the next same predefined interaction may exit the at least letter keys from the mode instance.

Some (e.g. standard and/or customized) functions may be assigned to the gliding actions on/from the Space and Backspace keys are being printed in abbreviation. Hereafter the list of some of them and description of their function if needed. Some of them are auto explicative:

- Comp Word Completion described before

- ErWr Erase Word erases the word before the cursor in the text

- RpBk Repeat Backspace

- LkMds Lock Modes enters all of the letter keys into their mode instance

- UsrGd User Guide

- RpSp Repeat Space

- DelWd Delete Word Deletes word before the cursor from the dictionary As mentioned throughout this application, providing a gliding action departing from a (e.g. letter) key of the first keypad may correspond to selecting a precise character of a predefined group of characters (letters) assigned to said key, and press (and holding) said key and providing a gliding action (e.g. in all of the zone, or in a predefined zone) outside said (e.g. letter) key on the screen may correspond to entering a corresponding character of another predefined group of characters (e.g. special characters such as digits 0-9, "NUMB" group) assigned to said key, or vise versus. Accordingly, a switching means or an interaction may be used to change the order of the assignment. For example, if each of the letters assigned to a key of the first keypad is designated to be entered precisely by providing a corresponding gliding action alone on or departing from said key, and each of the digits 0-9 assigned to said key is designated to be entered precisely by pressing (and holding) said key and providing a corresponding gliding action from a location outside said key, then by using said switching means this order my be reversed (e.g. back and forth). In this case and example, by using said switch once, each of the digits 0-9 assigned to said key will be designated to be entered precisely by providing a corresponding gliding action alone, on or departing from said key, and each of the letters assigned to said key will be designated to be entered precisely by pressing (and holding) said key and providing a corresponding gliding action from a (e.g. any or a predefined) location outside said key.

Correction on Sp/BkSp key

According to one embodiment of the invention, the departing point of the gliding actions corresponding to the correction procedure of the invention is predefined to be provided on or from a predefined zone of a touch sensitive surface such as a (an on-screen) key (e.g. the Space key of the first keypad, its Backspace key, an added key, etc.). Said zone may also be a (e.g. all of the) zone between the split keys of the first keypad (e.g on the screen).

According to one embodiment of the invention, at any moment the user may use a word already entered for modifying it. Said modification may be in different ways:

- if the user locates the cursor/caret at the end of an already entered word and enters new characters, said characters will be added as a new separate word at the end of said already entered word.

- if the user locates the cursor/caret at the end of an already entered word and provides a predefined interaction such as a press/gliding action on/from a key such as the backspace key informing the system to use said word, and enters new characters, then the system considers the characters of said already entered word as Certain characters of the beginning of a word including said new characters and behaves according to the principles of the data entry system of the invention when the user adds said new characters to said Certain characters.

- if the user selects an already entered word and provides a predefined interaction such as a (e.g. the same) press/gliding action on/from a key such as the backspace key informing the system to use said word, and enters new characters, then the system considers the characters of said already entered word as Uncertain characters of the beginning of a word including said new characters and behaves according to the principles of the data entry system of the invention when the user adds said new characters to said Uncertain characters.

Simplified Interface

According to one embodiment of the invention, a simplified user interface based may be created for the data entry system of the invention. Fig. 331a shows as an example, a device 331000 having a touch screen 331009 and the first keypad of the invention similar to some of those described earlier having the letter keys 331001-331004, the backspace key 331005, and the space key 331006. In this example, there is no Reference keypad on the screen.

The letters of the alphabet distributed among each of the letter keys 331001-331004 of the first keypad of the invention may be printed on said key as was described earlier and shown through figs. 308-309, such that each letter occupies the same location relating to the other letters on said key, as it does with said letters on a standard telephone keypad.

As an example, after the user taps anywhere on the keys 331004, 331002, and 331003, the system may propose the word "was" 331010 having the highest priority among the words corresponding said tapping actions. If this is not the desired word, the user may proceed to the correction procedure of the invention by pressing preferably anywhere on the screen outside the keys of the first keypad. At this time, an (enlarged) copy 331008 of the key 331004 of the first keypad corresponding to the first Uncertain letter of the proposed word may locate under the user's finger such that its center key being under the user's fingertip. Said copy 331008 of the key 331004 may be enlarged enough such that the user be able to see at least some of the letters of that key around his finger. The user then slides/glides his finger toward the first character "d" of his desired word (e.g. in this example, the word "day") and the system may proposes the word "day" which in this example is the word with highest priority among the words corresponding the Certain character "d" just entered and the ambiguous tapping actions already provided on the keys 331002, 331003. After providing said gliding action, said copy 331008 of the key 331004 may disappear.

Accordingly, as described before, the user may also enter Certain characters/letters by providing gliding actions on or departing from a letter key of the first keypad. With continuous embodiment of the invention, as shown in fig. 331b, as an example, if the user presses (e.g. and holds) on a key such as the key 331004, the system may preferably enlarge said key such that the center of said key being under the user's fingertip. At this time the user may provide a desired gliding action in the direction of a desired letter on said key to enter (insert) it precisely (at the cursor position in the current word). As an example, the user may provide the gliding action 331107 to enter the letter "d" precisely. He may then tap on the keys 331002, 331003, and the system may predict the word "day".

For entering a letter in the center of a key or on the center of its copy, the user may provide a slight gliding action for example upward.

With continuous description of the current embodiment, according to one method, as shown in fig. 331c, when user presses (e.g. and holds) on a key such as the key 331003 (hidden) of the (first) keypad of the device 331000, (e.g. in addition to enlarged picture 331213 of said key) the system may show an (a small) icon 331208 on the screen showing a copy of a picture of the second keypad of the invention preferably including the symbols of said key 331003 in the corresponding mode instance (e.g. said second keypad may be a picture wherein said symbols are arranged according to the arrangement of the keys of a telephone keypad). Then (preferably by holding said key 331003 in pressing position), as shown in fig 33 Id, if the user touches preferably anywhere on the screen with for example a second finger, the system may open an enlarged (picture of the) second keypad 331408 of the invention (e.g. enlarging that icon) in the corresponding mode (in this example, the NUMBER mode) and locates it under the user's second finger such that its center key 331425 may preferably locate under the use's fingertip. The user then may enter a desired symbol of said enlarged second keypad for example by providing a corresponding gliding action (in this example, 331407) as described throughout this application.

According to one method, after the user enters a symbol in a mode instance, the system continues to show said (picture of) the second keypad in that mode until the user provides an interaction with one of the keys of the first keypad. Meanwhile, the system remains in said mode instance and providing gliding actions on the screen may correspond to entering additional symbols of said mode.

According to one method, the name or some of the symbols of each mode instance (e.g. 331433) may be printed on or next to each key of the first keypad.

If the symbols of a mode instance include additional symbols, the user may provide a predefined interaction such as a long gliding action towards the lower right location 331442 of the (picture of the) second keypad and the system will print the next symbols of said mode instance on said or on a new (picture of the) second keypad, and so on. Providing a gliding action towards the lower left location 331441 may show the (picture of the) second keypad with the previous symbols.

As described before, same types of interaction applied with/on the letter keys may be provided with the space and backspace key, or maybe other additional keys if any. In the current embodiment, when the user presses (and holds) one of said keys, said key may be enlarged such that the center of the enlarged key being under the user's finger tip. The user then may provide a desired gliding action towards his desired symbol such as a special character or a function on said key. Similarly, when the user presses (and holds) one of said keys, in addition to enlarging said key, the system may show a (picture of the) second keypad of the invention on the screen relating to symbols assigned to mode key of said key. The user then may press (e.g. and hold) anywhere on the screen with a second finger and the system may locate an enlarged (picture of the) second telephone keypad under the user's finger wherein its center is under the user's fingertip. The user then may provide the desired gliding action based on principles described throughout this application.

It must be noted that a gliding action towards a symbol on a key from its center may end on said key or outside it.

Different Input Signals for different interactions

Providing a gliding action from any point on the screen towards a first predefined direction preferably provides a first input signal, and providing a gliding action from any point on the screen in a second predefined direction preferably provides a second (different) input signal. As described in this and other patent applications filed by this inventor, said input signals may correspond to interacting with appropriate keys/zones on the screen, or of an imaginary keypad model.

According to one method, two gliding actions in a same direction provided differently such as with different gliding lengths, or with different level of pressure on the screen, etc.,, may correspond to two different input signals.

It must be noted, that during the correction procedure, an instance of the second keypad of the invention including the letters of its corresponding key of the first keypad may appear on the screen. According to one method, gliding action in the same direction in another instance of the second keypad of the invention corresponding to interaction with another key of the first keypad may correspond to another letter assigned to that gliding action. As described, an input signal of the first keypad and an input signal of the second keypad may have several letters assigned to each of them preferably such that they have at most one character in common.

Selecting a letter in Current Word manually (for correction)

According to one embodiment of the invention, a character of a word (e.g. of the current word) may manually be selected for being corrected. Preferably, the corresponding word of that character may be/become enlarged such that the user can easily select that character. Based on the correction procedure type, said letter may be replaced using the first and/or the second keypad keys. These procedures have been described before.

Complex words as a word (e.g. email address)

According to one embodiment of the invention, some of the words of the database may include at least some characters such as @ and dot. Said characters may be assigned to both the keys of the first and the second keypad of the invention. This way, email addresses and URLs may form a word of the dictionary of the system and may be entered through the word predictive system of the invention. Japanese

The data entry system of the invention can be applied to all of the languages based on letter alphabets or phonetic alphabets. It can be applied to Roman-based languages and the languages such as Korean, Chinese, Japanese, Hundu, etc.

As an example, a word in Korean language is composed of one or more Hanguls which are entered by entering Jamos. Usually, each Hangul is composed more that one Jamos. Jamos may be considered similar to the letters and be distributed on the keys of the first and the second keypad based on the principles of the system. Preferably, an end-of-the-word-signal may be provided at the end of the entry of a word.

According to one method, words in Japanese language may be entered based on their phonetic alphabets through an English/roman keyboard (e.g. Romaji). In this case, as for the

Pinyins in the Chinese language, the user can enter portions of a word until it ends entering the word. He may either provide

According to one method, during the entry of the word, the system may print on the screen the corresponding Roman characters he is entering and/or predicted by the system. The user may correct the predicted word/portion of the word, if desired.

If the corresponding Japanese word (or a portion of the word) proposed based on the

Roman characters is not the desired word, the user may select another word for example as described for the Chinese language.

According to one embodiment of the invention, each time the user presses a key of the first keypad, the key enlarges and shows an image of the reference keypad of the invention on that key so that the user can easier find his next desired leter and its corresponding key.

According to one method, the pressed key will remain enlarged until the user interacts with another key.

It must be noted that the examples of groups, the name, shape, etc, have been shown only to describe the embodiments. Other groups, shapes, ideas based thereon can be considered by people skilled in the art. For example, the items corresponding to the letters of a key may not be written. According to another method, the items may be printed bt pictures. For example, instead of writing the name of an animal, its picture may be entered on the key.

Portion of a word

According to one embodiment o the invention, a predefined interaction with a key (of the first keypad) may correspond to entering a (e.g. an ambiguous) portion-of-word/stem. Preferably, said interaction may correspond to any of both, a precise character of a word or a (an ambiguous) portion of a word preferably beginning with said precise character. The portion-of- a-words/stems and their assignment to the appropriate keys has been described in previous patent applications filed by this inventor. As an example, the portion-of-a-words/stems beginning with a same character may generally be assigned to the key to which said character is assigned.

According to one embodiment of the invention, a word can be entered by combining different types of interactions such as taping actions on and/or gliding actions on or departing from each of the keys of the first keypad. According to one method, the predictive system may relate the taping actions ambiguously to the characters on the corresponding keys, and may relate the gliding actions to either precise characters or (ambiguous) portions-of-a-word assigned to said keys. The system then may consider those interactions to the related letters and/or portions and may propose/predict one or more words to the user.

According to one method, preferably, part of a word including some of the characters such the (few) beginning characters of a word may be entered by taping on the corresponding keys, and the other portions of that word may be entered by providing gliding actions corresponding to the beginning character of each of said other portions. It is understood that the part of the word entered character by character by taping actions provided on the ambiguous keys may also be any other part of the word such as the middle or end part, and the other portions may be entered by the gliding actions.

As an example, by considering the letter keys 331001-331004 of the first keypad of fig. 331a, in order to enter the word "revolution", represented in the database as "re-vo-lu-tion", the user may first provide the ambiguous key presses corresponding to the beginning letters "revo", and then provide the gliding actions on or departing from the appropriate keys for entering precisely the characters "1" and "t", respectively the beginning characters of the potions "lu", and "tion". By considering ambiguous key presses corresponding to those letters followed by the precise letters "1" and "t", the system may on one hand search for a word that begins with those ambiguous characters and ends by said precise characters. The system on the other hand may also search for a word that begins with said ambiguous characters followed by two (ambiguous) portions that begin with said two precise characters. According to one embodiment, the system may propose the word having the highest priority among the words selected by the system in both cases. In this example, the word "revolution", may be proposed by the system.

In the embodiments above, the correction procedure of the system may correct the letters as usual according to the principles of the correction procedure, and the system may change the proposed word to another as described above. According to one method, proposed word may be based on all oi the user's interactions corresponding to entering a word including ambiguous key presses, precise characters, and precise or ambiguous portions entered by the user.

Location of letters based on shape characteristics

The embodiment just described, may be useful for entering long words with few key interactions but with very high accuracy. It can also help people that are not sure about the spelling of a word, and also reduces the output fluctuations while entering long words through word predictive systems.

Words in the dictionary may be arranged based on portions that are divided by a separator type such as a special character.

According to one method, a precise character and a portion of word beginning with that character may be assigned to two different input-signals/key-interactions.

If each of one or more of the keys of a keyboard has only one character assigned to a pressing action on it (e.g. such as a QWERTY keyboard), then, according to one embodiment of the invention, a pressing action on one such key may correspond to that character precisely and another type of interaction such as a gliding action on or departing from said key may (ambiguously) correspond to a portion of a word for example beginning with said letter as described above. If said key if a key of an on-screen (e.g. QWERTY) keyboard, then, depending on the accuracy of the location of the finger's impact of a pressing action on said key, the system may relate said pressing action precisely to said letter assigned to said pressing action on said key or it may relate said pressing action ambiguous to a group of characters including said letter and/or any of the letters assigned to pressing on the keys adjacent to said key and/or said impact.

According to one method, the word database may be arranged on a portion-based for each word. According to one method, said portions may be defined based on the syllables of the word, or according to any other predefined word dividing principles into portions known by people skilled in the art.

One of the factors of distribution and arrangement of the letters of the keys of the first keypad that may be considered is a method of distribution such that at least the beginner user can easily find his desired letter on a corresponding key. According to one embodiment of the invention, the letters may be grouped in a number of (e.g. four) different groups such that letters of each group generally/substantially have a resembling appearance characteristic of at least a portion of them (e.g. a portion of the letters of a corresponding group in their capital font). As an example, letters of a Roman alphabet (e.g. English) may be grouped in four groups and each of said groups being assigned to a different key of the four letters keys of the first keypad of the invention. Preferably, the letters are distributed on the keys of the first keypad based on the principles of the data entry system of the invention. For example, the letters on a key of first keypad and the letters on a key of the second keypad have preferably at most one common letter in common.

As an example, fig. 336 shows four letter keys 336001-336004 of a device 336000, wherein each of said keys ambiguously represents several letters of the alphabet based on the similar appearance of a portion of each of said letters as described. As an example, each of the letters "AHKNRX" which are assigned to the key 336001 have two legs (e.g. if they were objects to stand on a surface, they would have had two touching point on that surface). For example, letter 'K" has two legs 336011, 336012. In this example, a second group of letters is assigned to the key 336002. Each of the letters of this group has the characteristic in which it has substantially the form of a complete (e.g. 336025) or non-complete (e.g. 336023) circle. In this example, this group has an exception because the letter "W" 336026 which has 2 legs is assigned to the key 336002 because there was no place for that character in the previous group wherein the character "X" which also has two legs was assigned to it. By considering a telephone-type keypad as the second keypad of the invention, letters 'X" and "W" preferably should not be assigned to a same key as described before because they are on the same key of the telephone keypad.

With continuous description of the current embodiment, a third group of letters is assigned to the key 336003. Each of the letters of this group has the characteristic in which a horizontal line constitutes a portion of its shape. In this example, a fourth group of letters is assigned to the key 336004. Letters of this group have a common characteristic in which if they were objects to stand on a surface, they would have had other than two touching point on that surface. In addition they do not have the form of a circle and substantially non of them has a horizontal lined portion except the letter "F". In this example, all of those letters have one touching point, except the letter "M" which has three touching points. In this example, character "F" could have been assigned to the previous key 336003, but because the letter "E" was assigned to that key 336003, character "F" is assigned to the current key 336004.

It must be noted that each of some special characters used as a portion of a word may also be assigned to a key of the first keypad wherein its letters have a similar characteristic with that special character. For example, character "@" 336027 may be assigned to the key 336002, character "-" 336033 may be assigned to the key 336003, etc.

According to one method, each of the keys of the first keypad may have a shape to remind its group of letters. For example, the key 336001 itself may have two legs 336021, 336022, and the key 336002 may have a circle 336028 or itself being a circle, etc. The arrangement of letters according to this embodiment may result in slightly lower accuracy of prediction by the system when entering word only through ambiguous key presses, but it has clearly a faster learning curve for memorizing/finding a letter on a key therefore resulting in faster data entry system and adoption of the system by the beginners. Once the user gets used to the data entry system of the invention and sees its merits, he may then switch to another letter arrangement on the keys of the first keypad with higher accuracy as described and shown before.

According to one embodiment of the invention, combined different characteristics may also be used to find/memorize a letter assigned to a key. As an example, fig. 337 shows the four letter keys 337001-337004 of the first keypad wherein the letters are assigned to those keys based on different characteristics. For example, each of the letters of the group of letters assigned to the key 337001 have the characteristic in which a horizontal line constitutes a portion of its shape, except the letter "S". The characteristic of the letters assigned to the second key 337002 are in which that it forms two words "FARM", and "TX" (e.g. "Texas"). The key may have green color therefore user can easily remember those words/letters. In this example the third letter key 337003 includes all of the letters forming complete and non-completer circles. It also includes letters that at least a portion of them resembles to the letter "V" (e.g. "Y", "V", "K"). The fourth key 337004 includes other letters hat a portion of the is non- complete circle and straight line(s). It also includes the letters "D" and "J", remembering DJ. The key may have the shape of a disc. In this example, the user has more freedom to assign the letters to the keys such that the accuracy of the text entry based on ambiguous key presses being very high.

It must be noted that the characteristics such as similar appearance of portions of the letters, number of group of characters/letters, number of keys to which said groups are assigned, and the methods of assignments to the keys may be considered differently be people skilled in the art, and the examples shown and described in the embodiments above are used for description of the principles only.

According to one embodiment of the invention, instead of assigning the group of letters to the keys as described above, the user may provide four different finger gestures such as

air. As such the word predictive system may relate each of said gestures to its corresponding group of characters as shown and described before and predicts a corresponding word. For example, by considering the first keypad of the invention of fig. 336, providing gestures may provide the word "this".

Fig. 338a shows a slightly different arrangement of the letters of the alphabet on the keys of the of the (first) keypad based of on principles described before compared to their arrangement on the keys of the (first) keypad of fig. 336. In this example, letter "W" 338011 is moved to the key 338001, and the letter "F" 338012 is moved to the key 338003. Now, all of the letters of the key 338001 have two touching point on an imaginary surface if they were objects standing on said surface (e.g. each of them has two legs), all of the letters of the key 338002 have the shape of a complete or non complete circle, all of the letters of the key 338003 have a horizontal-lined portion, and each of all of the letters of the key 338004 has a number of touching point on the imaginary surface which is different than two touching points (e.g. letter "M" has three touching points and all others have one touching point). In this example, if the second keypad of the invention is a telephone-type keypad then the letters W and X, which are on the same key of the telephone keypad and also on the same key of 338001 of the first keypad may be ambiguous when the user provides a corresponding gliding action in order to enter a precise letter during the Correction procedure or during the Insertion procedure. In this case, according to one method, the user may shorter or longer gliding actions in the same direction for entering respectively one of said letters.

According to another method, the second keypad may be considered as being any type of known or customized keypad. Fig. 338c shows as an example, the keys of the Fig. 338a in another key arrangement such that when the user enters a text in for example the English language he may type on different keys preferably evenly. In this example, the key 338004 is identified by a dot 338119 which is based on its main principle which is if the letters on said key were objects that are standing on a virtual surface, substantially all of them except the letter "M" may have a single contact point with said surface.

The second keypad of the invention may be of any type such as a known keypad or a customized keypad. As an example, the second keypad may be a slightly modified telephone- type keypad with slight exceptions as was shown in figs. 338a and is shown in fig. 338bc (e.g. said keypad itself is not shown here, but the arrangement of its letters as printed on the keys of the first keypad reminds the assignments of letters to the keys of that keypad), wherein its key to which the letters "TUV" are assigned may also include the letter "W", and its key to which the letters "MNO" are assigned may also include the letter "F". In this example, the lower-center position of the keys of the first keypad include the letters "TUVW", and the center-right position of the keys of the first keypad include the letters "MNOF" .

With continuous description of the current embodiment, during the correction procedure of the invention, in order to enter precisely the letters "W" 338115 or "F" 338118, the user may provide the corresponding gliding action (e.g. beginning outside the keys of the first keypad), downward or rightward, respectively. Providing gliding actions in the same directions respectively, on or departing from the corresponding keys of the first keypad may Insert said letters, respectively. The location of the letters as described (e.g. W, X) on the keys of the second keypad may be changed/swapped according to parameters such as the language. For example, in English language, the locations of the letters "W" 338115 and "X" 338116 may be swapped, and when using the French language the locations of said letters may be maintained as shown in fig. 338a or 338c. This is because for example, the letter "X" is rarely used in English language, so it is corrected less. Same may apply to the letter "W" in French language.

According to one embodiment of the invention, instead of using gliding actions during the Correction procedure or during the Insertion procedure for entering a precise character such as a letter, the user may handwrite the letter on the screen outside the keys or on the keys, respectively.

As an example, by considering the fig. 338a, by pressing on the keys 338002, 338001, 338001, the system may propose the word "far". The user then may proceed to the Correction procedure of the invention using handwriting for entering a precise character and write the letter "e" 338013 on the screen outside the keys. The system may consider said letter as being the first letter of the desired word and may propose another word, "ear" in this example.

In the example of the fig. 338b, for entering the same word, through the handwriting method, the user may write the letter "e" preferably on the key to which said letter is assigned. This way, the user may enter the letter "e" precisely. He then may press on the keys 338001, 338001, ambiguous corresponding to the remaining characters of the desired word. Based on these combined interactions, the system may propose the word "ear" without the Correction procedure.

It must be noted that entering precise letters during Correction and/or Insertion procedures may be provided on (other) different predefined sensitive zones (e.g. on said screen or on separate touch sensitive surface(s) than what described above.

According to one embodiment of the invention, out-of-dictionary words may be entered precise letter by precise letter through the handwriting system. the first keypad may be the space character. As such a word of the dictionary may also include a space (e.g. " ") character (e.g. such word is a compound/combined word, including two or more single words). A space character within a compound word of the dictionary may be represented by a special character. In this method, a word of the dictionary may also be defined to be formed from several single words. This is beneficial to reduce the number of corrections. For example, by considering the Fig. 338c, if the ambiguous letter " " 338117 is assigned to the key 338001, then by considering the letter layout of fig. 338c, normally, if the user desires to enter the words "I am" by pressing the space key 338006 between said words during entering them, the system may propose the words "I as", because the word "as" has higher priority than the word "am". In order to avoid the correction procedure, the user may enter the words "I am" as a single combined word by using the ambiguous space character 338117 between said words by pressing the key 338001 instead of the key 338006 between said words. For that, in this example, the user may press the keys 338004, 338001, 338001, 338004. In this example, the chances are very high that the combined longer word "I am" 338114 may be the one which has the highest priority among the (e.g. combined or single) words having the same key presses in the dictionary used by the system and may be proposed by the system without Correction. The system may propose said combined word as a single word without the need of correction. Obviously, if a correction is needed, at least during the entry of the combined word the system preferably considers said combined word as a single word and may act accordingly.

In the current embodiment, the system may not include said combined words, but every time the user enters an ambiguous space character " " 338117, the system may create a corresponding combined word on the fly when the user enters the remaining keypresses of the desired combined word.

With continuous description of the current embodiment, in order to not confuse the user, according to one method, the ambiguous space character may be shown to the user differently from the space character (which is generally not seen), for example, by a color or gray shading, by a unique shape and/or by another character infrequently used, such as "~". This display of the ambiguous space character may be presented on the key or otherwise in the key description for the user on the first and/or the second keypad and/or may be displayed on the screen when the ambiguous space character is entered, at least during the entry of a combined word. Even after the confirmation of the system's proposed word by the user, said connecting character (e.g. "~") may remain between the words of the combined word in the document. As such, according to one method, if the user desires to re-use (i.e., re-edit) said combined word, the system may use all of the words of the words of the combined word for said er-editing procedure. Eventually, for example on completion of writing a document, the connecting space characters are converted into regular space characters, automatically or by a user command.

When the user knows that a phrase including two or more words is in the dictionary or will be used often and should be entered into the dictionary, the user may use the connecting space character instead of the regular space, to indicate that the words of the phrase should be handled in the accessing the dictionary as a combined word.

According to one method, during the entry of a combined word, the user may enter the whole and/or a portion of at least some of the words of said combined word preferably each word/portion being separated by said connecting character. For example, the user may press/glide on the keys corresponding to one or more beginning characters of each of the words of the combined word while entering the ambiguous or precise connecting character (e.g. "~") between the pressing actions corresponding to each of the words of the combined word.

By considering the combined input information, and knowing that after each connecting character or the ambiguous pressing action corresponding to the connecting character, the system may predict a desired combined word.

Alternatively to the connecting space character representing only a space character, the connecting space character may represent several characters which the system is to guess based on a matching word or phrase including a plurality of words in the dictionary. Optionally, the connecting space character in this embodiment is interpreted as representing a completion of the current word and the space following the word.

As an example, in the combined word is the phrase "have~a~good~day" which is a combined word of the dictionary, then as an example, by considering the first keypad of figure 336, the user may press the keys 336001 (ambiguously for Ή"), 336001 (ambiguously for "A"), 336001 (ambiguously for "~"), 336001 (ambiguously for "A"), 336001 (ambiguously for "~"), 336002 (ambiguously for "G"), 336001 (ambiguously for "~"), 336004 (ambiguously for "D"). By considering said input information, and considering that each of the last ambiguous key press 336001 between different key presses may correspond to the connecting character, and that each sequence of key presses between said (ambiguous) connecting characters may correspond to at least the beginning characters of a word of a desired combined word, the system may be able to propose the desired combined word.

It is noted that, preferably, the connecting character may be entered as a an ambiguous character which the system selects from one of a plurality of characters of an actuated key or as a precise character by for example, providing the desired gliding action on/from its corresponding (ambiguous) key.

According to one method, the connecting character may be assigned to a key other that said ambiguous letter keys of the first keypad so that to be more easy recognizable by the system.

Preferably, according to one method, the combined words of the dictionary appear in the dictionary with said connecting character (e.g. "~").

According to one method, if the second keypad of the invention is a telephone-type keypad, in addition to being assigned to a key of the first keypad, then the connecting character may preferably also be assigned to the key of the second keypad to which the digit "1" is generally assigned. In this case, according to this method, in order to insert said connecting character or correct an ambiguously proposed character relating to pressing the corresponding key of the first keypad to said connecting character, the user may provide a corresponding gliding action towards upper-right on the screen, respectively, from the corresponding key of the first keypad or from outside the keys of the first keypad.

According to one method, because the combined word may be considered as a single word, the system may use the word completion system to propose a word before the user provides the input (as described above) corresponding to all of the words of the combined word.

According to one embodiment, instead of or in addition to the word database of the system, the system may use word databases of other resources such as those available in its corresponding device or on the Internet.

If the combined word entered by the user is not in the dictionary, the system may add it to the dictionary.

According to one embodiment of the invention, each time the user enters the input information of a current (single) word, the system considers/combines said input information with (input information) of one or more previous words and provides/creates a combined word based on the combined words of the dictionary.

According to one embodiment, in addition to single words, a 1-gram word database used by the system may include compound words (e.g. bi-grams, tri-grams, etc.) preferably connected to each other by a connector.

According to one embodiment of the invention, the gliding actions for entering precise character during Correction or by Insertion may be provided in predefined directions other than based on a telephone keypad arrangement. For example, the gliding action may be in direction of the ending portion of a letter when it is written by hand. As an example, fig. 338d shows the letter key 338001 being shown in enlarged position for describing this embodiment. The ending portion of the letters on the key 338001are shown by arrows in different directions. For example, letters "A, X" end by two gliding actions in different directions 338201 and 338202, respectively. During the Correction or Insertion procedures said letters may be entered precisely by providing gliding actions 338211, and 338212, according to said ending portions, respectively.

In the current embodiment, there may be some exceptions to consider. For example, the letters "K" and "R" may end in the same direction when writing them. In this case, for example, the gliding action corresponding to the letter "R" may be considered to be in a different direction 338203 relating to other letters on sad key, or as another example, it may be considered to include a curved trajectory 338213 to remind a portion of the letter "R". Other letters on each of the other keys may follow similar principles based on how described and as may be known by people skilled in the art.

According to one embodiment of the invention, during the correction and/or insertion procedures, the user may combine gliding actions as described above and handwriting method for entering different characters of a key precisely. For example, as shown on the key 338002 of fig. 338c, the letters "L" and "Z" may end in the same gliding direction. In this case, as an example, a gliding action rightward may correspond to entering the letter "Z", and the user may provide a compound gliding action " |^ " that resembles to handwriting the letter "L".

If the letter keys of the 2^nd keypad (imaginary/model) have key and letter arrangement resembling to that of a telephone-type keypad, then, according to one embodiment of the invention, the letters J,K,L, usually belonging to the center key of said (e.g. telephone-type) keypad, may be assigned to the (upper-left) key of said keypad wherein usually no letters are assigned. Accordingly, each of said letters J, K,L, may be the printed on/assigned to the upper- left side of its corresponding key of the first keypad.

As an example, fig. 339, shows the letter keys 339001-339004 of the first keypad. In this example, the second keypad of the invention is an imaginary telephone-type keypad model with the difference that here the letters J, K, L, are assigned to the upper-left key of the 2^nd keypad (model) and are printed on the upper-left side of their corresponding respective keys 339001, 339002, 339004, of the first keypad. Note that, in this example, the special characters , 339121

339123 and— 339124, are assigned to the center key of the 2^nd (imaginary) keypad

(model) and are printed respectively on the center of the corresponding keys 339001, 339003, 339004, of the first keypad. In this example, the special character Π 339122, represents the space character when it is part of a (combined) word. A combined word has been described in detail the previous patent applications filed by this inventor. It is generally constituted of more than one word separated/connected with a special character such as a space character, and is considered by the system as a single word. It is generally included as a single word within the dictionary used by the system. Said space character is generally considered by the system as a character/letter of a combined word and may be ambiguously assigned to one of the keys of the first keypad, and also be assigned to one of the keys of the second keypad, and be used with the (predictive) data entry system of the invention.

According to one embodiment of the invention, in some languages using ideographic characters such as Chinese or Japanese, if a word is not in the dictionary, the user can add it to the dictionary. After the user enters at least some of the transliteration symbols (e.g. pinyin, romaji, Bopomofo etc.) corresponding to a desired word, if the system does not include that word, the system may be informed by the user that he wants to add said word to the dictionary. In this case, because in such language the transliteration symbols are usually entered according to one method, the system may consider the input information provided by the user and may propose different sets of transliteration symbols based on said input information divided differently to provide each of said sets of transliteration symbols .

Each of the proposed different sets of transliteration symbol s may be presented on a key of the second keypad of the invention, and the user may provide a corresponding tap/gliding action to select the desired set of transliteration symbols . After selecting the set of

transliteration symbols , the system may present to the user the corresponding (ideogram) characters of each of the transliteration symbols of the set preferably, one by one, and the user selects the corresponding desired (ideogram) character for each of the transliteration symbols .

With the continuous description of the embodiment, according to one method, if the set of the transliteration symbols does not include all of the transliteration symbols of the desired word, the user may then provide the input information corresponding to the following transliteration symbol(s) , and the system may propose the corresponding (ideogram) characters, and the user selects his desired (ideogram) character. This procedure may be repeated until the user enters all of the transliteration symbols and the desired (ideogram) characters of the desired word. At the end, the user may provide an end-of-the-word signal such as a space character or a special character.

The assignment of the letters/characters to the keys of the first keypad based on their common characteristics such as a portion of their shapes may be applied to the characters of any language. Fig. 340 shows as an example the assignment of the Korean letters to four keys of the first keys pad accordingly. As an example, key 340001 mostly includes the letters that stand on one point (as referred to as "leg"), key 340002 mostly includes the letters that stand on two points/legs, key 340004 mostly includes the letters that stand on a horizontal line, and key 340003 mostly includes the letters that have a closed zone within them. For reasons such as for example to augment the accuracy of the prediction, there may be some exceptions in assigning the letters to the keys. For example, the letter "T" which stands on one point/leg generally belonging to the key 340001, is exceptionally assigned to the key 340003 in this example.

It is understood that the examples of distribution of letters on the keys of a (first) keypad based on the common shapes as provided above are only exemplary. Other kinds of distribution and/or other common portions of letters/characters maybe considered and accordingly assigned to any predefined number of keys by people skilled in the art.

Integrating the interface of the data entry system of the invention within a steering wheel of for example a car has been described before. According to one embodiment of the invention, said interface may be a standalone attachable/detachable interface that can be attached to or detached from the steering wheel.

According to one embodiment of the invention, the letters on the keys of the keypad may be distributed based on their common characteristics and the letters on the keys of the keypad may be distributed based on a QWERTY order. Fig. 341 shows as an example, the letter keys 341001-341004 of the first keypad distributed based on common shapes as described earlier. In this example, the letters on the keys of the imaginary second keypad model 341010 of the invention are distributed based on a QWERTY keyboard 341000 letter layout. In this example, the letter keys of each row of the QWERTY keyboard are divided into three groups, wherein each group is assigned to the corresponding key of the corresponding row of the imaginary second keypad model 341010. The order of the letters on the keys of the first row of the second keypad corresponds to the order of the letters on the first row of the QWERTY keyboard. For example, letters "QWE" forming the upper left group 341011 on the QWERTY keyboard 341000 are assigned to the upper left key 341021 of the second keypad 341010. Accordingly, the order of the letters on the keys of the second row of the second keypad corresponds to the order of the letters on the second row of the QWERTY keyboard, and the order of the letters on the keys of the third row of the second keypad corresponds to the order of the letters on the third row of the QWERTY keyboard. It must be noted that the examples order of distribution of letters on the keys of a PC keyboard such as the QWERTY to the keys as described above are only exemplary. Other kinds of distribution of letters/characters of said keyboard to the keys of the second keypad maybe considered by people skilled in the art. For example, the first group may include the letters "QWER".

As an example, and according to another embodiment of the invention, as shown in fig. 343, another type of distribution of the letters (e.g. of Roman/English alphabet) on the keys of the first keypad based on their common (partial) shape characteristics as described before, and another method of dividing the letters on a PC keyboard may be considered. In this example, letters of the alphabet are divided into four groups based on their common shapes

(characteristics) wherein each group is assigned to a different key of the first keypad of the invention, (e.g. one of the groups such as) The first group generally includes the letters that have a substantially closed substantially circle shape (e.g. they have a substantially closed circle in their shape). In this example, they are assigned to the key 343004 of the first keypad, (e.g. one of the groups such as) The second group generally includes the letters that if they were objects standing on a plane/flat surface they would have two touching points with said surface (e.g. or in other words they have two legs). In this example, they are assigned to the key 343002 of the first keypad, (e.g. one of the groups such as) The third group generally includes the letters that if they were objects standing on a plane surface they would have one touching point with said surface. In this example, they are assigned to the key 343001 of the first keypad, (e.g. one of the groups such as) The forth group generally includes the letters that have a straight and/or a curved lined base. In this example, they are assigned to the key 343003 of the first keypad. Fig. 343A shows the arrangement of the letters on the letter keys 343101-343104 of the first keypad as just described, and the common portions (e.g. shape) characteristics are shown on/under each letter on its corresponding key.

According to one method, the distribution of letters among groups may take in account that a common shape characteristic of one of the groups of letters may have priority over the common characteristic(s) of at least some of the other groups. For example, in this example, the first group may have priority over the other groups. In consequence, the letter "R" (e.g. having a closed circle, and standing on two legs) which could have been included in both, the first and the second group, is assigned to the first group which has priority over the other groups (i.e. in this example the "closed circle" group has a priority over the "two legs" group). In this example, the first keypad also includes a space key 343006 and a backspace key 343005 as shown in fig. 343A. In example of fig. 343, the letters of the keys of the QWERTY keyboard 343000 are divided into nine groups and assigned to the second keypad model 343010 of the invention having a three by three matrix form factor.

The letters of each group assigned to a key of the first keypad may be printed on the screen on a (e.g. an imaginary) three by three matrix based form factor such that the letters printed in a/the same position on each of the keys of the first keypad are the same group of letters which are positioned in a/the same position in the three by three matrix based second keypad of the invention. For example, letters "FGH" which are printed on the center of, respectively, the keys 343001, 343003, 343002, are also assigned to the center key 343025 of the (imaginary) second keypad model 343010 of the invention. Therefore, the user which is familiar with the PC (e.g. QWERTY) keyboard may provide blindly interactions such as tapping/gliding actions for example for entering precise characters during Correction and/or Insertion procedures of the invention.

According to another method, one of the groups may include substantially all of the letters having closed portion/form (of any shape) in their shape.

Preferably, a key of the first keypad and a/the key of a second keypad may have at most one common letter/character. As such, in the current example, some exceptions are being considered. For example, the letter "O" 343018 assigned to the key 343004 of the first keypad, which generally (for example, according to the QWERTY keyboard layout) should have been assigned to the upper right key of the second keypad 343010, is here assigned to the middle- right key 343026 of the second keypad 343010. Similarly, the letter "P" which is assigned to the key 343004 of the first keypad is also assigned to the upper right key of the second keypad. According to the principles of the invention described before, they preferably may not be assigned to the same key of the second keypad, and therefore the letter "O" is here assigned to the key 343026 of the second keypad 343010. In this example, the same principles are applied to the letters "T" 343017 which is assigned to the key 343003 of the first keypad, wherein normally also should have been assigned to the upper middle key of the sepond keypad 343010, but for the same reason it is here assigned to the upper-left key 343021 of the second keypad 343010.

The data entry system of the invention frees the screen from the input interface and the user may have the full screen available to see the output. Many applications may be created using the data entry system and/or principle of the invention (e.g. in the advanced/invisible mode wherein no keys appears on the screen). As an example, a chat/texting program may be created wherein two or more users may communicate with each other through their respective (e.g. mobile/fixed) devices and simultaneously see the image (e.g. picture, video/video stream, data stream, etc.) of each other or other selected images such as a TV program on their screens. By using the data entry system of the invention wherein preferably no keys are shown on the screen, the user may type on the screen while seeing the other person's images on the screen and simultaneously typing. The other person may be the person with whom he/she is chatting and may also see the text that said other person types. Preferably, in case said two persons are natives of two different languages, a translation program may be used to simultaneously translate the user's text being entered into the language of the receiving person.

According to one embodiment of the invention, two or more parties can (e.g.

simultaneously) videochat between themselves. A user interface may be created for such a purpose on the screen of each party, wherein while a first party/person videochats with a second party/person, he can see the second person's text sent to said first person along with a picture/video of said second person or of another image/video filmed and being send during chat by said second person to said first person, and vise versa. According to one method, the chat and the corresponding picture/video may be sent to all of the parties, preferably simultaneously. According to one method, the history of the chat may be shown on the screen.

Fig. 344 shows, as an example, the screen 344000 of a device during a videochat using the data entry system of the invention. The system includes the first and second keypad of the invention. In this example, the first keypad of the invention has four letter keys 344001-344004, a space key 344006, and a backspace key 344005. They may be visible, or they may be invisible (dynamic) keys as described before such that they do not occupy the screen. When the first party types a word, the current word 344009 and/or the text 344011 may be shown on said first user's screen (and eventually, in real time on the screen of the second party). The picture and/or the video of the second party with which the first party is videochatting or a picture or video being selected and/or provided/played 344013 by said second party may preferably simultaneously be shown on all of or at least a portion of the screen. A text/chat zone 344014 may be dedicated on the screen 344000 wherein the text of one or preferably all of the parties may be displayed on it. The screen may also include a (small) zone 344010 showing the small pictures/videos of, or the pictures/videos provided, by each of the chatting parties, preferably, the current chatting parties. The pictures/videos may be displayed on at least some of an at least three-by-three matrix zones of static or dynamic images. According to one method, when a first party touches a predefined zone on the screen covering at least a portion of the screen, said first party may provide a gliding/tapping action 344008 resembling that of the correction procedure as described, to select a corresponding picture 344007 identifying a second party with whom to videochat/text. The system then enters into a videochat instance (i.e. state) between the two parties, and preferably displays the enlarged picture/video provided by the second party on the screen of the first party and vice versa, and preferably, replacing that of a previous party. Preferably, when a party touches said predefined zone on the screen, the system provides said small zone 344010 and its corresponding videos/pictures under the party's finger as was described for the second keypad of the invention during the correction procedure.

It must be noted, that a picture/video of the second party may be a live/non-live video of the second party, filmed by said second party while chatting/texting, or it may be some other contents that may be sourced from other sources such as a TV program seen preferably at the same time by parties, etc. Said contents may be such as graphical videos/pictures, or other images. As an example, the TV program may be a sport match wherein parties watching the match may text-chat between each other about the match and about other subjects.

According to one method, the zone used to provide tapping/gliding actions on the screen and/or corresponding to interacting with the text and/or with a keypad of the invention, and the zone to provide interactions such as a gesture for selecting a party may be different zones on the screen..

Note that any of those fields and zones described above, may be placed anywhere on the screen and may be relocated based on the parties' manipulation.

According to one method, a party may simultaneously videochat with more than one other party, and may configure the screen in order to display enlarged pictures/videos of more than one other party on the screen.

According to one embodiment of the invention, interacting (e.g. preferably

simultaneously) with two predefined vertical zones each covering substantially all of a surface of a different side of an opposite side of the screen of a device in portrait or in landscape mode may be related to relocating/recalibrating the keys of the first keypad of the invention under the interaction location. In addition in order to access to the content under those two zones, a switching means such as predefined interaction for example, with a key of the first keypad may be provided to remove those zones and next time to bring them back on the screen.

According to another method, providing a press-and-holding action (e.g. for at least a predefined of time) with each of (e.g. at least) a pair of the user's finger (e.g. with two thumbs), preferably simultaneously, anywhere on a surface such as a touch screen may locate/relocate each of the (two) portions of the (e.g. split) first keypad of the system (e.g. such as the portions of the split keypad of fig. 101A) under a corresponding finger at the corresponding press-and- holding action (e.g. under the user's fingers), respectively. Preferably, the border between the two letter keys of each portion (e.g. or alternatively a specific key of each portion) may be under the appropriate touching point. According to one method, in order to eliminate the ambiguity between a mouse function and the calibration procedure, the system may include a software object in form of a virtual grid covering preferably substantially the entire said surface (e.g. of the touch screen). Said grid (which may be of a physical/virtual interface form) may be an invisible grid for example comprised of narrow continued or discontinued very thin (e.g.

crossing) lines or very small spots (e.g. one to few pixels in size, or made of one to few pixels, respectively) distanced far enough from each other such that to maximize the visibility of the output/content displayed on the screen but close enough such that when a user touches (e.g. anywhere on) the surface with a finger, the finger touches both the screen under the grid and the grid. By using such grid, during calibration, the data entry system may detects the user's finger(s) (e.g. thumbs) touching points/locations on the screen, along with the operating system of the device. According to one method, the data entry system may inform the system to ignore the mouse function and may relate said interaction to the calibration procedure, while according to another method, both the operating system and the data entry system operate in response the user's interaction (e.g. the mouse interaction event and the calibration of the keypad). Detecting user's interaction location on the screen using such a grid may be used for other purposes too. According to one method, such grid may be designed in such a way that detection of the user's interaction may occur without the user touching said surface.

The calibration/recalibration procedure just described, may enable a user to interact with the system (e.g. through a touch screen surface) without looking for the input interface. At any moment, the user may define where the interface should be/is.

By using this method of calibration, in other word, the user's fingers (e.g. thumbs) become the input interface. When a user provides a calibration procedure, the impacts of his fingers up and down, and/or left and right relating the touching ppints (e.g. in a predefined radius/zone around the calibration touching points), may relate to providing the input information corresponding the group of characters assigned to said impacts.

According to one embodiment of the invention, each of the (e.g. four) divided groups of characters (e.g. letters based on their common shapes characteristics and/or special characters based on categories described throughout this application) may be assigned to a different finger (e.g. all of then to four fingers) and be provided by interacting with an input surface/interface. The relationship between the locations of impacts of the user's fingers on the surface may defined the input signal and its corresponding related characters corresponding to the user's finger interacted with the screen. As an example, two of said groups may be assigned to two fingers of each hand of a user, each to a different finger. Space and backspace functions may be assigned to a third finger of each hand.

According to one embodiment of the invention, instead of or in addition to using a key to provide an input signal corresponding to a group of characters, the user may draw a shape of/representing said group for example (e.g. anywhere) on a touch sensitive surface such as a touch screen. Said groups of letters may be of any number of groups such as four groups respectively being based on kind of different predefined number of common characteristics such as a common shape characteristic, sound of the letters, etc., as described before, and may be used by a word predictive system for example that of the data entry system of the invention to predict a word.

Fig. 342A, shows as an example, a device having a touch screen 342000. In this example, each of the shapes 342011-342014 respectively represents one of the groups of ambiguous characters 342001-342004 used by the predictive data entry system. The characters of each group have generally a common shape characteristic such as those described before. By considering a dictionary of words/database (i.e. a word (e.g. N-gram) database, wherein preferably, N>0)„ and word predictive system, if the user draws the shapes, 342021, 342022, 342023, 342024, respectively corresponding to the groups of letters 342003, 342002, 342001, 342001, the system may propose the word "this" which has the highest priority (among the corresponding entries in the database) among the words corresponding to said drawing actions. Note that in this example, the user provides each consecutive common shape with a different finger (e.g. different thumb) each on a different side of the screen, thus permitting fast text entry.

The model of an imaginary second keypad model of the invention as described throughout this application may be used for correction procedure, special characters, and functions as before.

According to one method, the drawing actions for entering ambiguous letters may preferably be provided on the sides of the touch screen surface, and gliding actions

corresponding to correction procedure may be provided further away inside the screen.

The Insertion procedure of the invention may also be provided using the drawing method. According to one embodiment of the invention, after the user provides/draws one of the shapes with his finger, and by respecting providing a predefined principle such as keeping that finger in fixed position for a predefined laps of time, the system may show the letters relating to that shape on the (e.g. imaginary) second keypad of the invention on the screen, and the user may provide the corresponding gliding action for entering a desired character/letter. According to one method, mode instances and their corresponding groups of symbols may be assigned to the common characteristics such as shapes. According to one method, after drawing one of the common shapes by a first finger, and based on predefined principles such as for example, not lifting said finger from the screen for a predefined laps of time and/or touching the screen with his second finger, etc., the user may enter the system into the corresponding mode instance assigned to said common shape. After showing the characters/functions corresponding to that mode on the second keypad of the invention preferably under the user's second finger touching the screen, the user may select one of said symbols by providing a corresponding gliding action. In fig. 324B, which is shown as an example, the punctuations mark characters are assigned to the common shape "V". After drawing said shape 342101 with a first finger (e.g. and holding said finger in contact with the screen), the user may touch the screen with a second finger and the system may show the second keypad 342109 under said second finger. The user then may provide the gliding action 342102 corresponding to his desired punctuation mark character "-" 342115.

According to one embodiment, instead of drawing a circle corresponding to the group of characters, the user may tap on the screen, wherein according to another embodiment a tapping action may correspond to entering a dot. Accordingly, a short sliding action on the screen in lower right direction may correspond to a comma.

According to one embodiment of the invention, the system may include two predefined zones near opposite edges of the touch sensitive surface defining Space button 342006 and Backspace button 342005 of the invention and their corresponding various functions as were described before. In this case, preferably, the drawing/writing shapes for entering words/letters may be provided further inside the screen close to the edges as shown in fig. 342A.

According to one embodiment of the invention, the actions corresponding to drawing the common portions/shapes may be provided by the user in the air. The user may draw by using his hand, finger, a remote (laser) device, etc. In this case, the system may use a camera or another corresponding device for capturing said movements, a recognition system for recognizing them and attributing them to the corresponding groups of characters. Obviously, the system may include and use at least some of the features described throughout this application such as word predictive system.

By providing the embodiment just described, a writing method and alphabet including only few shapes such as four shapes/letters (e.g. four ambiguous letters) and using a predictive data entry system may provide high accuracy and fast text entry. Using such an alphabet may enable development of different devices and technologies.

According to another embodiment, the user may provide interactions with a surface or in the air with his fingers. For example, a tapping action with one finger (on the surface) or showing one finger may correspond to the group of characters standing on one point, a tapping action with two fingers or showing two fingers may correspond to the group of characters standing on two points, forming and showing a circle with the fingers may correspond to the group of letters having a closed circle in their shape, and forming a horizontal line with a finger may correspond to the group of letters having a large base.

According to one embodiment, a stylus having a motion sensor may be used to recognize said portions/shapes when the user writes them with said stylus. According to another method, a touch sensitive surface may be structured so that drawing said portions/shapes on said surface by a tip of a stylus provides a different sound for each of said portions/shapes. Since the number of common said portions/shapes is preferably limited, the corresponding recognition system may recognize the corresponding group with high accuracy. According to another method, said structured surface may be the tip of an stylus designed such that drawing said portions/shapes on said surface by the tip of said stylus provides a different sound for each of said portions.

As mentioned before, modifications to the second keypad of the invention may be applied. As an example, fig. 345A shows a first keypad of the invention similar to that shown in fig. 343 but with slight modification. In this example, the locations of the letters "O" 345017 and "P" 345018 printed on the key 345004 of the first keypad are swapped. This change is based on the exchange of locations of the letters "O", and "P", on the keys 345023 and 345026 of the second keypad 345010 of the invention. In addition, special characters, "-" 345013 which may be part of a word such as "semi-final", " ' " 345011 which may be part of a word such as "it's", "@" 345014 which may be part of a word such as "ben@snapz", and "Π" 345012 which may be part of a compound word as described before, may be assigned to the corresponding keys, respectively, 345003, 345001, 345004, and 345002.

It must be noted that according to one method, the arrangement of the letters on the keys of the second keypad of the invention may be decided and applied by people skilled in the art and/or even by the end user at his convenience. Accordingly, the appropriate letters on each key of the first keypad may be printed on said key of the first keypad according to the arrangement of the letters on said letters on the keys of the second keypad. It must be noted that, as generally described throughout this application, when a user touches or interacts with a key of the first keypad of the invention to provide a gliding action for entering a letter/character precisely, said gliding action generally duplicates (i.e. simulates) a gliding action provided on a (e.g. an imaginary) second keypad model of the invention in the direction of letter/character on said keypad model (e.g. same direction), wherein, generally, said gliding action on the keypad model is provided from a predefined key such as the center key of said keypad model towards another key on said keypad model including the letter/character desired by the user, wherein the locations of the letters on a key of the first keypad relating to each other are similar to the location of the same letters on the keys of the second keypad model relating to each other. This is the reason why in many embodiments, a second keypad of the invention is mentioned, wherein in many cases said second keypad of the invention is an imaginary keypad model.

According to one embodiment of the invention, a key of the first keypad of the invention and a key of the (e.g. imaginary) second keypad (model) of the invention may have more than one character in common. In this case, according to one method, during the entry of a word the system may consider all of the common characters, for example, during the correction procedure, if the user interacts with a key of the second keypad having more than character in common with its corresponding key of the first keypad related to a same ambiguous character within the current predicted word, the system may consider all of the common characters that they have in common and the rest of the input information provided for entering the desired word, and for example provides/predicts a new word among all of the words corresponding to the combined input information.

According to one embodiment of the invention, the second keypad of the invention may be shown to the user as a QWERTY keyboard such as the one 345119 shown in fig. 345B. The user may interact with its keys, by for example, pressing its keys. According to one method, the keys of said keyboard 345119 may be shown as forming several groups of character, in this example, nine groups, and the user may select a character for example during the correction procedure, by providing a corresponding gliding action 345116 departing for example, from the center group, in this example 345115, towards a desired group, in this example 345114, in which his desired character is included. In this example, exceptionally, characters "P" 345117, and "T" 345118, are respectively assigned to their neighboring groups as indicated by the arrows printed on the respective keys of the keyboard 345119, and are also printed, on the appropriate locations, respectively, on the keys 345104 and 345101, of the first keypad. According to one embodiment of the invention, during the correction procedure, if the selected ambiguous character within a predicted word is the same as the letter in the same position within the word that the user intends to enter, a tapping action (anywhere) on the screen preferably outside the keys of the first keypad, may indicate to the system to change the status of said selected character from ambiguous to precise. By considering said precise character and the rest of the input information provided by the user for entering his desired word, the system may propose/predict a new word corresponding to the combined input information provided by the user for entering said desired word. The system may then select a next (e.g. ambiguous) character of said new proposed/predicted word as described in detail before.

The tapping action just described may duplicate the gliding action provided for example from the center key of the (imaginary) second keypad (model) in the direction of the same desired letter on said key.

In the current embodiment, according to a preferred method, in order to enter precisely a character located in the center of a key of the second keypad (e.g. as mentioned, when a gliding action is provided on/from a key of the first keypad, said key may enter into its precise mode, therefore it may be considered as corresponding a second keypad corresponding to said first keypad) of the invention during the entry (e.g. using a correction or insertion procedure) of a word, the user may provide a back-and-forth gliding action, or a short gliding action, etc., (e.g. depending on the embodiment, respectively, on the screen outside the keys of the first keypad, or on the keys of the first keypad) as described before.

Fig. 346 shows another example of the distribution of letters on the keys of the

(imaginary) second keypad model 346010 of the invention, wherein the letters are also printed in the corresponding positions on each of the keys 346001-346004 of the first keypad. In this example, the center key 346025 of the second keypad does not include any letter/character, therefore, none of the keys of the first keypad have a letter printed on its center position. In this example, the letters "F", "G", and "H", are assigned to, respectively, the keys 346024, 346022, and 346022 of the second keypad 346010, and are also printed on the corresponding locations, respectively, 346011, 346014, and 346012, of the corresponding keys, respectively, 346001, 346004, and 346002. In this example, the letter "U" is assigned to the key 346023 of the second keypad 346010, and therefore, is also printed on the upper-right position 346013 of the key 346014 of the first keypad. Note that in this example, each group of letters assigned to a different key of the first keypad has the same letters as shown and described in fig. 345 A, with the difference being that the groups of letters that were respectively assigned to the keys 345003 and 345004, of fig. 345 A, are here assigned, respectively, to the keys 346004 and 346003 (the groups are swapped). It must be noted, that the assignment of the letters to the keys of the second keypad of the invention is such that to substantially remind their location on a PC (e.g. QWERTY) keyboard based on the principles as described before and shown as an example, in fig. 345A and fig. 343.

According to one embodiment of the invention, the system may be designed such that when the user provides the input information corresponding to a desired word and the system predicts a word based on that information, the system may select/highlight a (e.g. preferably, an ambiguous) letter of the predicted word such that to generally minimize required gliding actions and preferably maximize required tapping actions for entering the other words of the word database corresponding to the input information provided until that moment.

As an example, Fig. 347 shows an exemplary portion 347000 of the word database (e.g. a N- gram word database) of the system by considering the first keypad of the invention as shown in fig. 346. In this example, after the user taps the keys 346004, 346003, 346002, and 346004, the system may predict the word 'SOME". According to one method, as an example, the system then may select highlight the last letter "E" of the predicted word. Now, several scenarios may be considered:

1^st scenarios: As shown in the exemplary word instances 347011, after the system predicts the word "SOME", if the user provides a predefined interaction such as taps outsides the keys of the first keypad, the system may show another word of the portion

347000 of the word database, wherein said word corresponds to the combined input information provided until that moment (e.g. information provided during initial tapping actions and the information (e.g. in this example, the tapping action) provided during the correction procedure), and having for example the highest priority among the remaining words of the portion 347000 of the database. In this example, said word is the word

"COME". Preferably, as an example, the system then may select/highlight the letter "O" in the new predicted word "COME". Now, if the user again taps outsides the keys of the first keypad, the system may show another word of the portion 347000 of the word database, in this example, the word "GONE" wherein its second letter is the letter "O" and ends with the letter "E", and wherein said word corresponds to the combined input information provided until that moment, and having for example the highest priority among the remaining words of the portion 347000 of the database. Now, again as an example, the system may select/highlight the third letter 'TST of the last predicted word "GONE". At this time, if the user again taps outsides the keys of the first keypad, the system may predict/propose another word "ZONE" wherein its second letter is the word "O", its third letter is the letter "N", and its last letter is the letter "E", and wherein said word corresponds to the combined input information provided until that moment, and having for example the highest priority among the remaining words of the portion 347000 of the database. In the examples above, by providing only tapping action during the correction procedure, three different words of the portion 347000 of the word database can be entered.

2^nd scenario: As shown in the exemplary word instances 347012, after the system predicts the word "SOME", if the user provides a gliding action corresponding to the letter "G", the system may propose a new word having the precise letter "G" in its last letter position, and wherein said word corresponds to the combined input information provided until that moment, and having for example the highest priority among the remaining words of the portion 347000 of the database. In this example, said word is the word "LONG".

3^rd scenario: As shown in the exemplary word instances 347013, after the system predicts the word "SOME", if the user provides a gliding action corresponding to the precise letter "L", the system may propose a new word having the letter "L" in its last letter position, and wherein said word corresponds to the combined input information provided until that moment, and having for example the highest priority among the remaining words of the portion 347000 of the database.. In this example, said word is the word "GOAL". At this time, the system may select/highlight the second letter "O" of the new predicted word. Now, as an example, if the user taps outsides the keys of the first keypad, the system may predict/propose another word "COAL" wherein its second letter is the letter "O", and its last letter is the letter "L", and wherein said word corresponds to the combined input information provided until that moment, and having for example the highest priority among the remaining words of the portion 347000 of the database.

In the examples described above, most of the words of that portion 347000 of the dictionary are predicted by providing tapping actions rather the gliding action. The system preferably is configured such that after providing each precise character during the correction procedure, it dynamically highlights/selects a (e.g. preferably, an ambiguous) letter of the new proposed word minimizing the need of the gliding actions required for entering the remaining words of the database corresponding to the combined input information provided until that moment. By considering the embodiments and descriptions described and/or referred to throughout this application, a preferred embodiment of the invention is described hereafter.

Entering a word

To enter a word, preferably, the user taps on the appropriate keys of the first keypad of the invention to enter a word ambiguously. The system may predict a word as described throughout this application, and preferably selects/highlights the first ambiguous character/letter of the proposed word.

Correction - Case 1

After the user enters the word, if the system does not predict the user's desired word, but the selected/highlighted letter is what the user wants, he may provide a tapping action outside the keys of the first keypad on the touch sensitive surface. According to one method, the system may change the status of the selected/highlighted letter from ambiguous to precise, may predict a new word that may include that precise letter in the selected/highlighted position as described before, and may select/highlight a next (e.g. ambiguous) letter of the new word.

Correction - Case 2

After the user enters the word, if the system does not predict the desired word and the selected/highlighted letter is also not what the user desires, he may touch (e.g. and hold) the screen OUTSIDE the keys of the first keypad. According to one example/method, the system relates/regards the user's touching action to/as the beginning/activation

(providing) a correction procedure corresponding to the key of the first keypad relating to the said selected/highlighted letter. According to a preferred example/method, the system may enlarge the key of the first keypad relating to the said selected/highlighted letter and preferably, locates it under the user's finger. The user may SLIDE (his finger) in the direction corresponding to the desired letter on said key to enter said letter precisely. (Preferably, when the user removes his finger from the screen) the system may replace the selected/highlighted letter by the precise letter he just entered, may predict a new word including said precise letter in said position, and may select/highlight a next (e.g. ambiguous) letter of the new word. Preferably, when the user touches the appropriate key of the first keypad, said key may be enlarged under his finger.

According to a preferred example/method, the center key of said key may be located under the user's finger when he touches the touch sensitive surface.

Insertion At any time during typing a desired word, the user may touch (and hold) the appropriate key of the first keypad, and SLIDE his finger in the direction corresponding to a desired letter on said key to enter said letter precisely. For the other letters, he may tap on the appropriate keys of the first keypad to enter the other letters ambiguously. By considering the precise and ambiguous letters, the system may better predict a word as described throughout this patent application. Preferably, when the user touches the appropriate key for sliding, said key may be enlarged under his finger. According to a preferred example/method, the center of said key may be located under the user's finger when he touches the screen.

Note that the user may repeat any of the above correction and/or insertion procedures for more than one letter. To enter a word which is not in the dictionary or database used by the system, the user may spell out all of the letters precisely one by one using any of the procedures described above.

It must be noted that preferably the arrangement/printing of the letters on the keys of the first keypad may be such that to substantially correspond to a three-by-three matrixed keypad. Preferably, the letter arrangement/printing on said keys of the first keypad may reminds their location on a PC keyboard such as a QWERTY keyboard, or a telephone keypad, or defined by the user, or etc., as described throughout this application.

The common shapes/characteristics of the letters of a group of letters assigned to a key of the first keypad of the invention, as described above, may form at least one image such as a face. According to one embodiment of the invention, as shown in fig. 349 A, a face 349000 may be created by using the common shapes as described earlier. In this example, the nose may be represented by one point 349001 which is the common shape/characteristic of the letters standing on one point, the eyes may be represented by a two points 349002/1 and 349002/2 which is the common shape/ characteristic of the letters standing on a two points, the head may be represented by a circle 349003, which is the common shape/ characteristic of the letters having a closed circle in their form, and the lip may be represented by a large line/base 349004, which is the common shape/ characteristic of the letters standing on a large base.

According to one embodiment of the invention, the face 349000 may be used as an input interface of the system. Each of the common shapes/characteristics 349001-349004 may represent a (soft/hard) key of the system. To enter a word the user may press on the

corresponding keys on said face so that the system may predict a word. If the interface 349000 is shown on a sensitive surface the user may also enter precise characters through gliding actions on/from the keys as described earlier in this application. The user may also use a mouse to interact with said keys. The face just described may be used as a logo of the system of the invention and may be used within a game. Said game may be a game that uses text/data entry system of the invention using the face 349000.

According to one embodiment of the invention, the face just described may include additional features. As an example, fig. 349B shows the face 349100 having a headphone 349008 including two speaker shapes (features) 349005, and 349006, representing respectively, the backspace key and space key of the invention.

It must be noted that keys represented by the shapes of the face just described may include all of the symbols of the corresponding keys (e.g. input means) of the first keypad (e.g the first input means) of the invention.

According to one embodiment of the invention, a method of defining dynamically the keys of the (first) keypad of the invention may be provided.

According to this method, if the keypad model of the system is a keypad such as the keypad of fig. 346, then as shown in fig. 350A, the user may first provide a calibration procedure such as pressing on the touch sensitive surface 350009 simultaneously with two fingers. The system defines two imaginary borders 350011, and 350012 for each finger. Now, let's focus on the dynamic keys corresponding to the right portion of the (imaginary) keypad model by preferably using a finger (e.g. preferably, the corresponding thumb) on the right side of the screen, the finger related to the left portion of the keypad model (preferably on the left side of the touch sensitive surface) may follow the same rules that will be described hereafter.

If the user touches a location 350001 on the surface on the top of the border 350012, the system may consider said new touching action and its impact as interacting with the upper key of the two letter keys of the keypad model, corresponding to said finger. According to one method, the system then may consider a dynamic key at said location 350001, said key preferably having a limited size because if the user consecutively represses the same key, most probably said new pressing/touching action will not be provided far from said location 350001 on the screen. On the other hand, as shown in fig. 350B, if the user desires to interact with the other dynamic letter key relating to the lower-right side of the keypad model corresponding for example to the same finger, then he most probably will press a location 350002 under said upper dynamic key/location 350001 outside said key/location 350001. Now, the system may provide a dynamic key 350002 preferably with the size as described for the key 350001. The procedure just described may be reversed, if the user presses/interacts under a location of said border 350012 preferably with the corresponding finger, the first time after the calibration procedure.

It must be noted that each time the user presses a key consecutively, the system may adjust the location of the corresponding new key dynamically so that preferably the center of said key to be the under user's finger impact on the sensitive surface.

According to one method, after defining a new dynamic key such as the key 350002 based on the principles just described, the system may erase the location of the other key (e.g. the key 350001) corresponding to said finger.

According to one method, providing a finger interaction such as a tapping action on a location 350003 on the right side of the impacts 350001, 350002 (i.e., dynamic keys) as shown in fig. 350B, outside said impacts may correspond to interacting with the space key of the keypad model.

It must be noted that interacting with a location/zone may be any type of interactions such as a tapping action for entering an ambiguous character relating to the dynamic key and obviously to its corresponding key of the keypad model, or a gliding action to providing a precise character assigned to said dynamic key and obviously to its corresponding key of the keypad model.

Obviously, the principles of defining dynamically the letter keys corresponding to the right side of the keypad model (e.g. preferably by using a right finger) may be applied to the left side of the keypad model (e.g. preferably by using the left finger). In this case according to one method, providing a finger interaction such as a tapping action on a location on the left side of the impacts corresponding to the letter keys of the left side of the keypad model (e.g. preferably, by using the left finger), may correspond to interacting with the backspace key of the keypad model.

The principles of defining the dynamic letter keys of the first keypad of the invention as described, may give a great freedom to the user with very high accuracy.

It must be noted that the dynamic keys may be shown on a screen or they may be invisible.

It must be noted that the example of the dynamic keys just described is based on a keypad model with letter keys defining arranged as two columns of letter keys. Said keypad model may have any other configuration. For example, the letter keys of said keypad model may form two horizontal rows of keys. In this case, the definition of the location of the dynamic keys may be based on corresponding interactions impacts with the surface (e.g. touching, pressing, tapping, gliding, etc.) in horizontal manner with each finger.

According to one embodiment of the invention as shown in fig. 350c as an example, after the user provides a calibration procedure and defines the right and left impacts/borders 350011 and 350102, corresponding to calibrating left and right dynamic keys relating to the corresponding keypad model, the system may dynamically define different zones, preferably vertical zones, on the sensitive surface corresponding to defining different dynamic keys. As an example, the vertical zones 350021 and 350022 which include the corresponding impacts 250011 and 350012, may be assigned to dynamically interacting with the corresponding left and right letter keys of the keypad model, respectively. The zone 350025 situated between the zones 350021 and 350022 may be assigned to interacting with the second keypad of the invention, and the vertical zones 350023 and 350024 may be assigned to interacting dynamically respectively, backspace and space keys of the first keypad model of the invention. The zones just defined may be dynamically modified during dynamic interactions with the keys of the first keypad model, because a user almost never interacts in an exactly same location.

With continuous description of the current embodiment, as an example, after providing a dynamic interaction corresponding to interacting with the keys of the first keypad model of the invention, if a user provides another interaction such as a gliding action 350008 on the zone 350025 of the screen, far from the dynamic keys inside the screen (e.g. for interacting with the second keypad of the invention for example during the correction procedure or for entering a special character), the user may bring his/her finger into the zone near one of the calibration impacts such as for example, 350012 and may desire to provide an interaction such as a tapping action corresponding to a dynamic letter key of the first keypad on that zone. Most probably, the user may not remember the location of his previous tapping action on that zone on the screen. In this case, and preferably because the user generally provides the interaction with the second keypad of the invention while holding the device, according to one method, instead of providing a new calibrating action, as shown in fig. 350d, the user may preferably provide his/her current dynamic interaction with the desired key of the imaginary keypad model, either far up 350101 from the last calibration impact 350012 in said zone 350022 if he intends to interact with the upper right key of the keypad model, or far down 350102 in said zone 350022 if he intends to interact with the lower right key of the keypad model, preferably with the corresponding finger. By doing so, any ambiguity will be avoided regarding system identification of the dynamic key being interacted. With continuous description of the current embodiment, according to one example as shown in fig. 350e, after the user brings back his finger in the zone corresponding to provide a next dynamic interaction with the letter keys of the keypad model and for example provides an interaction such as tapping action far up 350101 on the screen as just described, if the user intends to again interact with the same key, then he may provide the next interaction at substantially the same location 350101 within the borders of the corresponding dynamic key created based on the corresponding previous impact, but if the user intends to interact dynamically with the lower right key of the keypad model, he then may provide the next interaction at any lower location (e.g. 350112) outside the dynamic key relating to the location 350101 in the corresponding zone 350022. Then, as an example shown in fig. 350f, the next dynamic interaction with the upper right key of the keypad model may be provided anywhere in the corresponding zone 350022, such as the location 350201, above the key borders of the last impact 350112 that was provided relating to the lower right key of the keypad model.

The examples above may be reversed if after the user brings back his finger in the letter keys zone 350022 for example after interaction in the zone 350025 corresponding to the correction procedure, and for example provides an interaction with the lower right key of the keypad model by for example providing an interaction on location 350102 as shown in fig. 350d. In this case, if the user desires to consecutively interact again with the same key of the keypad model, he may provide another interaction on the same location 350102 as shown in fig. 350d, but if the user desires to interact with the upper right key of the keypad model, he may provide an interaction anywhere above that location 350102 outside the corresponding dynamic key border. In this case, the next interaction with the lower-right key of the keypad model may be provided anywhere below the last impact corresponding to the upper-right key of the keypad model out side the last dynamic key.

Based on the principles just described, according to one embodiment of the invention, the procedure of calibration may be avoided. As an example, when a user wants to begin to type, no matter if he typed before or he simply wants to begin to type, he may provide (e.g. on the appropriate side of the touch sensitive surface) his first dynamic interaction corresponding to a letter key of a corresponding side of the keypad model, preferably far up or far down (e.g.

preferably near the corresponding edges) of the touch sensitive surface, depending on if he desires to interact with, respectively, the upper key or the lower right key of the appropriate side of the imaginary keypad. The user then, may continue to type by adjusting the following dynamic key interactions by interacting in the appropriate dynamic locations as described above. The principles of dynamic interactions applied/related to the keys of the right side of the keypad model may apply to the keys relating to the left side of the keypad model as well, preferably interaction provided by a left finger (e.g. thumb) of the user.

It must be noted that although in the embodiments above, interacting on right side and left side of a touch sensitive surface have been mentioned to describe the appropriate embodiments, the interactions relating to the keypad model as described, may be provided on a same side of a surface so that the system may be capable of relating a dynamic interaction to a key of a corresponding side of the keypad model based on the relationship between the interactions provided on two (left and right) imaginary vertical letter zones.

It must be noted that dynamic interactions relating to the keys of the keypad model may be any interaction such as tapping or gliding (actions) that may respectively correspond to entering ambiguous and precise characters as described throughout this application.

It must be noted that if the user provides the dynamic interactions corresponding to the letter keys of the keypad model too close to any of the edges of the sensitive surface such as a touch screen, then the system may alert the user, and may not consider them. On the other hand, if the user provides the tapping actions corresponding to the letter keys of the two sides of the keypad model too close to each other (e.g. reducing the middle zone to a size not being practical, for example, to provide comfortable interactions relating to the second keypad of the invention), then the system may alert the user accordingly and may not consider them.

According to one embodiment of the invention, each time the user provides an interaction corresponding to a dynamic letter key of the first keypad as described, the system may locate the corresponding backspace key or the corresponding space key in the appropriate location next to said letter key as defined in the keypad model.

It must be noted that the examples described above, are provided only to describe the corresponding exemplary embodiments. It is noted that any other arrangement of keys of the keypad model, and the corresponding zones of the dynamic keys and interaction may be considered by people skilled in the art. For example the letter keys of each side of the keypad model may be arranged horizontally.

The principles of dynamically interacting with the keys of an imaginary keypad model as just described, are described as relating to text/data entry, in the corresponding exemplary embodiments. Obviously, said principles may be applied for interacting with the keys of a keypad model in general for any appropriate purpose. Said principles may be applied to dynamically interacting with the keypad model in general, regardless of what is assigned to its keys. According to one embodiment of the invention, as shown in fig. 351 A, another type of configuration of the letters on the keys of the second (e.g. imaginary) keypad (e.g. model) of the invention may be considered. The printing locations of letters on the keys of first keypad may be defined accordingly to remind the gliding directions for the entry of precise characters. This arrangement method is provided on a three by three matrixed keypad 351020, substantially based on the locations of the letters on a QWERTY keyboard divided in a three by three group, as was shown as an another example previously such as in fig. 345 A. In the current embodiment, exceptionally, the letters Q and P are respectively assigned to the keys 351207, and 351029, and the letters G and H are assigned to the key 351022, of the second keypad of the invention 351020. The printing locations 351011, 351012, 351013, 351014, respectively, of the letters G, H, Q, and P, are shown on the corresponding keys of the first keypad, accordingly. Other letters of the keyboard 351020 are grouped based on their shapes and their locations on said keyboard and are assigned to the corresponding keys of the second keypad of the invention. The letters are grouped such that any key of the first keypad and any key of the second keypad have at most one common letter so that to conform to the rules of the correction procedure of the invention.

Combined words

Combined words (e.g. compound words) and the ways of using them with the word predictive system have been described before. According to one embodiment of the invention, the system may have/use a number of compound words integrated in a/the dictionary/database. According to another embodiment instead of or in addition to said number of integrated compound words, the system may permit a user to define (e.g. and add to the dictionary) a number of compound words being used by the system.

According to a first method, after a user enters a set of several, preferably consecutive, words the user may select them and ask the system to add them in its/a dictionary as a (e.g. single) compound word, and use it as a single word, for example, next time the user desires to enter said compound word. Said consecutive words may be single words being entered using the first keypad of the invention with or without having gone through the correction procedure of the invention.

According to another method, in addition to or instead of said first method, the user may select several, preferably consecutive, words of included in any document and inform the system to consider them as a compound word and add said compound word in its/a dictionary as a (e.g. single word. Said document may be any type of document such as a text document created by the user or a document created by a third party (e.g. an electronic document of a news agency).

With continuous description of the embodiment, Fig. 351 A shows as an example, an electronic device 351000, having a screen 351009 on which a text is printed/entered. In this example the document is created by the user. The device also includes the data entry system of the invention including a first keypad of the invention having the keys 351001-351006, including the letter keys 351001-351004. This type of keypad has been described before throughout this application.

In this example, the document may either being a text file opened by the user, or it may have been created by the user. As an example, if the document is created by the user, the words "MY NAME" 351051 may have been entered by pressing the corresponding keys of the first keypad, and using the correction procedure of the invention. For example, for the word "MY", the system at first may have predicted the word "AT" based on the tapping action provided by the user on the corresponding keys of the first keypad, and after the user provided a

corresponding correction procedure, the system may have predicted the word "MY". Same may happen with the word "NAME", wherein the system at first may have proposed/predicted the word "MAKE", and then the user may perform/provide the corresponding correction procedure so that the system predicted the word "NAME". With continuous description of the current embodiment, as an example, at any time the user may select some, preferably consecutive, words such as the words "MY NAME" 351051, by for example, using the native selecting functions of the device, and requiring/informing the system to add the selected words as a single compound word to a/the dictionary. According to one method, after the user selects the several, preferably consecutive, words, he may use a predefined procedure to add them (e.g. a single compound word) to the word database used by the system. As an example, the user may select several words such as the words "MY NAME" 351019, using a predefined function, such as the "EDIT WORD" (or e.g. the "USE WORD") function of the system. He/She may do so by, for example, providing a gliding action 351017, so that the system may activate/select those words "MY NAME" 351019 for editing. The user then may confirm said words by, for example, interacting with (e.g pressing, tapping, gliding, touching etc.) the space key or any other end-of-the-word character/function. According to one method, the system may consider the selected words as a single compound word (e.g. as a single word) including space character between the words of the compound word. As mentioned, according to one method, a compound word (e.g. 351019 presented to the user and/or included in the dictionary may include a connecting character such as the character "Π" 351016, or a "_" (e.g. underscore) character, or any other predefined character, instead of the space character between the single words of the compound word.

After entering the compound word, the user knows that the compound word exists in the dictionary. Next time the user wants to enter those single words (e.g. "MY NAME"), he may enter them as a compound word by interacting with (e.g. tapping on) the corresponding keys of the first keypad including interaction with the key (e.g. in this example, the key 351002) which includes the connector sign (e.g. in this example, 351018) between the key interactions corresponding to the word "MY" and the word "NAME". In this example, the only word of the corresponding dictionary of the words that are related to pressing/tapping action on the keys 346002, 346001, 346002, (e.g. corresponding to the word "MY" and the connector character "Π"), 346002, 346002, 346002, 346004, is the compound word "ΜΥΠΝΑΜΕ". By doing so, the system may be able to enter the words "MY NAME", by receiving the input information from the first keypad of the invention alone, without requiring additional information from the correction procedure.

The use of compound words with the data entry system of the invention reduces or eliminates the use of the correction procedure for predicting a word by the system. This is because a compound word is longer than each single word that is part of said compound word, and therefore the chances of predicting it by the system by eliminating or reducing the number of corrections are high.

It must be noted that according to one embodiment of the invention, a compound word may be considered a single word of the dictionary, therefore the correction procedure of the invention may be applied to a compound word as if it were a single word. As such, during the entry of a compound word, the first character of word predicted by the system may be highlighted. Said predicted word may be a compound word. By proceeding to the correction procedure, the user may replace the highlighted character by a precise character, regardless of the word being a compound word or a single word. Then, the system may propose another word, which may be a single word, or a compound word.

Using combined/compound word with the word predictive system may be beneficial in languages such as French, wherein many short single words are included in the dictionary.

It must be noted that after the user adds a compound word to the dictionary, the system may replace the space(s) between words of the compound word by a special character such as a "_" or "Π", as described before. Also as an example, when the user enters a word and the system predicts a compound word, the spaces within said compound word may be shown by said special character until the user confirms said compound word. In this case the system may replace the special character(s) by space character(s) in the corresponding document.

It must be noted that according to the one embodiment, a compound word may include two or more words, a phrase, a paragraph, or even a text document. Said text document may also include non-text portions.

According to one embodiment, the native word database of the system may include a number of compound words, and the user may add additional compound words to the database, while according to another embodiment, the native word database of the system may not include compound words, and the user may add compound words to the database.

According to one embodiment, if the user adds several consecutive words as a single compound word to the word database used by the system, if such compound word exists already in the word database, then the system may alert the user.

It must be noted that in order to define several words as forming a compound word and add the compound word to the database used by the system, one may use any method such as selecting words of a printed document on the screen, or typing several, preferably consecutive, words and adding them as a single compound word to the word database used by the system while said words are being entered, by for example, providing, preferably precisely, a connecting character between said words while said words are being entered. At the end of the entry of said words forming the compound word, the user may provide an end of the word signal such as a special character (e.g. space character) or a function.

According to one embodiment of the invention, the system may use one or more remote dictionaries or corpus of words for entering single or compound words. For example, each time a user enters a word including precise and/or ambiguous characters, the system may send the input information corresponding to one, some, or all, of the corresponding word(s) of the corresponding input information to a search engine of an internet service provider and based on the frequency of use response of the remote search engine, and/or other factors, if any, the system or the remote search engine may predict a corresponding single or compound word.

Note that, although the keypads of the data entry system of the invention are shown and described to include letters for entering text, obviously, the keys of the keypads of the invention may include any other type of symbol including or excluding letters for any type of entry for example entering shorthand writing or entering (e.g. combined) functions and/or commands. Preferably, said data entry may correspond to the principles of the data entry system as described throughout this and/or previous patent applications filed by this inventor. Selecting a portion of a text

According to one embodiment of the invention, selecting a portion of a text, such as selecting several words, may be assigned to a predefined key interaction. As shown as an example in fig. 352A, in order to select a portion of the text printed on the screen, a user may select a portion 352012 before the cursor 352014, by for example, pressing (e.g. and holding) an appropriate key such as the key 352005 and providing a predefined interaction such as a gliding action 352011 towards a left side such as the upper-left side on the screen. Said portion may be added as a compound word to the word database of the invention. According to one method, by performing a press and holding action on the key 352005 the system may open a second keypad of the invention 352013 relating to the functions assigned to providing an interaction such as the gliding action 352011 with an additional finger on the screen. This matter has already been described before.

With continuous description of the embodiment, as shown in fig. 352B, for selecting a portion 352111 of the text, located on the right side of the cursor 352014, the user may provide the gliding action 352115, towards a right side such as the upper-right side on the screen.

In the examples above, as an example, each gliding action may cause the system to select an additional unit of the text, such as a word or a character, as may have been defined.

Accordingly, a glide-and-holding action in left or right side may select several words/characters until the user removes his finger from the screen.

Proposing a word if there is only one choice

According to one embodiment of the invention, during the entry of a word, such as a single word or a compound word, if there is only one word corresponding the input information (e.g. key interaction) provided by the user, then the system may propose said word to the user even if the user has not finished to enter the whole input information corresponding to all of the characters of the desired word. If said proposed word is the desired word, then the user may enter it by providing a predefined interaction such as an interaction on/from a key. Fig. 352C shows as an example, the beginning of a combined word 352115 being displayed on the screen based on the corresponding input information entered by a user. At this time, the system may identify that there is only one word (e.g. in this example, the combined word 352117) in the word database corresponding to the input information provided by the user until that moment. The system then, may propose that word 352117 to the user. In this example, by providing a gliding action 352208 towards the right side on/from the space key 352006, the user may accept the proposed (combined) word 352117 and the system may enter it (e.g. prints it on the screen in the text).

Dynamic letter arrangement during correction

According to one embodiment of the invention, the default arrangement of the letters on the keys of the keypad relating to an imaginary second keypad of the invention as described before, may be considered in several different manners. For example, as shown on the key 352001 in fig. 352A, said arrangement may be such that when the user touches the key 352001 for example with his left thumb, he may be able to see at least the most commonly used letters on that key. Fig. 352B, shows said key 352001 and the letters printed on it when the user touches said key.

According to another method, as show on the key 352003, the letters may be arranged in alphabetical order on or around the surface of the key.

Note that, other methods of arrangement of letters on the keys may be considered by people skilled in the art.

Dynamic letter arrangement

According to one embodiment of the invention, the locations of the letters on the keys of the second keypad of the invention during a correction procedure may be dynamically defined based on their priorities within the group of words corresponding to a same input information, for example, the same key press sequence on the first keypad of the invention.

As an example, after receiving the input information corresponding to a user's desired word, in addition to predicting a word, the system may dynamically locate the ambiguous letters corresponding to the position of the highlighted ambiguous character of the predicted word on the locations on the second keypad of the invention such that to be easily visible and/or accessible to the user. According to a first method, said second keypad may appear under the user's finger when for example, the user touches the screen outside the first keypad of the invention, while according to another keypad, it may appear on a predefined location on the screen before the user touches the screen, for example, in a normal or in a reduced size. If it appears in a reduced size, then when the user touches the screen it may be enlarged under the user's finger/thumb.

With continuous description of the embodiment, as an example, by considering the database of words 353009 of fig. 353 A, after entering the input information corresponding to a desired word by for example pressing the keys 353001, 353004, 353002, 353003, the system may propose the word "YEAR" among the group of words corresponding to the same key interactions as shown in the portion 353009 of the exemplary word database used by the system. According to the current embodiment, if the user proceeds to the correction procedure, the location of the letters on the corresponding keys of the second keypad 353008in the current instance may dynamically be modified such that the user may easily see and/or access the desired letter relating to the highlighted letter position in the predicted word (e.g. in this example the position of the first letter "Y" of the predicted word "YEAR").

According to an arrangement method of the current embodiment, the system may locate/assign the highlighted letter to the center (key) 353011 of the second keypad 353008, such that if the user desires to enter a new word (e.g. in this example, the word "YEGO") with that letter "Y" in the same position within said new word, then he can just tap anywhere on the screen outside the keys of the first keypad (e.g. as described in detail before, according to one method, wherever the user taps on the screen is the center of the second keypad).

With continuous description of the current embodiment, the location of the other characters/letters on the keys of the second keypad 353008 in the current instance may dynamically be defined based on the other words position of the same group of words as shown in the portion 353009of the word database. For example, the first letter "F" of the word "FEAR" (e.g. having the next highest priority among the words beginning with a letter other than the letter "Y") may be assigned to the upper-right side/zone/key 353012 of the second keypad

353008, because if the user uses his left thumb to provide a correction, then, said location may be the most visible and/or accessible location to provide a corresponding gliding action. The next word in the group beginning with another letter in its first position is the word "TEND". The letter "T" which is the first letter of the word "TEND" may be placed on a location/key 353013 next to the letter "F" on the right side of the keypad 353008 such that also being easily accessible and/or visible to the user. Same principles may be applied to the next word beginning with a different letter, in this example, the word "JUMP" wherein the letter "J" is assigned to the visible/accessible position/key 353014, and so on. Letters that are not the first letter of any of the words of the group, or those relating to the words with lowest priorities may be placed on less comfortable locations/keys of the second keypad 353008.

With continuous embodiment of the invention, the same procedure may preferably be applied for the following highlighted letters of the predicted word. For example, if the user selects the letter "F", then, as shown if fig. 353B, the system may propose the word "FEAR". If this word is not the desired word, then, the other relevant words of the group after entering the precise character "F", are most probably, the words "FUND', "FLAP", and "FEND" having the letter "F" at the first letter position. Now, the highlighted letter of the predicted word is the second (e.g. ambiguous) letter, "E". Letter "E" may preferably be assigned to the center (key) of the second keypad 353108 in the current instance, so that if the user taps on the screen outside the keys of the first keypad, the system predicts the word "FEND" having the letter "E" in its second letter position.

This instance of the second keypad may preferably be used with the right finger, therefore, the second letter (e.g. "U") of the word "FUND" may be assigned to the upper left side (e.g. upper-left key) of the current second keypad 353108. The other letters that occur in the second letter position in the relevant words.may be assigned to other visible and/or accessible locations of the second keypad 353108 in the current instance. For example, the other different second letter of the relevant words of the group is the letter "L" of the relevant word "FLAP". In this example, the system may place that letter on the visible/accessible location/key 353113 on the second keypad 353108. In this example, if the user provides a gliding action corresponding to the letter "U", the system predicts the word "FUND", but if the user provides a gliding action corresponding to the letter "L", the system predicts the word "FLAP".

In the current embodiment, the procedure described above may be applied to any of the highlighted letter(s) of the proposed word (e.g. in the example above it may be applied to the following highlighted letters), regardless of if a said highlighted letter(s) is/are being (e.g.

consecutively or automatically) selected by the system or (e.g. arbitrarily/randomly) selected by the user.

It must be noted that according to one embodiment, the system may be defined so that the locations of the letters on the keys of the second keypad may be dynamically defined based on which instant of the second keypad is used with which user's thumb so that to enable the system to show in the best manner preferably all of the letters of the second keypad being interacted in appropriate visible/accessible locations on or outside (e.g around) said keypad.

According to one embodiment, the system' may be designed so that when the user touches the second keypad, the portion of the said second keypad being hidden by the user's thumb/finger, preferably, includes only the highlighted letter on its center location/key, as described.

Highlighting the relevant letters

According to one embodiment, instead of using a dynamic arrangement of letters of the second keypad, the default arrangement of the second keypad may remain the same, but the letters on said keypad corresponding to at least the most relevant words in a corresponding instance of the second keypad during/for the correction procedure may be dynamically highlighted to enable easier detection by the user. In the example of fig. 353A, instead of providing a dynamic letter arrangement, as shown in fig. 354, in the current correction instance, the letters "F" 354011, "T" 354012, and "J" 354013 of the second keypad 354008,

corresponding to the three most relevant words "FEAR", "TEND", and "JUMP" of the appropriate portion of the word database 353009 (fig. 353A) are dynamically highlighted.

According to one embodiment of the invention, the system may be designed to include substantially at least all of the compound/combined words having at least two words of less than three characters, wherein at least one of the single words of each compound word requires a correction procedure if it was entered as a single word.

The examples provided to describe the embodiments just described are used only as an example. Other methods and features according to the embodiments described may be used by people skilled in the art. For example, the second keypad of the invention may have distinctive keys shown to the user before he touches the screen, and the user can tap on its keys instead of providing gliding actions. Also as an example, other methods of showing the letters dynamically assigned to the keysAocations of the second keypad may be considered by people skilled in the art.

Derivatives usine one hand

According to one embodiment of the invention, the system may use a method to switch between different instances of the (e.g. second) keypads of the invention (e.g. by preferably using one hand). Said different instances may be of any kind and/or based on any principles such as those but not limited to instances described throughout different patent application filed by this inventor. For example, it may be used to enter derivative characters and functions.

According to one method, said means may be a back-and-forth gliding action provided on/from a surface such as the surface of a key.

As an example, fig. 355 A shows a device having a touch sensitive surface/screen, and having the first keypad of the invention including the letter keys 355001-355004, a backspace key, and a space key. In this example, the user just tapped on the key 355004, and the system predicted the letter "s". As an example, if the user desires to replace/correct the ambiguous letter "S" by the precise letter "e", he/she may first provide a gliding action 355011 outside the first keypad (e.g. on the second keypad 355008) in the direction of the letter "e". Then, as shown in fig. 355B, preferably, without removing his finger from the screen/surface, he may provide a reverse gliding action 355111 back approximately to the departure/center zone. At this time, the system may show a second keypad 355108 in an instance wherein each of its keys includes one of the accents/accented-characters corresponding to the letter "e" selected by the user. Now, as shown in fig. 355C, the user may provide a gliding action 355211 towards the desired accent/accented-character "e". According to one method, the gliding action 355211, may preferably be provided without removing the finger from the screen after providing the back and forth gliding actions 355011, and 355111, while according to another method, the gliding action 35521 lmay be provided after removing the finger from the screen after providing the back and forth gliding actions 355011, and 355111.

According to one embodiment, after providing the back and forth gliding actions

355011, and 355111, if the user removes his/her finger from the touch sensitive surface, then the system does nothing. At this time, according to one method, if the user interacts with one of the keys of the first keypad, the system removes the second keypad from the screen, while according to another method, after providing the back and forth gliding actions 355011, and 355111, if the user removes his/her finger from the touch sensitive surface, the system removes the second keypad from the screen.

According to one embodiment of the invention, the (e.g. the correction) procedures as described above may be applied to insert characters from the keys of the first keypad (e.g.

Insertions procedures of the invention that have been described before in detail), such as to insert derivative characters of a character. For example, the user may insert the character "e", by first providing the gliding action corresponding to the letter "e", from/on the keys 355004. Then the user provides the gliding action back to the departing point of the previous gliding action, preferably without removing his finger from the surface. Then the system may show the accents/accented-characters on a second keypad instance under the user's finger, and the user provides the desired gliding action on said keypad.

In the embodiments above, according to one method, if the user makes a mistake and provides the first gliding action in a wrong direction, then, preferably, the user may rectify his gliding action by continuing his/her gliding action by rectifying the trajectory of the gliding action towards the desired character without going through the departing point. After achieving (e.g.or arriving at) the intended character zone, the user may direct his/her gliding action towards the departing point and achieve (or arrive at) it, so that the system shows the derivative characters corresponding to the right character being selected.

It must be noted, that the procedures of using one hand as described above, may be used to provide consecutive second keypads of the invention such as those described regarding entry of any symbol such as an accented character, word completion, derivative of a character such as a dot, derivatives of a function, derivatives of a symbol representing currency symbols, etc. As an example, for languages such as Hindi, wherein the number of characters/letters assigned to a key require at least two instances during the insertion of a precise character or during correction procedure, then the first instance for example, having the characters used more often may be assigned to a first instance, and the characters used less often may be assigned to a second instance, wherein switching between said instances are made by using the procedures just described.

Changing modes (e.e. using one hand)

According to one embodiment of the invention, a means such as providing a predefined interaction with a key of the keypad of the invention, may enter at least all of the letter keys into their corresponding mode instances (e.g. a global mode instance). Fig. 356A shows as an example, a device having a touch sensitive surface/screen 356009, and having the first keypad of the invention including the letter keys 356001-356004, a backspace key 356005, and a space key 356006. In this example, the space key 356006 may be designed such that when the user provides a gliding action 356007 upward from/on said key, the system enters at least all of the letter keys into their corresponding mode instances as shown in fig. 356B, wherein the special characters on said keys are arranged according to (e.g. an imaginary) second keypad of the invention, as described throughout this application. The user now, may enter his/her special characters by interacting with the appropriate keys, by for example providing gliding actions (e.g. 356107) on/from said keys.

According to one method, tapping on a key of said keys 356001-356004 during the mode instance, may correspond to entering the center character, as described in detail before. In order to exit said global mode instances, the user may provide another type of interaction, or he may repeat the same interaction (e.g. 356007).

According to one embodiment of the invention, (e.g. during the global mode instance) the center keys of the second keypad of the invention may not include any special character, and if the user provides a tapping action on any of the letter keys, the system automatically may exit from the global mode instance. This avoids the additional interaction required in the

embodiment before to exit from the global mode instance. The user may provide gliding actions on/from the keys 356001-356004 to enter special characters during the global mode instance, and provide a tapping action on any of said keys to exit from the mode instance. Preferably, said tapping action relates to entry of an ambiguous character/letter on said key. According to one method, providing a backspace key interaction or providing a space key interaction exits the system from the global mode instance, in both embodiment described above.

Fig. 357 shows as an example, a function assignment method to the backspace key 357005, and the space key 357006, according to a preferred embodiment of the invention. In this example, said keys are shown in their enlarged status. As an example, the function "Shift" 357011 relating to capitalizing a next letter is assigned to the key 357005, and the function (change) "Mode" 357012 relating to entering (or exiting) at least allthe letter keys into (or from) a global mode instance, is assigned to the key 357006. Also as an example, the function 357013 confirming a single word corresponding to the input information of the beginning characters of the word presented to the user is here assigned to the keys 357006.

It must be noted that in some embodiments described, the letters on each of the letter keys of the first keypad are exemplary arranged in order to remind their assignment to the keys of an imaginary second keypad of the invention, preferably having an at least three-by-three matrix of keys, and in some cases having a matrix of three-by-four keys. Therefore, when the user provides the gliding actions on a key of the first keypad, said key in that instance (e.g. or a copy of said key that is relocated under the user's finger), may be considered as a second keypad of the invention duplicating a virtual second keypad model of the invention, and said gliding actions may duplicate similar gliding action provided on said keypad model, wherein the departure point of said gliding action preferably, correspond to the center key of the keypad model.

Mode - press/elide on any key + glide/tap on another key

The data entry system of the invention may include a means for entering the keys of the first keypad of the invention into their corresponding mode instances. According to one embodiment, (e.g. if the keys of the first keypad are located on a touch sensitive surface such as a touch screen) providing a gliding action towards a predefine direction such as downward on a key such as the space key, and/or the backspace key, of the first keypad may enter at least some, preferably all, of the keys of the first keypad of the system into the Special Symbols Mode instance. At this time, preferably, the system may show/display the special symbols such as special characters and functions on at least one of the corresponding keys of the first keypad (Displaying the special symbols on a key of the first keypad may be based on assigning them to the keys of a (e.g. an imaginary) keypad model such as a telephone-type keypad or a keypad having at least a three-by-three keys matrix. Selecting a special symbol of said key of the first keypad may be based on an appropriate tapping/gliding action provided on/from said key wherein tapping/gliding action duplicates a tapping/gliding action provided on the keypad model. This matter has already been described in detail).

After providing said gliding action (e.g. with a first finger/stylus):

according to one/first method, if the user removes his finger from the screen, the keys of the first keypad may enter into their corresponding special symbols/character mode instances. The user then may interact (e.g. tap for the center character, or glide in an appropriate direction for an appropriate special character) with an appropriate key of the first keypad to enter a desired special symbol such as a special character. After that, the system may automatically enter the keys of the first keypad back to their previous/default mode instances. Fig. 358A shows as an example, a device similar to that of fig. 356A having a sensitive surface 358009. In this example, the backspace key 358005 and the space key 358006 are shown in enlarged format 358015, and 358016, respectively. As shown as an example on the enlarged Space key 357016, said key includes a Mode means located on the zone 358012 on said key. Said mode means may be actuated by providing a gliding action 358032 downward (anywhere) on/from said key 358006 (e.g. as described before, when the user touches a key of the first keypad, a copy of said key may be shown on the screen, wherein its center zone being located under the user's finger impact on the screen). After providing said gliding action, the system may show at least one of the letter keys 358001-358004 in their special characters mode.

- according to another/second method, at this time, if the user (simultaneously or not simultaneously) interacts (e.g. with a second finger/stylus) with another key of the first keypad, said another key may enter into its corresponding special symbols mode instance (e.g. keys 358101 of fig. 358B), and the user may provide an appropriate interaction (e.g. tap, or glide 358111) with said another key to enter a desired special symbol assigned to said another key. According to this method, the user may enter consecutively several special symbols.

According to one embodiment, the system may include two means such as two gliding actions in each in a different direction, wherein a first of the means corresponds to the first method, and a second of the means corresponds to the second method just described above.

According to one embodiment of the invention, a back-and-forth gliding action on a key such as a back and forth gliding action on/from the space key in the direction of a zone on said key to which the Mode function is assigned/printed (e.g. in the example of fig. 358A, in the downward direction) and removing the finger from the screen may enter all of the keys of the first keypad into their corresponding mode instances. After that, interacting with the keys of the first keypad may correspond to entering special symbols, until the user provides an interaction such as repeating a similar back-and-forth interaction, to exit from said special symbols mode instance. In the example of fig. 358A, said mode means may be actuated by providing a back-and-forth gliding action 358031 downward (anywhere) on/from the key 358006. After providing said gliding action, the system may enter and show the kejs 358001-358004 in their special characters mode. Those keys may remain in said mode until the user provides a predefined interaction to exit said mode.

According to a preferred embodiment, each side of the split keypad may include a Mode means. In the example of fig. 358A, the backspace key 358005 may also include a mode means that may be activated by the gliding actions (e.g. here, 358033, and 358034) similar to the gliding actions 358031, and 358032, that are assigned to the space key 358006. It must be noted that the system may include several mode instances per key, and wherein by using said mode switching means the user may scroll/switch/activate the system into different mode instances.

The procedures just described may also be used for entry of capitalized characters by for example, using a Shift means. As an example, providing a gliding action towards the shift zone, on/from a key, for example the backspace key, may enter the system into the shift instance. In the example of fig. 358A, the Shift mode may be actuated by providing a gliding action 358022 upward (anywhere) on/from said key 358005. After providing said gliding action, the system may switch to shift instance and show the keys 358001-358004 in their capitalized characters mode. At this time:

- According to one method, if the user removes his finger from the screen, the next letter he types will be capitalized.

- According to another method, if the user does not remove his finger from the screen, and simultaneously types several letters, those letters will be capitalized until he removes his first finger from the screen.

In the embodiments just described and in other embodiments of this application, the arrangement of keys, the location of letters/symbols/functions on the keys, types of interactions with keys, etc., are provided as an example to describe the/different principles/features/concepts of the data entry system of the invention. Other arrangements of keys, the location of letters/symbols/functions on the keys, types of interactions with keys, etc., may also be considered. For example, after the key(s) is/are entered into a special character mode, instead of a gliding action towards a symbol, a tapping action on (the zone corresponding to) said symbol may enter said symbol precisely. According to one embodiment of the invention, a back-and-forth gliding action in the direction of the Shift means on said keypad and removing the finger from the screen may enter the system into a CapsLock mode, until the user provides an interaction such as repeating said back-and- forth interaction, to exit from said mode instance. In the example of fig. 358A, said mode means may be actuated/activated by providing a back-and-forth gliding action 358021 upwards (anywhere) on/from said key 358006. After actuating/activating said means, the letter keys 358001-358004 may enter into the capitalization (CapsLock) mode. Those keys may remain in said mode and letters entered may be capitalized (e.g. unless a shift mode is also

actuated/activated), until the user provides a predefined interaction to exit from the said mode.

According to a preferred embodiment, each side of the split keypad may include a Shift function means. In the example of fig. 358A, the space key 358006 may also include a Shift means that may be activated by the gliding actions (e.g. here, 358023, and 358024) similar to the gliding actions 358021, and 358022, that are assigned to the backspace key 358005.

It must be noted, that the Special Character Mode and Shift functions may be assigned to any other predefined key of the first keypad by people skilled in the art.

At least any of the space and backspace keys may also be entered into their

corresponding mode using any of the procedures just described. For example, after the user activates the mode instance by providing a back-and-forth gliding action 358031, the backspace and space keys may also enter into their corresponding Mode instances. Fig. 358C shows those keys 358205, and 358206, in their mode instance. Fig. 358D shows as an example, the key 358006 in its mode instance being enlarged after for example the user touches said key and provides the gliding action 358315 relating to a procedure of switching the system into a/another language. Then, preferably, the system may propose a list of different languages. The names of such languages may be shown on a second keypad of the invention 358418 as shown in fig. 358E. Preferably, said second keypad may appear under the user's finger after the user touches anywhere on the screen to provide a corresponding tapping or gliding action to select another language. In this example, the user provides the gliding action 358415 to switch the system to Spanish language. At this time, the system begins to use a database of words in Spanish language. According to one method, after providing a gliding action (e.g. 358315) if the user does not desire to select any other language, then an interaction such as a backspace interaction or another interaction such as continuing typing letters/words by tapping on a letter key may cancel said procedure. To switch back to the previous mode, according to one method, the user may provide a predefined interaction such as a back-and-forth gliding action as described above, such as the back-and-forth gliding action 358431.

Changing instances using back-and-forth gliding action

As described before, a back-and-forth gliding action may be used to enter/switch the system, the keypad, or at least one of the keys of the keypad, into a different mode. According to one embodiment of the invention, if the back and forth gliding action provided on/from a key is provided towards a symbol having derivatives, then, the system (e.g. said key) may be entered/switched into the instance corresponding to the derivatives of said symbol. On the other hand, if said symbol does not have a derivative, then the system (e.g. said key) may be entered/switched into another predefined instance. As an example, during the insertion or correction procedures for the entry of a precise character, if a user provides on an appropriate key a back-and-forth-gliding action towards a letter (e.g. "e") having derivative accented characters (e.g., "β", "β"), then said key may be entered/s witched into the accented symbols instance, but if the user provides a back-and-forth-gliding action on/from a key towards a letter that does not have derivative characters (e.g. the letter "b"), then the system (e.g. said key) may be entered/switched into another predefined mode instance.

As another example, during the entry of a precise character, if a user provides a back- and-forth gliding action on/from a key towards a letter (e.g. "S") having one or more derivative characters (e.g., "s'"), then, the second keypad (model) of the invention may enter/switch into the derivatives mode instance of said letter. Preferably, the apostrophe may be assigned to the center key/zone of the mentioned second keypad in the derivative instance.

As another example, several (e.g. two) groups of letters may be assigned to different instances of the same key of the first keypad. For example, if the key includes Hindi characters, the letters that are more frequently used may be assigned to the key in the first instance, and the letters that are less frequently used may be assigned to said key in the second instance.

According to one method, a back-and-forth gliding action on said key may switch the system between the two instances.

Yet, according to another example, if a key of the first keypad includes several special characters, a first instance may be related to the letters assigned to said key, and a second instance may be related to ambiguous special characters. Examples of assignment of symbols to different instances of a key just described are presented to demonstrate the principles of these embodiments of the system. Obviously, many other symbols may be considered by people skilled in the art and may be assigned to different instances of a key.

It must be noted that the system may include more than one second keypad of the invention, and wherein each of said second keypads may have several instances.

Showing a selection of words

According to one embodiment of the invention, the system may include a means such as a predefined interaction with a key. After the system predicts a word (e.g. a complete word or a stem) based on the input information relating to at least some of ambiguous and/or precise characters of said word provided by the user, the user may scroll within a list of alternative words corresponding to the input information provided until that moment.

According to another embodiment, the system may show at least some of the alternative words, preferably those having the highest priority, that correspond to the input information provided until that moment, wherein preferably each of said words is represented by a key/zone of the second keypad. Preferably, said second keypad may be presented to the user when he/she touches a (touch) sensitive surface of a device such as a touch screen surface. The user then may provide a corresponding interaction such as an appropriate gliding action to select a desired word.

It must be noted that in order to differentiate between an interaction such as a gliding action corresponding to the correction procedure and an interaction corresponding to selecting an alternative word, the following may preferably be taken into account:

According to a first method, each of said interactions may be provided on a different predefined zone on a touch sensitive surface (e.g. on the right side and on the left side of a touch screen, respectively). Fig. 359A shows as an example, a device having a touch sensitive surface 359009, the data entry system of the invention having four letter keys, and uses a

dictionary/database of words wherein a portion of it 359010 is shown. In this example, after pressing the appropriate keys, the system predicts the word "GOES". If the user touches the screen on the right side/zone to provide a gliding action 359011, the system relates said interaction to the correction procedure of the invention and predicts another word "COLE". Now, if the user touches the screen on the left side/zone to provide a gliding action 359111 as shown in fig. 359B, the system relates said interaction to the procedure of selecting an alternative word among the alternative words relating the input information corresponding to the desired word provided by the user until that moment. Said alternative words may be shown on a second keypad 359118 when the user touches the screen on said left side.

According to a second method, a predefined means such as a switching method may be used to switch between the instance of the second keypad corresponding to the correction procedure and the instance corresponding to the alternative words. For example, during the entry of a word, said switching means may be a back-and-forth interaction on the screen outside the keys of the first keypad. As an example, a gliding action not ending at its departing zone may correspond to the correction procedure, and a back-and-forth gliding action (e.g. preferably, in the direction of a letter/symbol that does not have a derivative) may inform the system to show the alternative words to the user (e.g. under his finger) and then the user may provide, for example, a gliding action to select one of the alternative words.

According to a third method, during the entry of a word, a predefined interaction such as a gliding action on/from a key such as the backspace key, may inform/cause the system to show the derivative words to the user. Other methods for the same purpose may be considered by people skilled in the art.

Showing presenting the alternative words may be based on different predefined procedures that involve/consider at least the beginning letters of the predicted word:

According to a first procedure, during the correction procedure, when the system predicts a stem/word/N-gram, if the first character of the predicted word is ambiguous, in addition to said predicted stem/word, the system may also select/use at least some alternative stems/words corresponding to the input information provided until that moment wherein their beginning characters may be different from each other, and preferably show said stems/words on a screen. Preferably, the system may show said alternative words, after the user provides a predefined interaction such as touching a predefined zone of the touch sensitive surface as described. An alternative word may be an entire word and/or a portion of a word (e.g. a stem). Said alternative words may be shown on the keys/zones of an instance of the second keypad of the invention such as a keypad having a three-by-three keys matrix. If there are more words, they may be assigned to a second instance of the second keypad. The user may switch to said second instance by using a switching means.

According to a second procedure, during the correction procedure, when the system predicts at least a portion of a word, if the first character of the predicted word is ambiguous, in addition to said predicted word, the system may also select at least some of the alternative words corresponding to the input information provided until that moment wherein their beginning characters may be the same as the beginning characters of the predicted word and preferably show them on a screen. Preferably, said alternative words are shown on the screen when the user provides a predefined interaction such as touching a predefined zone on the touch sensitive surface. After touching the surface, if the user does not select an alternative word and removes his/her finger from the sensitive surface, according to the first method, the system does nothing, and the status of the first character remains unchanged (e.g. remains ambiguous), while according tothe second method, the system changes the status of the first character from ambiguous to precise.

According to one embodiment, during the correction procedure, when the system predicts at least a portion of a word wherein at least one of its characters such as at least the beginning character(s) is (are) precise, in addition to said prediction the system may also select at least some alternative words corresponding to the input information provided until that moment, and preferably show them on a screen. Obviously, said alternative words

selected/showed by system may include the same precise characters as the predicted word. Preferably, said alternative words are shown to the user when he/she touches the surface as described above.

Using tapping action to scroll between alternative words

According to one embodiment, during the correction procedure, if the first letter of the predicted word is ambiguous and correct, then, the first tapping action on the touch sensitive surface may correspond to changing the status of said correctly predicted letter, from ambiguous to precise, wherein the system generally predicts another word beginning with said precise letter. According to one method, each following tapping action may inform/cause the system to propose another word corresponding to the input information (e.g. during entry and/or correction procedure) provided until that moment. If the user provides one or more precise characters of the predicted word by for example providing corresponding gliding actions, then again each additional tapping action may inform/cause the system to propose another word corresponding to the input information (e.g. during entry and/or correction procedure) provided until that moment.

During/while/when providing the first input information (e.g. key pressing actions) corresponding to the characters of a word, the system may also show at least some of the characters/letters corresponding to each of at least some of the input signals (e.g. key presses) provided by the user, for example, in a column under each letter of the word being predicted (e g. hereafter referred to as "letters column(s)") on a screen. Said letter columns may include or exclude the corresponding character/letter of the proposed word. The elements (e.g. letters, different zones on each letter column to which a character is assigned, etc.) of each column of said letter columns may be responsive to user's interaction and therefore another type of correction procedure according to an embodiment of the invention may be considered, which has been described in detail in the previous patent applications filed by this inventor. The procedure of correction according to an embodiment of the invention may replace or be used as an additional type of correction procedure of the system.

According to one embodiment of the invention, interacting with the zones of one or more letter columns may be provided by using corresponding gliding action(s). This may permit a quick and intuitive correction procedure, enabling the user to enter one or more precise character during the correction procedure of the invention.

Fig. 360A shows as an example, a device 360000 having a touch screen 360009, and including the data entry system of the invention having at least four letter keys 360001-360004. In this example, based on the user's interactions with the appropriate letter keys the system has proposed the word "LAST" in a predefined location 360010. According to one method, if the user touches anywhere (e.g. 360011) outside the keys of the first keypad on the screen, the system may show the letter columns 360008 corresponding to the input information (e.g. key presses) provided by the user, preferably under/close to the user's finger. The predicted word may also be shown on the top of the letter columns under the user's finger. In this example, the first (e.g. ambiguous) letter of the predicted word is selected/highlighted. Now, as shown in fig. 360B, the user may enter the system into the correction procedure by providing a gliding action 360107 towards a desired letter (e.g. "S") in the first column to enter said letter precisely. After reaching the zone corresponding to the letter "S", as an example, if the user removes his finger from the screen, the system may replace the selected/highlighted letter "L" by the precise letter "S" just selected, and may predict a new word which in this example is the word "SHUT"

360111. Now, the system may select/highlight the next (e.g. ambiguous) character (e.g. "H") of the predicted word.

Preferably, while providing the gliding action over the zones corresponding to the letters within the letter columns, the letter corresponding to each zone being interacted may be shown in a highlighted manner 360113 in a visible location.

With continuous description of the current embodiment, if the new predicted word is not what the user wants, he may continue the gliding action towards a letter of the next letter column, in order to replace, for example, a highlighted letter in the predicted word by one of the letters of the corresponding column, and so on. In the current example, as shown in fig. 360C, the user continues his gliding action 360207 towards the letter "A" in the next column. After reaching the zone corresponding to the letter "A", as an example, if the user removes his finger from the screen, the system may replace the selected/highlighted letter Ή" by the precise letter "A" just selected, and may predict a new word which in this example is the word "SALT" 360211. Now, the system may select/highlight the next (e.g. ambiguous) character (e.g. "L") of the predicted word.

The user can repeat the current/said correction procedure to enter precise characters corresponding to the positions of each of the following characters of the desired word until the system proposes the word he/she desires. As an example, as shown in fig. 360D, the user may continue said procedure by providing the gliding action 360307 which relates to entering the precise letters in the third and forth character location of the desired word, correspondingly. In this case, it is the word "SAGT" 360311 that, in this example, is not included in the word database.

It must be noted, that in order not to confuse the system during the gliding actions over the letter zones while traversing from one column to another (the next) column, preferably, after selecting a letter in a zone, the user may provide an accented gliding movement towards the next column. This matter is known by people skilled in the art. Fig. 360E shows as an example, an accented gliding action 360407 provided by the user with respect to the same word "SAGT".

According to one embodiment of the invention, when the user touches the screen for entering into the correction procedure, the touching impact/point on the screen may correspond to the precise letter of the first ambiguous character position of a word having the next highest priority among the words relating to the input information provided by the user. Based on that, the system may locate the letter columns relating to the predicted word in an appropriate location under the user's finger. Fig. 360F shows an example, wherein the system initially predicted the word "LAST" (not shown), and wherein when the user touches the location 360514 to enter into the correction procedure, the system relates said location to the zone corresponding to the letter "E" which in this example is the first character of the word "EASY" which has the next highest priority among the words relating to the input information provided by the user. Now, as an example, if the user removes his finger from the screen, the system predicts the word "EASY".

It must be noted, that during interacting with the letter columns, the use may also go back in the previous letter columns to correct a letter of a previous column. According to one method, if the user desires to interact with a specific letter column, he may glide from outside the letter columns within said letter column.

According to one method, if the system provides a gliding action on one or more characters of the predicted word, then said characters may be considered as precise characters.

According to one embodiment, if the user desires to select one or more letter columns to interact with, then he may touch the corresponding letter(s) of the predicted word, and the system shows/provides the corresponding letter column(s).

According to another embodiment, the touching point may be corresponded/related to the first letter of the predicted word.

If the letter column has many columns, then according to one method, the system shows few columns after (e.g. and few columns before) the current letter column being interacted.

According to one method, during providing a compound gliding action (as described herein) on the letter columns may be provided in several sessions, wherein each session may correspond to one or more precise characters. After providing a first portion of a compound gliding action, if the user removes his finger from the screen and touches the screen again, then, the system relates said touching action to the next, preferably ambiguous, column of the letter columns.

According to one embodiment of the invention, during/while providing a gliding action on the letter columns, if said gliding action does not precisely correspond to all of the desired characters (e.g. for example, a portion of the gliding action may ambiguously relate to any of the two adjacent letters, then the system may propose several corresponding words to the user so that the user for example selects one of them.

It must be noted that the correction procedure as described above can replace or be an alternative to the other correction procedures described before, and may be used with any of the embodiments described throughout this application and other patent applications filed before.

Note that for better memorization, the letters on the letter keys 360001-360004 of fig. 360E, are distributed in an alphabetical order.

As mentioned before, a first type of interaction with a letter key such as a tapping action on a letter key may correspond to an ambiguous character on said key and a second type of interaction with a letter key such as a gliding action on/from a letter key may correspond to an appropriate precise character on said key. According to one embodiment of the invention, during the entry of a word, if the predicted/current word includes at least one ambiguous character, then a gliding action provided on/from a key may correspond to the correction procedure of the invention (e.g. replacing an ambiguous character of the predicted word by the selected precise character through the gliding action, and eventually predicting a new word). If the current word does not include an ambiguous character, the gliding action provided on/from a key may correspond to the insertion procedure of the invention (e.g. adding a precise character for example at the end of the current word, at the cursor position, etc.). As an example, fig. 361A shows a device similar to that of fig. 360 A, including the keypad of the invention having the letter keys 361001-361004, a backspace key 361005, and the space key 361006. Fig. 361A also shows a portion 361010 of a database of words used by the system. In this example, by tapping on the keys 361001, and 361003, the system predicts the word "TO" 361008. All of the letters of the predicted word are ambiguous. According to this embodiment, if the user provides a gliding action 361007 on/from the key 361001 towards the letter "F", then the system may replace the highlighted (e.g. first) letter "T" of the predicted word by the letter "F", and may predict a new word if available. In this example, as shown in fig. 361B, said new word is the word "FR" 361108. Now, the second ambiguous character of the new predicted word is highlighted. In order to replace said ambiguous character by a precise character, the user then, may provide a gliding action, in this example 361107, on/from the key 361003 corresponding to selecting an appropriate letter (e.g. "D") of the key 361003 of the keypad. At this time, preferably when the user removes his finger from the screen, the system replaces the highlighted character of the predicted word 361108 by the letter "D" just selected, and provides the word "FD", as shown in fig. 361C. Now, all of the letters of the proposed word 361208 are precise. If the user provides an additional gliding action such as 361207 on/from a key 361001, then, the selected character (e.g. "J") will be added to the current word, to provide a new word 361308, as shown in fig. 361D. All of the letters of the new word 361308 are precise. Providing additional gliding actions on/from the letter keys will add additional precise character to the current word.

Fig. 362A, shows another example of the current embodiment, wherein the user begins entering a word by providing a gliding action 362007 on/from a key 362001. In this example, because there is no current word before that gliding action, the system considers said precise character as the first character of the current word to be entered and preferably, after the users removes his finger from the screen 362009, the system shows the current word 362008. Then, as shown in fig. 362B, if the user provides another gliding action 362107 on/from a key 362003, then, (e.g. preferably when he removes his finger from the screen) the system adds the corresponding precise letter/character to the current word, to provide a new current word 362108. On the other hand if the user provides a tapping action on a letter key, the system may add an ambiguous character to the current word, and generally predicting a new word. In this case according to one method, the next gliding action on/from a key relating to a precise character, may replace the ambiguous character of the new word by said precise character.

The correction procedures through/relating-to gliding actions provided on/from the letter keys of the system as just described, may be in addition-to or in replacement-of the correction procedures through gliding/tapping actions provided outside the letter keys during the entry of a word.

By replacing the correction procedures through gliding/tapping actions provided outside the letter keys during the entry of a word by procedures through gliding actions provided on/from the letter keys, interactions on the screen/touch-sensitive surface can correspond to mouse functions, even during the entry of a word.

Switching the letter keys into Precise Characters Mode Instance

According to one embodiment of the invention, a switching means for example in form of a predefined interaction with a key may be included within the system such that after providing said interaction, the letter keys of the keypad may enter into a mode wherein if the user touches a letter (e.g. a zone assigned to a letter) on a letter key, said letter may be inserted precisely (e.g. hereafter said mode is called/referred to as "Precise (Character) Mode instance"). This method of entering precise characters may be used with the embodiments of the word predictive system of the invention including those based on combination of ambiguous and precise characters. Preferably, during the Precise Character Mode Instance, the letter keys may be enlarged so that it is/becomes easier to access a letter zone. The/a (letter/special character) key in the Precise Mode instance may be considered as a second keypad of the invention. This is because in the Predictive Mode, a tapping action anywhere on a key may ambiguously correspond to any of a plurality of characters assigned to a key, and in the Precise Mode, a tapping action on a specific zone on said key may correspond to a specific character assigned to said zone. Therefore, a plurality of said zones on said key may be considered to form a corresponding second keypad of the invention replacing said letter key during the Precise Mode instance. In the Precise Mode, the borders between the zones/keys of the second keypad may be marked by lines as shown in Fig. 367.

Switching Means

According to one embodiment of the invention, the switching means to enter the letter keys into the Precise Character Mode Instance may be a gliding action on a key such as the space key or the backspace key. Fig. 363 shows as an example, a/the Space/Function key 363006 of the system in enlarged position. As shown as an example, providing a gliding action 363011 towards the upper-right side on/from said key may enter the letter keys into the Precise Character Mode Instance. At this time the following may occur:

According to one method, if the user removes his finger which provided the gliding action, from the screen/sensitive surface, after entering an appropriate single precise character by tapping on a letter zone on a letter key (e.g. on a corresponding second keypad), the system may automatically exit from the Precise Character Mode Instance, and preferably enter the letter keys back into their original (e.g. ambiguous) mode instance.

According to one method, if the user does not remove his finger which provided the gliding action, from the screen/sensitive surface, then, the user may be able/permitted to enter multiple precise characters/letters until he removes his finger which provided the gliding action from the screen.

According to one embodiment of the invention, another type of interaction such as a back-and-forth gliding action (e.g. 363012) with a finger, with/on a key followed by removal of the finger from the screen, may enter the system into the Precise Character Mode Instance. In this case, all of the letters entered by providing tapping actions on the letter zones on the letter keys in their precise letter mode (e.g. as mentioned, a letter key in its precise mode is represented by a corresponding second keypad of the invention) may be entered precisely, until the user provides another interaction, such as the same back and forth gliding action again (e.g. 363012), to enter the letter keys back into their original (e.g. ambiguous) mode instance.

It must be noted that other switching means for switching between different modes may be considered by people skilled in the art. For example, if the device using the data entry system of the invention includes a movement sensor, for switching (e.g. back-and-forth) between two modes, the user may provide a predefined movement such as shaking the device.

Correction Switch / Insertion Switch

The examples of assigning a predefined interaction to, a symbol such as a character or function, to a mode instance, etc, as described throughout this application is only used to describe different embodiments of the invention. Obviously, other types of interaction, and other mode instances, etc. may be considered and implemented by people skilled in the art. For example, according to one embodiment, a first type/kind of interaction with a key may enter the letter keys into a Precise Character Mode Instance for Correction, and a second type of interaction with a key may enter the letter keys into a Precise Character Mode Instance for Insertion, wherein in the Precise Character Mode Instance for Correction, the precise characters entered by using the letter keys replaces ambiguous characters of a predicted word, and in the Precise Character Mode Instance for Insertion, the precise characters entered by using the letter keys will preferably be entered at the cursor position (e.g. in the word being entered).

In the example of fig. 363, a gliding action 363013 towards the right side from/on the key 363006 may correspond to confirming a unique word relating to the input information relating to the beginning characters of said word.

Precise Character Mode - method 1

With continuous description of current embodiment, according to a first method, entering precise characters by using the letter keys during the Precise Character Mode Instance, may be applied to both the Correction and the Insertion procedures of the invention according to the principles of entering precise characters during the entry (e.g. correction, insertion) of a word by providing gliding actions on/from the letter keys of the system as explained throughout the embodiments just described above and shown throughout figures 361 A through 362B (e.g. by replacing gliding actions provided from the letter keys).

As an example, according to one embodiment, during the entry of a word, if the predicted portion/word includes ambiguous characters, and the user switches the (e.g. letter) keys into Precise Character Mode Instance, providing a precise character by tapping on a letter (e.g. letter-zone) on a letter key (e.g. on a key/zone of the corresponding second keypad), may replace an ambiguous character of the predicted portion/word, and, eventually, the system may propose/predict another word/portion. On the other hand, if the predicted portion/word does not include an ambiguous character, the said precise character may be inserted (e.g. added) at the cursor position within said word/portion.

As an example, fig. 364 shows the letter keys 364001-364004 of the invention after the system has entered into the Precise Character/Letter Mode Instance. In this example, said letter keys are being enlarged, as described above. As shown, providing a tapping action on a letter zone 364007 on key 364001 of the keypad may enter said letter precisely. In this example, because the letter being entered is the first character of a word being entered, said letter is added/inserted without replacing any character in the Current word (zone) 364008 shown on the screen 364009. As an example, by providing an end-of-the-word symbol such as a space character, the user may accept said letter/character (e.g. as a word to be entered in the text/document).

Precise Character - method 2

According to another embodiment of the invention, entering precise letters by taping on their zones on their corresponding keys during the Precise Character Mode Instance, may cause the system to insert (e.g. add) said precise letters at the cursor position. Said cursor position may be the cursor position in a word being entered, even if said word includes ambiguous characters (e.g. before and/or after) said cursor position. According to this method, entering precise characters using the letter keys during the Precise Character Mode Instance, preferably may be applied to the Insertion principles of the invention, wherein a precise character is generally entered at the cursor position. Said cursor position may be within a word being entered, wherein said word may include ambiguous characters. According to this method, the precise characters being entered through Precise Character Mode Instance may not replace the ambiguous characters.

Other methods

According to one embodiment/method of the invention, the gliding action provided on/from a letter key may correspond to inserting (e.g. adding) a precise character at the cursor position within a word being entered even if said word has an ambiguous character, and proving a precise character using a letter key during the Precise Character Mode Instance may correspond to correcting a highlighted/selected ambiguous character of the word being entered if said word being entered includes an ambiguous character.

According to another embodiment/method, during the entry of a word, when entering a precise letter/character (e.g. correcting/replacing or inserting/adding) through the Precise Character Mode Instance (e.g. by tapping on the corresponding letter/letter-zone on a letter key):

- if one of the letters/characters of the (e.g. current) predicted word is

selected/highlighted, said precise letter/character may replace said

selected/highlighted letter/character of the predicted word, and the system eventually may predict (e.g. if any) another word including said letter/character. This procedure may be repeated until there is no ambiguous character within a predicted word.

- if the (e.g. current) predicted word does not include a selected/highlighted

letter/character (e.g. if all of the letters/characters of the predicted word are precise), said precise letter/character preferably is inserted at the cursor position (e.g. at the end of the current word. According to one embodiment, at any moment during the entry of a word, if the input information provided by the user does not correspond to a word of the word database used by the system, then, the system may automatically enter into the Precise Character Mode Instance. The user then, may use the letter keys in said mode to enter precise character(s) (e.g. by tapping on the corresponding letter/letter-zone on a letter key). As a first example, during the entry of a word, if the user provides ambiguous key presses, several scenarios may be considered such as:

1- If at some point, the system does not find a word corresponding to said key presses, then the system may automatically enter into the Precise Character Mode Instance, and the user may use the letter keys in said mode to enter precise characters replacing the predicted characters. After replacing said predicted characters, if the user continues to enter precise characters, they will be Inserted at the cursor position (e.g. they will be added at the end of replaces characters)

2- After providing all of the ambiguous key presses corresponding to his desired word, if the system predicts a none desired word, and the user may perform a Correction procedure to for example correct at least some of the characters of the predicted word, and at some point the system does not find a word corresponding to the combined input information provided by the user, then the system may automatically enter into the Precise Character Mode Instance, and the user may use the letter keys in said mode to enter precise characters replacing the predicted (e.g. ambiguous) characters of the predicted word. After replacing said predicted characters, if the user continues to enter precise characters, they will be Inserted at the cursor position (e.g. they will be added at the end of replaced characters).

Correcting a mistyped letter

According to one embodiment of the invention, during the entry of a word, a character (e.g. ambiguous and/or precise letter) of said word may be mistyped (e.g. the user may press a wrong letter key). In this case, when said mistyped character is highlighted/selected, the precise character/letter provided during a correction procedure using the letter keys as described above (e.g. using gliding action on/from a letter key in a normal (e.g. ambiguous) mode, or a tapping action on a letter zone on a letter key when said key is in Precise Character Mode Instance) may replace said mistyped character, even if said precise character is provided/entered by

using/through a letter key other than the letter key to which said mistyped character belongs or through which it has been entered.

According to one embodiment of the invention, during the Precise Character Mode instance, a tapping on the center zone of a letter key may keep a selected/highlighted character of a predicted word, as is. After providing said tapping action, the system may select/highlight another (e.g. ambiguous, or precise) character of the word being entered.

According to one embodiment of the invention, when the system is in a/the Precise Character Mode Instance, a gliding action on/from a key towards a predefined direction may correspond to Correcting an ambiguous character of a word being entered regardless of the cursor position, and a tapping action on a letter zone on a letter key of the system may correspond to

inserting/adding an appropriate precise character in the cursor position (e.g. within/at-the-end of the word being entered), or vise versa.

According to one embodiment of the invention, if a character (e.g. letter) of a word is manually selected by the user (e.g. and highlighted) by for example using a In-Word selecting means (e.g. a character selecting means to select a character of the current word to, for example, correct it. A correction procedure generally applies to a/the selected character. This matter has been described in different patent applications filed by this inventor) or by tapping on said character within a word (e.g. being entered), then providing a precise character through a letter key may replace the selected character by the said precise character, even if said precise character and said selected character belong to different keys.

According to one embodiment of the invention, a predefined interaction such as a back- and-forth gliding action on/from a letter key relating to a letter on said key may insert or place said letter at the cursor position (e.g. within the current word, in the document, etc.).

Switching between word databases

According to one embodiment of the invention, a switching means for example in form of a predefined interaction with a key may be included within the system such that after providing said interaction, the system switches from using a first (N-gram) database to using another predefined (N-Gram) word database. For example, the user may enter multilingual texts by switching between different word databases in different language. According to another example, a first word database may be a common (e.g. a general purpose) word database, and a second word database may be a specific database such as a professional vocabulary/word database such as a medical vocabulary/database. As an example, a doctor may enter a report wherein if during his report he uses medical terms, he may switch the system from using a general word database to using a medical word database on the fly.

According to one embodiment of the invention, a single application may use several different word databases. For example, if the application relates to GPS, then, the system may use a database of words including the names of the countries, a database of words including the names of cities in one or more specific country, a database of words including the names of the roads/streets in a specific city, and so on. The different database of words may be combined in a single database or they may be separate databases.

In the case of a GPS dictionary, an entry such as a name of a street may be or include a compound word. A compound word having connectors between its words is preferably considered (by the system and/or user) as a 1-gram word.

According to one method, the entries of a word database used by the system may include compound words. A compound word is a sequence of words connected to each other by a connector such as special character, a value/code or even a space character, and wherein (e.g. during search) the system and/or the user considers said compound word as a single word.

An entry of an N-gram database of words may include at least one compound word. According to a first embodiment, each of the entries of a specific database used by the system such as the database of streets may be constituted of a single/compound word. Said word may be an independent word (e.g. "Royal") or a compound word (e.g. "Royal Street"). The system may include a means such as a key interaction to be able to enter a connector. While providing the input information corresponding to a compound word, the user may enter said connector between the input information corresponding to individual words of compound words.

According to a first method, a pressing action on the Space key of the system may preferably be used by the user to accept/confirm a predicted word (e.g. a single word, or a compound word). According to a second method, a pressing action on the space key may be considered as providing a predefined connector (character) between individual words of a compound word, and another predefined interaction may be used for confirming a word predicted by the system.

According to another embodiment of the invention, a specific database such as the names of the streets may be organized as having N-gram words. According to one method, the entries of such a database may preferably be constituted of 1-gram words and/or N-gram preferably 2-gram words. The 1-gram words entries and the N-gram (N>1) entries may either constitute a single database or they may constitute separate databases. For ease of search and ease of interaction a content such as the name of a street may be divided into one or more N- gram words each assigned to a different entry of the database(s). As an example, if the database is related to title of songs, the title "sympathy for the devil" may be divided into "sympathy", "sympathy for", "the devil", and "devil", so as to permit a content search that may be based on partial information corresponding to a desired content. It must be noted that according to one method, the words of each of the entries just described may also be considered as a single compound word.

According to one method, during a content search, if the input information provided by the user corresponds to more than one entry,or word the user may either proceed to a correction procedure of the invention, or the system may propose said entries (e.g. all at once, several at a time, or one by one, etc.) to the user so that he selects one of them.

According to one embodiment of the invention, when using an application such as a GPS application wherein entering out-of-dictionary words (i.e. the words that do not exists in the database used by the system) are not permitted, at any moment during the entry of a word such as the name of a street, if the input information including one or more input signals provided by the user does not correspond to any entry of the corresponding/said database, then the system may ignore the last input signal provided by the user and may inform him by an alerting means such as a beep.

Using professional/specific/custom dictionaries/databases may contribute to entering professional/specific terms with high accuracy because such databases may generally have a limited number of entries. For example, a specific database such as a city street names database may be very limited compared to a common/general purpose database.

Depending on the design of a content search system, different types of arrangement of the entries of databases may be considered and used by the appropriate searching method. For example, according to a first method, the names of the streets may be arranged as single compound (e.g.l-gram) words (e.g. including connectors) to form a 1-gram database. According to a second method, the names of the streets may be arranged to constitute N-gram words, wherein N>0.

The data entry system of the invention may be adapted for use with different (e.g.

vertical) applications wherein usually use of database of words having a limited number of word which are generally dedicated to said applications. Those databases usually include single and/or combined word. In many cases, those words are long words. This structure of database permits the data entry of the invention to have extremely high rate of word prediction.

A page (e.g. screen layout) of an application such as a vertical application may include more than one field, wherein each field may include/related to a different word database. For example, a GPS application may use different word databases such as the a database of names of countries, a database of names of cities of all of the countries, different databases each corresponding to a single country, etc. According to one method, those databases may be downloaded/used dynamically, based on user's demand or by the system. For example, according to one method, if the user enters the name of a country, then, the system may use a sub-dictionary corresponding to the cities of that country. According to another method, the dictionary corresponding to the names of the cities is used based on the location of the GPS device in a corresponding country. According to another example, an order sheet of an ecommerce garment company, may include several fields such as the name of the article, its color, its size, credit card company name, etc., wherein at least some of said field may be related to a different database of words such as the name of articles, the colors, etc. Some fields such as numeric fields may not be related to a dictionary. These matters are known by people skilled in the art.

As mentioned before, one of the characters assigned to letter keys of the system may be a special character to indicate the space character between the words of a combined word. Said character may be a space character or it maybe another character such as a connecting character replacing the use of space character between the individual words of a compound word.

Alternatively, said special space character/connector may be assigned to a key other than the letter keys, for example to the space key of the system for preferably being entered precisely. As an example, a gliding action in a predefined direction on/from said key may be assigned to entering said characters precisely. Providing the connecting character precisely, will help the system to more accurately predict a word. The prediction of a desired word by the system may be possible before the user enters all of the input information corresponding to a desired combined word.

According to one embodiment, when using a dedicated word database, the user can enter a normal space character between words of a combined word and the system may consider said space character(s) as part of a combined word being entered. For example, during the entry of a combined word "Winston street", the user may enter a normal space character between the input information corresponding to the words "Winston" and "street", and the system may preferably considers said input information including said space character as relating to a combined word of the dictionary, and predicts a word.

Content search such as GPS - tap to accept

According to one embodiment of the invention, providing/performing a gliding action in four different directions provided anywhere on a touch sensitive surface may correspond to interacting with four letter keys of the system. Preferably, the directions may be towards any of the following: upper left, upper right, lower left, and lower right. Furthermore, gliding actions in other directions such as left, right, up, and down, may correspond to other symbols such as special characters, and functions. For example, a gliding action leftward may correspond to backspace function, and a gliding action rightward may correspond to providing a space character. Also, according to one method, a tapping action on a sensitive surface may correspond to another symbol such as a confirmation function.

The directed gliding actions such as those just described above may be used to enter data in devices such as those mainly using a sensitive surface to input data. One of those devices is the Global Positioning System ("GPS") device. According to one embodiment of the invention, as shown as an example in fig. 365 A, a GPS device having a touch sensitive surface 365019 to enter data, and a separate screen 355009 to output data, may include the data entry system of the invention. Gliding actions provided anywhere on the touch sensitive surface 365019 may imitate/duplicate gliding actions provided from the center of an imaginary keypad model 365000 towards a corresponding key of said keypad model 365000. For example, the gliding action 365013 provided on the touch sensitive surface 365009, imitates the gliding action 365023 corresponding to interacting with the key 365003 of the imaginary keypad 365009. As shown, by using the touch sensitive surface, according to one method, if the user provides the gliding actions, 365014, 365014, 365013, 365013, 365014, 365014, 365012 (for

space/connector), 365013, 365013, 365012, 365013, the system may propose the corresponding street name comprising the combined word "George Road".

According to one method, the system may include an N-Gram (e.g. word) completion system so that, during the entry of an N-Gram (e.g. single word, combined word) of the relevant N-Gram database, if there is only one corresponding word (e.g. a longer word), wherein this word corresponds to the user's interaction(s) with keys, and is other than the word or chain of characters being entered, then the system may also propose the longer word.

According to another method, during the entry of a word using a (e.g.an N-Gram) database, if there are two or more corresponding words (longer words), wherein these words correspond to the input information provided by for example user's interaction(s) with keys and differ from the word or chain of characters being entered, wherein these words may preferably be assigned different frequency values, the N-Gram (e.g. word) completion system may also propose the longer word of the highest frequency. Said method may preferably be applied to all the embodiments of the invention. According to one embodiment of the invention, a tapping action may be assigned to confirming a proposed/predicted word. Also as an example, a gliding action provided from anywhere in a direction such as upward may relate to a previous page/field, and a gliding action provided from anywhere in a direction such as downward may relate to a next page/field, or vice versus.

According to one embodiment of the invention, the dynamic database used by the system may be based on the context of a text/document at a relevant instance.

According to one embodiment, the keyboard/keypad mode instance may automatically change according to a different context. For example, if a field requires entry of letters, the

keyboard/keypad may be in letter mode (e.g. the mode wherein interacting with the four letter keys generally correspond to letters), and if a field requires entry of digits and does not require entry of letters, then the keyboard/keypad may automatically enter into the special characters mode such as at least one of the letter keys enters into the digit mode, and preferably, also other letter keys represent different group of special symbols as described throughout this application. In the example of fig. 365A, according to one method, if the user is asked to enter a house number, then the keypad 365019 may be defined such that gliding actions in different directions on it correspond to entering numbers based on a telephone keypad keys as described earlier in this application. In this case, a different interaction such as a long/double pressing action may be related to confirming the number entered.

In the current example, if the input information provided by the user may correspond to more than one words/candidates of the database, then, the system may propose them to the user by means such as using a text to speech means/software to pronounce them, or showing them on a separate screen 365009 of the invention as described before. In this case, the user may provide an appropriate interaction such as pronouncing a candidate word he/she has selected, whilst a speech to text means/software is being used, or, for example, performing an appropriate gliding action to enter/select one of them. According to one method, when candidates are proposed to the user, a gliding action provided on the screen (e.g. preferably towards a candidate) may be related to selecting a candidate. According to one method, the system may show them on a three by three matrix based presentation as described earlier. Said selecting interaction/gliding action may be provided on any of the sensitive surfaces 365019 and/or 365009. According to one method, instead of a single gliding action, a back-and-forth gliding action in the direction of the desired candidate may be provided to select one of the candidates proposed to the user. According to one embodiment of the invention, the system may be designed in a way that the user does not have to enter all the elements (e.g. words) of an N-Gram such as a name of a street (e.g. when he/she is looking for a map, is using the GPS, is searching for a movie, is using any other content search application, etc.), In such a case, the system may use information corresponding to a portion of one or more entries of the database to select/consider them.

A gliding action provided from anywhere upwards (e.g. 365017) and a gliding action provided from anywhere downwards (e.g. 365018), may, for example, be assigned to functions such as going to the next field, or going to other pages such as a homepage, a GPS directory, etc. As an instance, when the user begins to enter an address, the system may first require from a user to enter a country or a state name by, for example, locating a cursor/focus on a field (on a screen) corresponding to the names of countries. After entering and confirming it, by, for example, providing a tapping action on a surface, the user may provide a gliding action 365018 and, as a result, the system may focus on the next field corresponding to entering a city name. According to one method, the gliding action downwards may confirm the country name being entered and move/skip the focus to the city field. In this case the tapping action for confirmation may not be required. The gliding actions downwards and upwards may navigate the focus of the system between the fields back and forth.

According to one embodiment, multiple gliding actions in a predefined direction such as a gliding action downwards may navigate the focus of the system on different fields in a loop manner. In this case, the gliding action in another direction (e.g., 365017 may be used for other functions such as entering the system into its homepage, etc.

In a case when the system is focused on a field designed to receive digits, a predefined interaction such as a long pressing action may be used to confirm the digits entered and eventually exit (e.g. remove the focus) from that field.

According to one embodiment of the invention, in some cases such as if an application field requires the entry of ambiguous characters/letters ( which may imply using the word predictive mode of the system) alongside with precise characters, then the interactions of the first type (such as quick gliding actions) may correspond to entering the ambiguous characters, and interactions of the second type (such as slow gliding actions) may correspond to entering the precise characters. A slow gliding action has been described before by this inventor. For example, a slow gliding action may be a slow motion during the entire instance of the gliding action or during a portion of it such as in the beginning or in the end of said gliding action. According to another method, the interaction of the second type may be another predefined interaction such as a back-and-forth (or, for example, forth-and-back) gliding action

corresponding to a character such as a digit assigned to a predefined second keypad model (e.g. a telephone-type keypad) such as those described before., According to other methods, said interactions of the first and the second type may be, correspondingly, short and long gliding actions, soft and hard gliding actions provided on a surface sensitive to different levels of pressure applied to it, etc.

In the example shown in 365 A, the GPS screen may be the screen of a first mobile device such as a smartphone, and the separate pad for entering data may be the screen of the second mobile device such as a smartphone. A wireless/wired communication means may be applied between the two devices. The (e.g. processing of) data entry system of the invention may be executed either in any one of the devices or in both of them. The same applies for any number of mobile devices used in accordance with this embodiment. With continuous embodiment of the current example, the input means and the output means may be the same means such as a touch screen of a device (such as a GPS device or a mobile communication device 365119 as shown in fig. 365B). According to one embodiment, both the input and the output means may be integrally associated with one device. According to one embodiment, the system may include at least two input surfaces/pads. As an example, fig. 365C shows a steering wheel 365200 of an automobile wherein two input surfaces 365201 and 365202 are provided on it. As an example, providing interactions such as gliding actions on the surface 365201 may correspond to entering numbers, and providing interactions such as gliding actions on the surface 365202 may correspond to other functions such as entering letters, space, backspace, and up/down arrow functions as described for figs 365 A-B. This assignment may be changed/reversed. The term steering wheel may apply to any navigational means of any transportation means. The split surfaces used with the data entry system described above may be either integrated within the steering wheel or it may be a detachable/attachable unit attached-to/detached-from the steering wheel. The split surfaces may be wirelessly or by wired connected to the rest of the system.

Two pads integrated or add-on

According to one embodiment, as shown in fig. 365D, the system may include two input surfaces/pads wherein tapping actions on predefined zones on them represent interacting with the keys 365301-365306 of the split keypad of the invention as described and shown earlier. On the other hand, gliding actions on one of the surfaces may correspond to a certain function such as a correction procedure of the invention and/or selecting a candidate, and gliding actions provided on the other surface may correspond to another function such as entering digits and/or selecting a candidate.

According to one method, a connector character may be entered precisely during the input of information corresponding to an (e.g. N-Gram) enrty of a database. The connector character may preferably be assigned to another key such as the space key, or a key other than the keys of the first keypad. In this case, the ambiguous connecting character represented by the letter key (e.g. 365302) may be removed from said key. As mentioned before, a precise input of a connecting character may significantly augment the prediction accuracy, so that the system often may predict the desired content before the completion of entering the entire input information, hence saving (reducing the number of) interactions.

Note that the methods of content search described throughout this application are not applied only to the contents mentioned: they may be applied to any type of content. For example, the methods of data entry in the GPS can be applied for another type of content search such as searching songs, videos, movies, filling the fields of an e-shopping page, etc.

According to one embodiment of the invention, instead of (or in addition to) displaying the text/word being entered by the user, the system may use a text to speech software to pronounce the text/word being entered by the user. By combining this method with using the gliding actions for entering text a completely eye-free text entry solution may be provided.

The data entry system just described, may be used for any vertical application wherein preferably the corresponding word database has a limited number of word. One example of those applications may be such as an SMS application using mainly SMS customized word and/or abbreviations. Another example may be a data/text entry application for secret agents or army personals using customized or even coded words. It must be noted that the ambiguous keys of the keypad of the system may include any character such as letters and/or special characters.

Text entry using gestures

According to one embodiment of the invention, the data entry system with which the user interacts mainly by gestures as just described and shown, may be used for general data/text entry. For example, after the system proposes a word based on the user's gliding actions corresponding to the letter keys, a back-and-forth gliding action in a predefined direction may correspond to a precise letter assigned-to/located-on the key (press) that provided the first ambiguous character of the predicted word. The system may propose another word

corresponding to the combined input information, and so on. Accordingly, if a user desires to enter a special character, he/she may first provide a gliding action corresponding to a letter key to which his/her desired special character is assigned when the system is in special character mode. At the end of the gliding action, the user may pause for at least a predefined lap of time. Said pausing action may enter the system into the special character mode. Now, according to one method, the user may provide a second gliding action towards his/her desired special character. Said second gliding action may be provided without removing the finger at the end of the first gliding action or after removing the finger at the end of the first gliding action. In both cases, preferably, the corresponding key in the special character mode may be shown under the user's finger as described throughout this application. According to another method, the second gliding action provided after the first gliding action (and pausing) without removing the finger at the end of the first gliding action may correspond to a function such as selecting a precise character (e.g. letter), whereas the second gliding action provided after the first gliding action (and pausing) after removing the finger at the end of the first gliding action may correspond to another function such as selecting a special character.

According to another embodiment, a predefined interaction such as a gliding action provided from the outside/edge of the screen inwards (on the screen), preferably, in a predefined direction may correspond to entering the system into a mode instance suc¾ as the correction procedure. Same interaction in another direction may correspond to entering the system into another mode instance such as the special character mode, etc. A gliding action upward, or downward may correspond

According to one embodiment of the invention, tapping actions on different letter zones such as four letter zones on a sensitive surface may correspond to entering (e.g. ambiguous) letters/N-Grams, whereas gliding actions on that surface may correspond to other functions such as correction procedure of the invention based on gliding actions as described throughout this application.

As an example, after the user provides tapping actions corresponding to a word being predicted by the system, the user may provide an appropriate gliding action relating to a key of the second keypad (e.g. based on a keypad model) of the invention to enter precisely one or more characters of the desired word, as described before, and the system may propose/predict other word(s), accordingly (e.g. preferably, a key of the first keypad, and a key of the second keypad by having at most one common character as described before, help the system to precisely recognize a common character on those two keys).

Fig. 366 shows as an example, a device having a touch sensitive surface virtually or physically divided into four zones 366001-366004, wherein each zone corresponds to one of the letter keys of the system as described before. According to one embodiment, a tapping action on a zone (e.g. 366002) corresponds to entering an ambiguous letter of that zone, and a gliding action (e.g. 366023, 366028) provided in predefined direction on said surface may correspond to interacting with one of the keys (e.g. respectively, 366013, 366018) of an imaginary second keypad (e.g. 366009) of the invention, for example, for correcting a letter of a word being proposed. As an example, such a keypad may be used for many eye-free applications such as a GPS application. Content search

The data entry system of the invention may be used for searching contents. In many cases the entries in a database such as a content database may include one or more words. The database used by the system may include compound word having at least two words in an entry. The space character between the single words of an entry may be replaced by or considered as, a connecting character. According to one embodiment of the invention, when a user enters a keyword including two or more words into an application to be used by a search engine, the application and/or system may be designed in such a way to consider at least a portion (e.g. including at least two words) of the keyword as a compound word and may provide a search based on said compound word, anywhere within the entries of the database. The data entry system of the invention may be used for entering data in general, and searching content such as ordering a movie/video (e.g. video on demand) procedure. The user may use at least the letter keys to enter a keyword. Because the number of entries of a database of any content are generally limited, and an entry generally includes more than one word, then, generally, the system predicts the right entry mostly without requiring correction procedure, although requiring correction procedure may be possible.

As an example, the method just described may use a remote control of a device such as a television ("TV"), having any type of keys such as soft or hard keys. The user may use a number of letter keys such as the four letter keys of a keypad of the invention (e.g. they can be hard or soft keys wherein letters may be arranged on them as shown in fig. 365 A) to enter the input information corresponding to his desired content such as the title of a movie, title of a song, a portion of the script of the movie/song, etc.. As an example, the keys of a remote control used as the four letter keys of the invention may be the volume-up, volume-down, channel-up, and channel-down. The system may many times find the right entry/movie before the user completes the input of the all the information (e.g. key presses). In some cases where there may be more than one entry corresponding to user's input, according to one method, the system may show at least some of them on for example at least a three-by-three-key keypad as described throughout this application, or as the items of a list wherein preferably the items having sequential numbers, and the user may select one them by, for example, using/pressing a corresponding key of the numeric keypad (e.g. the telephone-keypad layout keys) off/on the remote control. Preferably, the system may show the selection to the user during the entry of content title/name or other related keywords where there are less than 11 candidates corresponding to the user's input.

According to one embodiment, any of the proposed candidates may be selected by using a speech recognition system. In this case, as an example, the user may either articulate the content or he/she may pronounce its number in the proposed list, etc. According to one method, an additional key such as a separate key may be used to enter a connector character preferably precisely during providing the input information corresponding to the keyword. In this case, the ambiguous connecting character represented by a letter key (e.g. the key having letters standing on two points) may be removed from said key. As mentioned before, providing a connecting character precisely may significantly augment the prediction accuracy, and many times the system may predict the desired content before ending to enter the whole input information.

According to one embodiment of the invention, a content word database may have several (e.g. sub) directories based on different categories. This way, the number of entries reduces and the accuracy of prediction augments.

According to one embodiment, the content search may use the correction procedure of the invention. Preferably, the telephone-keypad layout keys of the remote control may be used for such purpose. This matter has been described in detail in different patent applications filed by the inventor. In case that the same the telephone-keypad layout keys is used for both the correction procedure mode and selecting a proposed content mode, then a switching key may be used for switching the between these two modes. According to one embodiment, any of the proposed candidates may be selected by using a speech recognition system. In this case, as an example, the user may either say the content or he may say its number in the proposed list, etc.

It must be noted, that the remote control device may be of any kind, such as that of a TV, or of any other device/system. It may also either be a remote device exclusively dedicated to TV, or a home control remote control according to another embodiment, the remote control, may be in form of a software integrated/used-by a device which its main functionality is other computing such as a mobile phone, a smartphone, a tablet, etc. It must also be noted that the content search using a remote control may not be limited for searching a content but also being applied to any other task such as searching a channel by entering its channel name/appellation (e.g. for example by entering "CNN"). The TV may then change the channel to the one that user has (e.g. just) entered. Enhanced letter arrangement

In order to enhance the user's performance and permit a faster text entry in the Precise Mode Instance, the letters belonging to a letter key may be arranges such that they can be easier to access and less finger-traveling while tapping on them/their-zones. For that, according to one method, the letters that are used more frequently may be arranged such that to be closed (near) to each other. Fig. 367 shows an example of such arrangement. In this example, letters that are more frequently used on the upper keys 367001, 367002, are located on the lower row 367013 on their corresponding keys, and letters that are more frequently used on the lower keys 367003, 367004, are located on the upper row 367014 on their corresponding keys. Based on this principle, some of the other letters which have medium frequency are located on the middle row on their respective keys, and so on.

Two groups on each side

It should be emphasized that in all the embodiments of the invention the functions/modes assigned to each one of the four keys (the main keys) may change correspondingly to the user's actions (such as gliding, tapping, pronouncing words or numbers etc.). If the function/mode of a key changes (e.g. if the system switches to a correction mode, or to a precise characters mode etc.), said key is considered a different key despite the fact that its visual representation may have remained the same or undergone a slight modification only (e.g. a tiny grid may have occurred between the letters).

Thus the keys of the first keypad (i.e. the keypad that corresponds to the predictive mode) and the keys of the second keypad (i.e. the keypad that corresponds to the precise

characters/correction mode) are different keypads, although in some cases they may look (almost) identically to each other.

Hence, it must be noted that in the Precise Mode although a key of the main (e.g. the first) keypad of the invention is shown to be divided into several keys, (e.g. to form one of the second keypads of the invention. As an example, if the first keypad has four letter keys, each of them in the Precise Mode may be considered to form one of the second keypads of the invention as shown in fig. 367) wherein each one of them corresponds to/represents an identified character, it is understood that each of said several keys may be considered as an independent key.

According to this principle, all the keys relating to one side of the device may be considered as independent keys (i.e. not as constituents of the (main) keys on which they are visually placed). of the first keys including them, or not part of a corresponding second keypad).

In reality, there may be considered a keyboard wherein its letter keys are organized in two groups of letter keys, wherein each group of letter keys includes one or more groups of letters (e.g. two groups on each side of the screen of the corresponding device) wherein each group of letters has a common characteristic in their shapes such as those described throughout this application (e.g. standing on one point, standing on two points, standing on a large base, and having a closed circle in their shape). Other principles of letter/character grouping, and accordingly, different groups of letters/characters may be considered by people skilled in the art.

According to one embodiment of the invention, the keys of the first keypad in their Precise Character Mode instance (e.g. the corresponding second keypads of the keys of the first keypad) may cover substantially the whole width of the screen of the corresponding device. As an example, According to another embodiment, those keys may be hard or soft keys. Fig. 368 shows a device 368000 having a touch screen 368009 and an on-screen keyboard 368001as just described, as an example, after the user rejects a proposed word, or when the user enters manually the system in the Precsise Character (Letter/Special Character) Mode, the system may display the a the screen a corresponding second keypad of each of the (letter) keys of the first keypad (e.g. four second keypads in total), wherein, generally each of the keys a second keypad relates to an identified character of the group of letters assigned to its corresponding key of the first keypad. According to one method, said second keypad may constitute a single keyboard. Preferably, the backspace key 368002, and the space key 368003, of this keyboard may have functions resembling to the functions of the backspace key and space key of the first keypad of the invention such as those described and shown throughout this patent application.

Other arrangements may be considered by people skilled in the art. For example, this arrangement may slightly be modified in French language.

Instead of a sensitive surface, the system may include a movement detection means to detect and perceive the movement of the user in the air, for example, based on images received by a camera, or a signals received through a sensor in the user's hand, etc. The user may provides movements to imitate/duplicate any of user's interactionrelating to entering data/text (e.g. by using the data entry system of the invention) provided on a touch sensitive surface, in the air. The perceiving said movements and may provides the corresponding input information to (e.g. processor usd by) the system.

Tap for Special Characters

According to one embodiment of the invention, when the system is switched into the Special Character mode, preferably, the system shows the special characters on the corresponding keys of the first keypad of the invention. A special character may be entered by tapping on said special character or its dedicated zone (e.g. on the corresponding key). Alternatively, the system may show the special characters on a corresponding second keypad of the invention. As mentioned before, the second keypad (e.g. keypad model) of the invention may have a three- by-three-keys matrix. The special characters assigned to a key may be arranged in a three-by- three matrixed arrangement on said key of the first keypad. In case of the digits, according to one method, they may be arranged on a three-by-four matrix-key form factor similar to a telephone keypad. According to another method, the digits 1 to 9 may be managed on a three- by-three-keys matrix on said key and each of them may be entered by a tapping action on said digit. In this case, according to one method, a gliding action downward on the corresponding key of the first keypad may correspond to entering the digit "0".

Derivatives

According to one embodiment of the invention, when entering a precise character or function having derivatives (e.g. through correction, insertion, precise mode instance), the system may show its derivatives on a second keypad of the invention on the screen. The user then, may provide a corresponding interaction such as a tapping /gliding action as described before to enter its derivative. As an example, if the user enters a precise letter having accented derivatives, the system may show its accents or accented characters on a second, and the user may select said accent/accented character. If the user does not provide a required interaction for entering a derivative, and for example, provides another interaction such as continuing pressing on the keys of the first keypad, then the system hides the second keypad and ignores it.

According to one embodiment of the invention, a combined word (e.g. considered as a single word of the database used by the system) may include several words being attached to each other without a connector character or a space character.

According tone embodiment of the invention, the system may include a means such as a slider to enlarge/retract the keys of the system in each mode.

As mentioned before, the user may point to or select one word and use a means of the system such as a predefined gliding action so that the system consider said word as the current word (e.g. for example, for correcting said word, or saving it). Using said mode consequently such as providing multiple such gliding actions may select multiple words before the cursor and consider them as a combined word by the system. The user then may either modify it and/or add it as a single word to the word database.

According to one embodiment of the invention, an auto-correction system relating to the Precise (e.g. Letter) Mode Instance may be implemented with the system. As an example, said auto- correction system may consider neighboring characters of a user's finger impact on a key to provide an extremely accurate prediction during the entry of words through the Precise Mode Instance. This is feasible because the letters on each letter key of the system are preferably are located/arranged based on a three-by-three matrix key keypad, wherein preferably its center key does not include a character.

Common shapes being shown

According to one embodiment of the invention, during the entry of a word, in addition to or instead of the word being predicted/proposed (or even being entered in a precise mode), the system may also show the common shape of each of the letters of said word to the user. By providing such feature, the user can, at any moment during the entry of a word, check to see if he has typed/used the right/appropriate/correct keys while typing. Fig. 369 shows as an example, a device using the data entry system of the invention, through which the user has provided information corresponding to the beginning of a desired word. In this example, the system has predicted the chain of characters "comedy", which are the beginning characters of a word and which may not be the beginning of the desired word of the user. In this example, the keys being pressed may also correspond to the words "lowery", conspiracy", "comedian", "grasping", etc. At this time or point during typing, for example, if the desired word of the user is the word "conspiracy" and the user wants to know if he has typed the right/appropriate/correct keys he may look at the common shapes of the letters for each of the keys being pressed which in this example are shown in location 369007 on the screen 369009. The user may easily understand whether, in this example, he has typed the right/appropriate/correct keys and whether he may continue the typing actions corresponding to the remaining characters of his desired word. On the other had if one of the common shapes does not correspond to an appropriate letter of his desired word, then he may proceed to correcting his mistake. For example, he may erase that character by backspacing 369005 until that character (including it) and retype it (continue to type the appropriate/correct characters of the desired word).

As mentioned before, symbols such as special characters and digits may also ambiguously assigned to the keys of the first keypad. In this example, the special characters, " . " (dot) , and " , " (comma) are also added to the appropriate keys 369001 and 369003, respectively. As an example, the special character " , " may be frequently used during content search in titles such as "Star Trek, The", etc. Preferably, as shown in fig. 372A, the common shapes 372019 corresponding to the key interactions provided by the user during the entry of a word may be shown in large font, and the predicted word letters 372018 may be shown in small font so that the main focus of the user being on the shapes. The common shapes may be printed in the color of their appropriate keys. Instead of a character of a predicted word (e.g. for example for being corrected), its shape 372017 may be selected/highlighted. If the user provides a correction procedure and the system replaces the appropriate ambiguous character by a precise character, then, the corresponding common shape may also be replaced by the precise character. As an example, fig. 372B shows, the word "test" being predicted by the system after the user provided a correction procedure by providing the gliding action 372113 corresponding to the first letter of the originally predicted word "just" (not shown). By replacing the (ambiguous) letter "j" by the (precise) letter "t" the system now proposes the word "test". In addition, the system also here replaced the first common shape by the common shape associated with the precise letter "t" 372116, and selects/highlights this common shape 373117.

Show all words, if available, completion

According to one embodiment of the invention, during typing a word, in addition to the predicted word/stem, the system may present to the user at least some of the other choices of words corresponding to the input information provided by the user until that moment.

According to one method, said other choices may be assigned to / shown on the keys of the second keypad of the invention. Preferably, the system may be designed such that the system may first consider all of the words having the length corresponding to the keys being interacted. If the number of word choices is less than the number of keys of the second keypad or the number of word choices having at least a predefined frequency rate are less than the number of keys of the second keypad, then in addition the system may show some of the words/stems wherein their beginning characters corresponds to the information provided by the user until that moment, for example, based on their priorities (e.g. word completion).

According to one method, the second keypad may have a three-by-three matrix of keys, as described before. According to one method, the word having the next highest priority after the predicted word with the highest priority (which preferably being shown separately as shown throughout this application), may be shown in the center of said second keypad so that a predicted word of the next highest priority may be entered by a tapping action on the screen preferably outside the keys of the first keypad. According to one method, the second keypad (e.g. including the choices) may be shown automatically, while according to another method, it may be shown at user's request by for example, providing a predefined interaction, such as touching anywhere on the screen outside the keys of the first keypad or providing a gliding action from a key such as the space/backspace key.

According to one embodiment of the invention, if the word predicted is not the desired word, the user may touch anywhere on the screen outside the keys of the first keypad and the system may show, preferably under the user's finger, the other choices of words to the user. If one of the words is the desired word, then, the user then may provide a gliding action towards said word. If none of the other choices is the desired word, the user may remove his finger from the screen and the system may automatically enter into the Precise Character Mode. At this time, the user may begin to enter his word precisely by entering its characters one by one. According to one method, the first precise character entered may replace the first ambiguous character, and the system may propose another word, and so on. If one of the words proposed is the desired word, the user may accept it, but if the words proposed are not the desired word, the user may continue entering all of the remaining characters of his desired word, precisely. According to one method, after the user ends entering his desired word, by providing an end-of-the-word signal such as a space character, the system may automatically enter back to

ambiguous/predictive mode.

According to one embodiment, if the user selects one of the proposed words/stems other than the main predicted word/stem, for example by providing an appropriate gliding tapping action, then the system may propose said selected word/stem as a new main proposed word to the user. According to one method, the system considers all of the letters of the new main word as precise characters, and may show words/stems beginning with said main proposed word on the keys of the second keypad. At this time, if for example, the user desired to enter the main proposed word, he may provide an end-of the-word signals such as pressing on the space key. On the other hand, if the user desires one of the words/stems proposed on the new second keypad he may provide a corresponding tapping/gliding action, and so on. Alternatively, if the user does not desire any of the proposed word, he may provide more input information for example by tapping on keys of the first keypad corresponding to at least one of the remaining characters of his desired word.

Fig. 370 shows as an example, a predicted word "goes" being proposed by the system based on the input information provided by pressing actions provided by the user on the appropriate keys of the first keypad. In addition to the main predicted word, the system also shows other words corresponding to said input information on a second keypad of the invention 370008. At this time, if the user provides a gliding action 370011 (anywhere on the screen, preferably outside the keys of the first keypad) the system may select the word "urge" and consider it as the main proposed word 370110 as shown in Fig. 370A. The letters of the new main proposed word "urge" 370110 may be considered as precise letters. The system then may show a variety of words (of the word database), for example, based on their priority, beginning with said precise letters on the keys of the second keypad 370108. The user may either enter the main proposed word "urge" by providing for example a space character, or he may select one of the words proposed on the keys of the second keypad. In this case, the same procedure may be repeated for the words beginning with a new selected word, and so on.

According to one embodiment, by considering the Fig. 370 as an example, after the system proposes the word "goes" and the other words on the second keypad 370008, then as shown in fig. 371A, if the user proceeds to the correction procedure of the invention by providing precisely the first character "U" of his desired word, for example, by providing the gliding action the 3710211, then the system may predict a word 3710210 corresponding to the input information provided by the user until that moment, and preferably, the system may provide on a second keypad 370208 of the system other words corresponding to said input information (e.g. words beginning with the letter "U" and corresponding to the key presses provided by the user). In this example, one of the proposed words on the second keypad has four letters, and because the database used by the system may not include other words (e.g. with the same number of characters) corresponding to the input information provided by the user, then, the system shows other word(s) being longer than four letters wherein their beginning characters corresponds to the user's input information.

According to another example, after the system proposes the word "goes" and the other words on the second keypad 370008, then as shown in fig. 371B, if the user desires to provide a word beginning with a proposed word in this example the word "goes", then he may changes the status of all of the letters of the predicted word to precise, by for example using the function "Lock Word" as described earlier, by for example providing an appropriate gliding action 370311 from the key 370005. At this time, the system may show word beginning with said precise letters "goes", on the keys of the second keypad 370308. The user now, may either select one of said word(s), or he may provide more input information for example by tapping on keys of the first keypad corresponding to at least one of the remaining characters of his desired word.

According to one method, if the user selects one of the words having more characters than the number of characters corresponding to the input information, then, providing a backspace may bring back the system to the word previously being shown as the main predicted word.

The embodiments above describe methods of word completon which may be also shown for stems being entered precisely through the precise mode instance of the invention. According to one method, at any stage, if there are not enough words/stems to be proposed on the keys of the second keypad, then the system may propose some auto-corrected words on the keys not being occupied by the proposed words/stems.

Digits part of amb.

According to one embodiment of the invention, symbols such as special characters, digits, etc., may be (e.g. ambiguously) assigned to the keys of the keypad of the invention. A word of a database used by the system may include one or more said symbols, preferably in addition to letters. They may be used with the predictive data entry system of the invention. According to one method, any of said symbols may be assigned to a key of the keypad of the invention based on a common characteristic such as the common (partial) shape, as described before. As an example, fig. 372A shows an example, a first keypad of the invention similar to that of the fig. 371 A but here in this example the digits 0-9 are also distributed within its letter keys based on their common shapes characteristics. For example, digits 0, 6, 8, and 9, that have a closed circle in their shape are assigned to the key 372003 which also represents the letters including a closed circle in their shape, digits 1, 4, and 7, that stand on one point are assigned to the key 372001 which also represents the letters standing on one point, digits 3, and 5, that stand on a large base are assigned to the key 372004 which also represents the letters standing on a large base. In this example, digit 2, may either be assigned to the key 372004, if it is considered to have a (straight) large base, or alternatively, it can be assigned to the key 372002 (e.g. including letters that stand on two points) if considering it to stand on two points which is the case depending on the font used to represent the digit "2".

Slide outside/ tap inside to select 1 of 9

According to one embodiment of the invention, tapping on a key of the second keypad of the invention corresponds to interacting with said key. According to one method, providing a appropriate gliding action outside the second keypad may also correspond to interacting with said key. As an example, in fig. 372A, taping on the key 372021 of the keypad 372008 and the gliding action 372013, both correspond to interacting with the key 372021 of said keypad 372008. According to one method, said outside zone may include even the second keypad zone,.

Short for addition/long for correction (vise versa)

According to one embodiment of the invention, during the entry of a word, a short gliding action on/from a key may be assigned the correction procedure and a long gliding action on/from a key may be assigned the insertion procedure (e.g. or vise versus).

According to one embodiment of the invention, during the entry of a word, a gliding action on a key within the borders of said key may be assigned the correction procedure and a gliding action departing from a key and ending outside said key may be assigned the insertion procedure (e.g. or vise versus).

Switch on/off for word proposal

According to one embodiment of the invention, proposing candidates to the user during the entry of a word may be optional. A switching means such as a predefined interaction with a key may be used for the purpose of controlling whether the system may propose candidates. As an example, Fig. 372B shows a gliding action in a predefined direction as such a switching means. In this example, providing the gliding action 372123 departing from the key 372005 may correspond to hiding the second keypad relating to showing the candidate words during the entry of a word, and providing the same gliding action again mar correspond to showing the second keypad relating to showing the candidate words during the entry of a word (e.g. or vise versus).

It must be noted that obviously, each embodiment or methods described and shown throughout this application may be applied to or used by or incorporated into any other embodiment or method described in this application. For example, the three-by-three matrix keys of the second keypad relating to candidate contents described in the embodiments relating to the content search may be applied to the embodiments relating to the GPS application or the general data entry system of the invention as described throughout this application (e.g. or other applications filed by this inventor).

It must be noted that the embodiments relating to content search such as those described throughout this application (e.g. GPS, remote control) are used as examples to describe some principles of the derivative systems or applications of the data entry system of the invention. Obviously, said principles may be applied to any other kind of content/information.

The general data entry system of the invention, and its derivatives such as those relating to vertical markets such as content search or GPS, may open the door to a whole new world of applications that until now were either impossible or cumbersome.

Ordering songs in-car and other mobile situations such as portable media players

As an example, one such application may relate to ordering audio contents such as songs, books, etc., in environments wherein the user can not or is prohibited to look/focus on an input device/unit. One of such case is when a person is driving or walking. According to one embodiment, the (e.g. content search) data entry system of the invention may be implemented within a vehicle such as a car. The keys of the invention may be implemented or used for example, as described before. As an example, during driving, many times a user may want to listen to different songs. In many cases, this is impossible with the current methods of search (currently in the industry). By using the (e.g. the content search) data entry system of the invention, the user may enter (e.g. completely or partially) his desired song/content keyword while driving. According to one embodiment, the corresponding content and/or its

corresponding database may be accessible locally (e.g. within a server/computer in the car) or remotely (e.g. through a wireless communication such as satellite, (e.g. mobile) internet, on a remote server through a cellular/WiFi communication, etc.).According to one method, said server may be part of an organization that charges a user for each song/content used. According to one method, the user may pay for the content download, or he may be able to use said content for a predefined number of times such as according to several predefined times based on business model(s) of the (e.g. content) supplier. According to one method of advertising, each time a user searches to download a content such as downloading a song video during-driving/on the road, an audio/video advertising

content/message may be sent to the user's car to be listened/viewed preferably before the downloaded content is played. This may replace the fee or reduce the fee being asked to be paid for the downloaded content. Many attempts in the industry to provide advertisements by showing indicating businesses in a nearby area during driving (e.g. a car) have been provided without real success, because the user generally has no reason to accept such advertisement. The method of advertisement just described may encourage the user to accept the receiving of an advertisement in exchange of a free or reduced fee download of a content.

This method of using content and its corresponding business model may be applied to any other device/environment such as for example portable media (audio/video) players such as walkman/ iPods, and other contents such as texts/eBooks whrein a text-to-speech software may be used to read the book to the user (e.g. while driving walking).

On-board

According to another example of use, the data entry system of the invention may be used in inflight or on a naval environment. For example, passenger on-board of a plane may use the data entry system of the invention and/or its derivatives such as the content search system to, respectively, interact with each other or select contents to use. They may be able to video-text with each other, use a server computer of the plane for computing such as to use or create content, during the flight. They may use a personal memory device to use or store their information used or created during the flight, etc. It must be noted, that although through this application, the exemplary second keypad (model) of the invention is described as having a three-by-three matrix of keys/zones, and gliding actions from the center of said keypad in different predefined directions (e.g. 8 directions) are described to correspond to interacting with the keys of said keypad, obviously, said keypad may have other key/zone arrangements for providing the same/similar gliding actions. For example, fig. 373, shows an instance 373002 (e.g. corresponding to a key of the first keypad grouping the letter keys that stand on two points) of a round-shaped second keypad of the invention, wherein gliding actions in eight directions may be similar to those used with a keypad having a three-by- three matrix of keys. In embodiments described throughout this application wherein a connector is used to provide a combined/compound word, according to one method, at least some of said connectors may be eliminated and replaced by space characters being considered as connecting words. For example, several words of an entry of the database may be separated by space characters but being considered as a single compound word. Also as an example, several words of a keyword for searching a content may be separated by space characters but being considered as a single compound word, etc.

According to one method, a keyword for searching a content may include one or more words including connectors or space characters (e.g. as connectors). For example, the user may enter a portion of the title of a song, or o portion of the script/lyrics of a song, and the system may search the song/content based on the keyword entered.

According to one embodiment, the keyword entered by the user may be transmitted to the search engine of another (e.g. a remote) application. For example, if the user desired to listen to a music, the keyword entered may be transmitted to the search engine of an application such as YouTube, or Google.

According to one embodiment of the invention, in languages wherein the accuracy of prediction using four ambiguous keys may be low, a system having more letter keys (e.g. 6 to 8, letter keys) may be considered to augment the prediction accuracy. According to one method, the letter keys may preferably be divided into two groups, wherein preferably each group being assigned to one of the two of the user's fingers /thumbs. The letters may be grouped based on their shapes as described before, and each group being assigned to a corresponding key.

Beginners

According to one embodiment of the invention, the data entry system of the invention may be designed such that to include different levels/versions for different user's level such as a beginner or an advanced user. As an example, the version targeted to the beginner user may have less functions on for example the Space and the Backspace keys. According to one method, each of said keys may have four functions assigned to four different gliding action assigned to each of those keys.

Simplified version As mentioned before the arrangement of keys of the keypad may vary. Fig. 374, shows another arrangement of the keys of the (e.g. first) keypad of the invention having four letter keys.

Also the distribution of symbols among the keys may vary. According to one embodiment, a simplified version of the system having few functions assigned to its keys may be provided. This may be beneficial to the beginner users of the system. Fig 374 shows the keypad of the system wherein the keys 374005, and 374006, each include few functions. In this example, Precise Mode ( ree") and Special Character Mode ("123") are assigned to the key 374005. Also, Precising (e.g. changing the status of a letter from Ambiguous to Precise) a

selected/highlighted letter of a proposed word (e.g. by using function "FixLtr") in this example is assigned to (e.g. a gesture towards right of) the key 374005. As mentioned before, after Precising a letter, the system may propose another word beginning with said precised letter and corresponding to the input information provided by the user until then. According to one method, if the user holds said gliding action at the end of the gliding action (i.e., glide and hold action), the user may Precise one by one the characters of the proposed word until the user removes his finger from the touch sensitive surface. In this case, preferably, the system Precises said letters without changing the other characters of the proposed word. Using this method, may permit the user to reach a desired character within a word in order to correct said

character/word, After correcting, the system may propose another word. The Undo/Edit functions here are assigned to the key 374006. Based on the instance/state in which the system is in, one of the functions Undo and Edit may be available.

In this example, an interaction such as a gliding action on/from the key 374006 rightward may correspond to the "Next/Tab" function, a gliding action on/from the key 374005 towards left may correspond to hiding the keypad until the user again touches the screen, and a gliding action on/from the key 374005 towards upper/lower-left may correspond to hiding the keypad until the user manually (e.g. by providing gesture from outside the screen, as described before) brings it back on the screen, and so on. In this case, providing an interaction on the touch sensitive surface (e.g. touch screen) may correspond to a mouse interaction. Those gestures/interactions may be used for other purposes too.

The number of keys may also vary. As an example, fig. 375 shows an arrangement of the keys of the (e.g. first) keypad of the invention having six keys. Having more letter keys can augment the accuracy of the system and may be beneficial for entering text in languages having large number of letters/symbols (e.g. Hindi) and/or languages having many short words such as those not including vowels in the written text (e.g. Arabic). In the example of 375, the letters of Roman alphabet are distributed on six keys based on their common shape characteristics. Letters having a closed zone on the upper side of their shape are assigned to key 375004, letters having a horizontal line in their shape are assigned to key 375007, letters having circled shape (i.e. substantially circular shape) are assigned to key 375002, letters standing on one point are assigned to key 375001, Letters standing on two points are assigned to key 375003, and letters standing on a curved base are assigned to key 375008. SMS

Words used in an SMS text are usually customized words and/or abbreviations. Because those words are usually short, if they are integrated within the general data entry word database, they may reduce the accuracy of the word prediction of the system. According to one embodiment of the invention, the data entry system of the invention may use more than one database of words. As an example, the system may use two databases, a first general database including the words of a language, and a second database including words used by an SMS user. As an example, when user provides key interactions corresponding to a desired word, by default, the system may use one of the databases (e.g. the general database) and may propose a corresponding word of that database. At this time, touching a first zone such as the right zone of a touch-sensitive surface may cause the system to show more corresponding candidates of the general word database to the user, preferably around the finger touching the screen, and touching (e.g. not simultaneously with the interaction of the first zone) a second zone such as the left zone of a touch-sensitive surface may cause the system to show one or more corresponding candidates from the SMS word database to the user, preferably around the finger touching the surface. As an example, fig. 376 shows a device having the keypad and the data entry system of the invention. In this example, after the user presses the keys 736002, 736001, and 736002, the system may propose the word "him" from the main/general database used by the system. At this time, as an example, if the user touches on the right side 376007 of the screen 376010 the system may propose additional words from the main/general database corresponding to said key presses. On the other hand, as an example, if the user touches on the left side 376008 of the screen 376010 the system may propose words from the SMS database of word used by the system corresponding to said key presses. This way, the system/user may faster predict/find a desired word. TTS for candidates

According to one embodiment of the invention, a Text To Speech system may be used to speak the candidate(s) to the user, and the system may be designed such that the user may select at least one of them by providing a predefined interaction (e.g by speech or touch interaction) before, during, or after the speech.

Content app using 4k

According to one embodiment of the invention, the data entry system of the invention may be designed such that only (e.g. four) letter keys and preferably tapping interactions on them for entering words may be used. At least some of the other symbols such as functions (e.g. such as spacing and/or backspacing functions) may be assigned to other interactions such as gestures provided from the keys or from outside the keys. This embodiment may preferably be used for content search applications where preferably no correction is required. According to one method, providing a gliding action from a letter key towards a first predefined direction such as rightward may correspond to a first function, for example, providing a space character, and providing a gliding action from a letter key towards a second predefined direction such as leftward may correspond to a second function, for example, the Backspace function.

Menu / Home Page

According to one embodiment of the invention, a system such as an entertainment application may be designed to use general and/or customized (e.g. for vertical market) data entry of the invention, as an example, a car entertainment application may have a menu/homepage page wherein different sub-applications such as car-functions, audio, video, TV, GPS, email, etc, may be arranged such that each gliding action in a predefined direction may preferably correspond to one of the/those sub-applications. According to one method, at least the main functions of such applications may be provided using gestures. For example, if the system is in the Home Page, the use may provide on a surface a predefined gesture to select one of the application, for example, "Car Functions" application. At this time, another page including car functions such as "Lights On/Off", "Air Condition", "Radio", etc., arranged and shown as described above, on that page. Any of these functions/applications/sub-applications may be selected as described before. The sub-applications may also include sub-applications, arranged in the same manner, and so on. Preferably, at any moment, the user may provide a predefined interaction such that the application goes back to the homepage. Hereafter, an exemplary car entertainment/function tree- based application and its sub-applications may is described: In-Car Home Page

Car functions

Lights

Glide in different directions to select a light

Glide upward/downward for on/off that light

Air condition

Glide upward/downward for Temp high/low

Radio

Channel

Glide Leftward/rightward to select channels

GPS

Music

Video The Menu/Home Page just described may be used to include the appellation (e.g. names/icons) of several different functions/applications. Preferably, they may be shown to the user such that by providing a single gesture provided in a corresponding direction, the user may be able to select a corresponding application. The gliding action relating to interacting with one of the keys of the matrix has already been described in detail. Preferably, the arrangement of the

names/icons of said application may be such that to remind a three-by-three matrix of keys wherein each key relating to one application.

Vol/channel with sliding

According to one embodiment, in a predefined instance, providing a gliding action from anywhere towards four different directions may correspond to volume down, volume up, channel up, and channel down. For example, providing a gliding action towards, upper-left, lower-left, upper-right, lower-right direction, may respectively correspond to volume-up, volume-down, channel-up, channel-down. Search content in a text, based on precise letters

According to one method, while searching for content , if a keyword entered includes only precise character, then, according to one method, the system may search corresponding contents without using the word predictive system. News titles on the fly

The databases of words used by the predictive data entry system may be updated automatically. For example, in case of an application relating to search of news, the keywords related to the news may be updated automatically for example depending on the news at any moment and/or for example, relating to user's references. For example, if the system uses one or more databases of words such as a database of sport keywords and/or athletes, the database of news relating to the (e.g. recent) sport news (e.g. relating to an athlete) may be permanently

(continuously and/or constantly) updated. The permanently updating method, may be an updating procedure executed automatically at any moment, at every predefined laps of time, based on corresponding news events, etc.

TTS with caller's voice

Using a Text To Speech (TTS) system to read aloud (i.e., to speaks) a text/content to a user is known by people skilled in the art. Such system has many use cases. For example, an email or a SMS, or etc., may be drafted by a first person and sent to a device of a second person. The device of the second person may use a TTS system to read aloud said email/message to the second person or to another person. Usually the TTS system uses an automated system based on the voice of a predefined (non-related, and/or integrated) third person to read the message.

According to one embodiment of the invention, the TTS system may use an automated system based on the voice of the first user drafting and/or sending a message, or of a third party decided by for example, any of the, sender or receiver, of the message. For such purpose, according to one example, a system/software may be created and used such that a user can create a database of phonemes and/or spoken words based on his/her own speech/voice. The TTS system using said database may be integrated within and/or used by the device of the second person such that (e.g. each time) a message is received from a first person by the device of the second person, the TTS system may read said message with the voice of the first person, or optionally, with a voice of a third person which may be decided by any party. The TTS system may use a different, personalized, database of phonemes or spoken words for each different person's voice. The database may be integrated within the device of the second person, and/or it may be in the "Cloud" (i.e., on a server) used by a device of the second person. According to one embodiment, when reading a message, a background music/audio may be played along with the TTS reading the message/text, said music/audio being decided/selected and sent by one of the parties or by a third party, to the receiver's device along with the message, or from another source. Shoes

According to one embodiment of the invention, the letter keys of the system may be integrated within the shoes of a user or may be represented by different interactions provided on/with the shoes of a user. For example, each shoe of a user may include two of the letter keys preferably on a bottom surface on the shoe, such that each key being activated separately. In this case a pair of shoes of a use may include the four letter keys of the system, and be used with the data entry system of the invention. For example, while walking/jogging the user may search for a content such as a song by for example providing one or more pressing actions with a different portion of his shoes relating to one of said keys. Said keys may be hard keys or soft keys. According to one method, instead of keys, the shoes may be used with a motion detecting means to transform different predefined user's shoe motions to different input signals assigned to at least the letters.

Words including special characters

As mentioned before, the ambiguous keys may include special characters and letters such that by interacting with them the user may enter ambiguously a desired word such as a URL (which may be considered as a single word) and the system may predict it. In this case the words of the database used by the system may include words that include letters and special characters. As mentioned throughout this application, in addition to some letters, each of the keys of a keypad may include a group of special characters. Those characters may also be assigned to said keys as ambiguous characters. Therefore, according to one method, if the words of the dictionary are long words such as URLs, the chances of entering such words without requiring correction procedure is high.

According to one embodiment, if a correction procedure is required to enter a word including special characters, the system may be designed such that the correction procedure may

preferably be applied only/mainly to the letters of the word being entered, while preferably, ignoring the special characters.

It must be noted that according to one embodiment, during the entry of a word (of the dictionary/database) including/containing special characters, the user may press the keys corresponding to the letters and special characters of the word being entered. Regarding the special characters, the user has just to remember which special character belongs to which group, and press the key corresponding to said group. As an example, fig. 377shows a keypad of the invention including ambiguous letters and special characters. In this example, mathematic group of characters is assigned to the key 377001, digits group of characters is assigned to the key 377002, punctuation group of characters is assigned to the key 377004, and "other/rare" group of characters is assigned to the key 377003. The characters of each group are shown as example if fig. 356A. as an example, in order to enter ambiguously the beginning characters "http://" of the URL 377010, the user may press the keys, 377002, 377001, 377001, 377004, 377004, 377001, and 377001. In this example, character ":" belongs to the punctuation characters group, and character "/" belongs to the mathematic characters group.

According to another embodiment, the correction procedure may also be applied to the special characters of a word being entered. In most cases, it may be enough to indicate to the system that the selected character position corresponds to a symbol of one of the group of the special characters without specifying which character. For this purpose, according to one method, a first predefined interaction such as a first gliding action in a predefined direction on/from a letter keys may correspond to the group of special characters assigned to said key. If desired, one of the special characters of the group may be entered precisely in said position. As an example, after the user provides said first gliding action, he may provide a second gliding action corresponding to a special character of said group. For this purpose, according to one method, after terminating the first gliding action, the system may show the corresponding special characters under the same finger, or under another user's finger touching the screen, and the user may provide the corresponding second gliding action to select one of the special characters.

Visible/Invisible modes

According to one embodiment of the invention, the system may be designed such that in the visible mode (e.g. when the keys are shown on the screen), the Precise Mode instance (e.g. of the keypad) may be available to the user when requested/required. In the visible mode, when the system is entered into the Special Character Mode, preferably, in order to enter a special character the user may tap on the corresponding special character. On the other hand, if the system is in the invisible mode, preferably, the user may not have access to the Precise Mode instance of the keypad for interactions such as tapping interactions. In the invisible mode, if the system is in the Special Character Mode, then in order to enter a special character, preferably, a gliding action as described before may be required. This permits a full blind typing ability for experienced users. Preferably, the system may b equipped with TTS software/system to at least read/speak aloud the current word being entered and/or corrected. In this case, the invisible mode permits the user to enter any text without requiring looking at the screen/device at all.

Invisible, but showing shapes According to one embodiment of the invention, in the invisible mode the approximate/exact location of the invisible letter keys on the screen may be marked by the corresponding common letter shapes (characteristics) if the system is in Letter Mode, or by the corresponding icons/informing means representing the corresponding group of special characters, if the system is in Special Character Mode. Figs. 378A, and 378B, show examples of such cases, respectively. In the examples of figures, 378A, and 378B the exemplary limits/borders of the predefined zones of the keys are shown with discontinued frames lines because the borders/limits of the zones of the keys are preferably not (meant to be) shown on the screen/corresponding-surface. By using this embodiment, the screen of the device is free of keys and the user only may see four shapes each relating to its corresponding invisible key on the screen. According to one method, the location of the invisible space and invisible backspace keys (e.g. on the screen) may be marked by two icons for example such as "->" and accordingly.

Slide towards the 1^st Kev

A gliding action provided over several keys of a virtual keypad (e.g. or provided anywhere on a surface relating to gliding actions over several keys of an imaginary keypad model) relating to interaction with said several keys has been described in detail in previous patent applications filed by this inventor. As described, said interactions may be used with the word predictive system. As mentioned before, according to one method, a gliding action provided-on/relating-to several keys may be a compound gliding action having one or more changes of direction.

According to one embodiment of the invention, the first portion of such gliding action may relate to a gliding action provided from the center of the keypad (e.g. model) towards a first key to be interacted. Then, if needed, the user may change the direction of the gliding action, towards the second key to be interacted, and so on. Fig. 379A shows as an example, the letter keys of a keypad model 379010used by the system. As shown, providing a compound gliding action 379007 provided on the screen 379000 of a device may correspond to the word "THE" (e.g. having the highest priority among the words corresponding to the same key interacting actions) predicted by the system. In this example, the first portion 379001 of the compound gliding action 379007 corresponds to interacting with the key 379011 of the keypad model 379010, the second portion 379002 of the compound gliding action corresponds to interacting with the key 379012, and the third portion 379003 of the compound gliding action corresponds to interacting with the key 379013. At the end of the entry of the word, the user may provide an end-of-the-word signals such as a space character. According to the method, the gliding actions may be provided in different separate portions of an entire compound gliding action wherein the user may remove his/her finger from the sensitive surface between said separate portions. At the end of the entry of a word using this method, the system/user may enter an end-of-the-word signal such as a space character. Fig. 379B shows an example of the current method for the entry of the word "THE". In this example, the user removed his finger after the entry of the first portion 379101, and provided the remaining portions 379102 of the compound gliding action 379007 of fig. 379 A. According to one method, one or more portions of a compound gliding action may be sufficient for predicting a word.

Hold letter kev(s) to relocate the keypad

According to one embodiment of the invention, the keypad on the screen may be relocated by press-and-holding at least one of the keys, preferably, at least a letter key, of the keypad or a predefined area on the screen of the device, and dragging the keypad. If the keypad is a split keypad, then according to one method, by dragging one portion, the second portion may also be dragged accordingly. According to another method, the user may use two fingers (e.g.

preferably, simultaneously), each dragging one of the portions of the split keypad. lock numbers and arith chars

According to one embodiment of the invention, preferably, when the system is entered into the Special Character Mode, if the user enters a symbol belonging to a first type of group of symbols such as Numeric, and/or Arithmetic characters, and/or functions, then the system may not automatically go back to Letter Mode (e.g. until the user switches it back manually). On the other hand, if the user enters a symbol belonging to a second type of group of symbols such as those other than the Numeric, and/or Arithmetic characters (e.g. such as a group of symbols comprising punctuation mark characters), then the system may automatically go back to the Letter Mode after said symbol is entered. In either case, if the user wants to repeat a symbol other than the Numeric and/or Arithmetic character, then he may provide a tap-and-hold, or a glide-and-hold action corresponding to entering said symbol. The system may repeat entering said symbol until the user does a predefined interaction such as removing his/her finger from the touch sensitive surface. Obviously, the example of automatic and/or manual switching actions between different modes as just described can be applied to any group of symbols.

Precise punctuations

According to one embodiment of the invention, during the entry of a word, if a special character such as a punctuation mark character is entered precisely, then the system may add said special character at the end of said word being entered. According to one method, at this time, if the user continues to enter additional ambiguous characters, then the system considers all of the previous characters of the Current word as a single word, and enters it, and relates the ambiguous characters entered after said precise special character as part of the new Current Word. Preferably, In this case, the system continues to show the previous Current word together with the new Current Word until the user provides an end-of-the-word signal such as a space character.

As an example, in order to enter " 's " at the end of a word, such as the word "Robert", the user may first enter said word, then, provide enter the apostrophe character by for example providing the corresponding gliding action on/from its corresponding key. At this time (i.e. after the entry of the apostrophe character), the system may consider the word "Robert" as entered, although, preferably, showing it still as the beginning of the current word to be completed by additional characters (e.g. that may follow the apostrophe character) which all together may not be in the dictionary as a word. The user then, may add the word "s" to the end of the chain of characters " Robert' ".

This principle may be applied to any other language. For example, in French language, there may be words ending with apostrophe such as " qu' ". Such a word may be the beginning of another word such as the word " qu'avec " which may not be in the dictionary. In the manner just describe, such words may be entered.

According to one embodiment of the invention, a special character may also have derivative character(s). As an example, the special character apostrophe may have the derivative " 's ". Such derivatives may be entered in a manner as described in this and in the previous patent applications filed by this inventor.

Correction + entering additional amb. character(s) - lock until additional character(s)

According to one embodiment of the invention, during the entry of a word, after ending the correction procedure to enter a precise character, if the user begins to add additional ambiguous characters of the word, the system Locks the previous characters (i.e. turns the status of said characters from ambiguous to precise) of the Current word until the first character of the additional ambiguous characters .

Remote KB for a computer According to one embodiment of the invention, a first device in which the data entry system of the invention is implemented may send/transmit the characters/words/functions and/or input information to be entered into a second device/application such as into a text editor of a PC. In such case, preferably, the screen of the first device may show at least a portion of the content printed/shown on the screen of the second device, preferably, a zone around a cursor/pointer on the screen of the second device. For such purpose, according to one method, the second device may transmit, wirelessly or through wires, the image of said zone to the first device. Said cursor/pointer may be manipulated through the first device, for example, by interacting on the screen of the first device.

Using speech

According to one embodiment of the invention, the correction procedure of the invention may be provided through speech, and analyzed by a speech recognition system. For example, while driving and texting, the user may correct an undesired proposed word by speaking the first character of the desired word. According to one embodiment, the user may speak his desired word replacing a non-desired proposed word. Correcting a word based on providing key interactions and speech has already been described in detail in different patent applications filed by this inventor. Additional features

According to one embodiment of the invention, in the Menu, one of the menu pages may correspond to applications such as social application such as Facebook, twitter, etc. By providing a gesture from anywhere towards the application title/icon, the user may switch the system focus towards the selected application.

Around phone

According to one embodiment of the invention, a social system may be designed for communication devices. For example, the devices of users within a group may include a tracking system that permits to track the devices of each other within a predefined size of perimeter. This method of tracking may permit the users of the group within perimeter to for example meet with each other.

Linguistic Rules -

Accepting / rejecting a predicted word According to one embodiment of the invention, a predictive system predicting words generally based on a 1-gram word database, may be combined with linguistic rules and a very simple method of a user's alerting/informing the system about the status of a predicted word (e.g. accept/reject a predicted word) to provide a high accuracy of prediction. According to one embodiment of the invention, by considering a plurality of suggested individual words as a combined predicted word (e.g. N-gram) and presenting it to the user, and the user's ability to assign a status of accepted or rejected to each individual word (e.g. single word) of said combined predicted word, and processing of the input information corresponding to said combined predicted word within the corresponding group of combined (e.g. N-gram) words with the corresponding N-gram word database, the system may provide a very high rate of accuracy of word prediction.

Preferably, the system may be designed such that user's simple interaction may lead (e.g. cause or enable) the system to accurately predict at least one of the beginning words (e.g. single words) of the combined predicted word. The user then, may confirm/accept said at least one of the beginning words.

Hereafter the principles of such system are described. Fig. 383, shows an exemplary diagram demonstrating some of the general principles of this embodiment of the invention. Given a situation where the user enters an individual word and where there is no input information relating to a combined word (e.g. the user enters a word, which may become the first word of a combined predicted word, or which may be considered to be the first word of a combined predicted word), and where a variable, for example N, designates the position or index value a word, possibly in a combined predicted word. Given the described situation and given N=0, the system (i.e. an implementation of the embodiment) is considered to be in the Start state (383001) of the embodiment.

From the Start State, the system will increase N such that N=N+1 (383012). The system may respond to the entry of a new current word (e.g. hereafter may be referred to as Current Predicted Word), such that, in general, as the new current word is being entered the system may change (383007) some of the other words (i.e. those words that do not have confirmed word- state) in a combined word. Given the case that N=l, the user may choose (383003) to (a) confirm the current word and consequently, the word-state (e.g. status of a word) of the current new word (e.g. being predicted/proposed) may be considered confirmed (383010) and then N will be decremented by 1 such that N=N-1 (383011), or (b) reject (383004) the new word in one of two ways; (1) by modifying (383008) the word (e.g. by correcting the word by some method) in which case the system may, for N>1, change some of the other words (that do not have confirmed word-state) in the combined word, or (2) by explicitly rejecting (383005) the new word where, in this case, the word-state (and/or the word) will be considered, by the system, to be in a rejected state (i.e. an ambiguous or unconfirmed state). The system may consider this rejected word to be part of a new combined word and increment N (383012) to designate the index for the new current word of the combined word, such that, in this case for example, N=N+1=2. After a word is confirmed, the system may find that N=0 (383002) in which case the variable, N, will be incremented (383012) and the user may enter further information. If the system finds (383002) that NoO (i.e. N does not have value of zero), then the user may make a choice (383003), as previously described above. A situation may arise where a new word may be entered and rejected and/corrected in several iterations such that N is greater than 1 without limit (N>1).

Given that N>1, a new current word may be entered (383006). Again, similar to the situation before, above, the system may respond to the entry of the new current word (e.g. with index N=2), such that, in general, as a new current word is being entered the system may change some of the other words in the combined word (e.g. the previous word with rejected word-state, for the situation where N=2).

Now, for any N >0, the user may make choices (383003, 383004) as before in a similar fashion, and so on, such that N may increase (383012) and/or decrease (383011) at various steps (as shown in Fig. 383) as the user enters new words (383006), modifies the new current word (383008), and/or chooses to set the word-state of each word to the status of confirmed (383010) or rejected (383005).

According to one method, if a word is being entered, the system may consider a number of preceding words (e.g., and, if needed also considering a number of following words after they are/will-be entered) such as at least the last word before the words being entered in order to better predict the word being entered. If said preceding word is provided before a special character such as a dot or comma, etc., the system may preferably not have to consider the preceding word. Accordingly, if the system proposes a desired word, the system may provide an end-of-the-word signal such as a space character. On the other hand if the word being entered is not the desired word, then the user may inform the system that the predicted word is not his/her desired word. The system then may propose another word by, preferably, still considering said at least one preceding word. If, the new predicted word is the desired word, the user may provide an end-of-the-word signal, if not, the user may either proceed to the correction procedure of the invention, or preferably, he/she may proceed to entering the next word, wherein, in the last case, the system may know that the word proposed in not what the user wants and may keep it on the front. Now, the system may consider the previous undesired word and the new word being entered and predict a set of two words, which at least one of them may be a new word.

With continuous description of current embodiment, several scenarios may happen:

According to a first method:

1. if the first word is predicted correctly, the user may provide a first end-of-the- word signal to enter it. If not, according to one option, he may proceed to correcting it, and entering it. By correcting said word, the system may propose another word as the second word. After correcting the first word, the user may provide an end-of-the-word signal to enter it.

2. if the second word is correct, the user may provide an end-of-the-word signal to enter it, if not, the user may proceed to either correcting said word, or he may proceed to entering a new word; in the last case, the system may know that the word proposed is not what the user wants and may keep it on the front.

And the procedures above may be repeated.

According to a second method:

1. if the first word is predicted correctly, the user may provide a first end-of-the- word signal to enter it. If not, according to one option, regardless of if the second word is predicted correctly or not, the user may proceed to entering a third word, and the system may predict a set of three words, wherein at least one of them may be a new word. According to this method, either the user may repeat this procedure until one or more consecutive beginning words of the set are predicted correctly, so that he may enter them by providing the corresponding number of end-of-the-word signals to enter them, or he may proceed to correcting at least the first beginning word of the set and enters it by providing the end-of-the-word signal. In this example, each beginning word of the set which is correct may be entered by providing an end-of-the-word signal.

In the embodiment above, the number of words of the current word may vary depending on how many words are signaled to the system as being non-desired words.

Using linguistic rules by considering two or more words, may be facilitated by using list(s) of N-gram words, such as one-gram, bi-gram, three gram, and more gram words. Such N-gram words/lists may each constitute a separate databases based on the value of N, or at least some of them may be combined together in a same database, based on different defined values of N. Preferably, the entries of the N-gram database may each have a frequency/priority value.

The embodiments just described are provided as examples to described the principles of entering words using a word predictive system and/or considering linguistic rules. Modification and derivations may be applied to said embodiments by people skilled in the art. For example, is all of the words of the set are correct, the user may provide a single end-of-the-word signal to enter all of them.

It must be noted, that principles just described may be applied to any word predictive system including the word predictive systems of the invention.

In the above mentioned embodiments, when a word is being entered:

- according to a first method, the system may preferably not consider the previous correct word that has been entered. In this case, as an example, the word having the highest priority may be proposed to the user.

- according to a second method, the system may consider the previous one or more words if one of them is not the desired word.

according to a third method, the system may consider the previous one or more words regardless of being correct or not.

Non-desired combined predicted word

A non-desired combined predicted word is a combined word including at least two (e.g. single) words, wherein at least the first word is a non-desired word. When a word predicted by the system is a non-desired word, the user may inform the system that the word is not what he/she wants and proceed to the entering of the next word. After entering the next word, if still the first proposed word(s) (e.g. in the embodiment, the first proposed word(s) may change during or after the entry of the next word) is a non-desired word, and even (e.g. despite the fact that) the second predicted word is correct, the system/user may consider the combined word as non-desired. At this time, different scenarios may be considered:

- according to a first scenario, the user may inform the system by providing a same or another interaction. The user then may then proceed to entering a next new word of the combined word, after which at least the first predicted word of the combined predicted word may be predicted correctly. If not, either the user again informs the system that the new combined word (e.g. which now, includes one more word) is still non-desired, or he/she may proceed to correcting the first word, which may result in automatic correction of additional consecutive words.

- according to a second scenario, the user may proceed to the correction procedure of the invention, to at least correct the first word of the combined word. According to one embodiment, each time the first word of the combined predicted word is correct, the user may provide a predefined interaction such as providing a space character to confirm/enter it. Then the system may remove said word from the combined word. This may apply to (e.g. each of) the following correct words of the combined predicted word. According to one embodiment of the invention, if the first word of a combined predicted word is not correct, then regardless of the status (i.e., desired/correct or non-desired/not correct) of the following words of the combined predictive word:

- if the user proceeds to correcting the first word and confirms it, and the second word is being an undesired word, the user may begin typing a next new word. The system then understands that the last predicted word was also undesired/not correct. According to one method, the system may also add a space character between the last predicted word and the one being entered.

It must be noted that based on the principles above, many more scenarios can be considered by people skilled in the art, but we show those requiring less and simple interaction from the user. . For example, according to one embodiment, a user may provide a first predefined interaction such as a first type of gliding action to inform the system that the word proposed is incorrect, and may provide a second predefined interaction such as a second type of gliding action to inform the system that the word proposed is correct. By considering the status of each word of the combined predicted words that may include two or more words (wherein preferably, the first word in non-desired), and comparing it with the corresponding entry of a corresponding N-gram dictionary/database the system may better predict the entire combined desired word, or at least the first word(s) of the combined predicted word. The user may provide a number of predefined interactions such as providing (e.g. for a number of times) the space character(s) relating to the number of correct words of at least the beginning of the combined predicted word. According to one embodiment, the second type of interaction may be avoided and replaced by user's behavior. For example, after entering input information corresponding to an additional word of the combined predictive word, if the user proceeds to correcting the first word of the combined predictive word, then, the system may consider that the new word just entered as a correctly predicted word.

Example - 1

An example of a preferred embodiments of the invention is provided hereafter. By considering the keypad 379010 of fig. 379 A, as an example, if the user presses the keys 379012, and

379013, to provide the key press value 23 for the first word (e.g. "as") and informs the system that the corresponding predicted word "as" is not correct, and provides additional key presses corresponding to the key press value 2334 for the second word. The corresponding 2-gram words of an exemplary 2-gram database in the English language may be as follow:

Table 1 - key value comb, word frequency

23 2334 we need

23 2334 he also

23 2334 we also

23 2334 he held

23 2334 we keep

23 2334 we help

In this example, the combined predicted word "as also" may preferably not be included because both of the single words (in the combined predicted word) have highest priority in their corresponding groups of single (e.g. 1-gram) word database. As an example, after entering this key press information, by knowing that the word "as" is not the desired word, the system may propose the combined predicted word "we need". At this time:

1^st exemplary possibility - If the word "we" is the desired word, then the user may provide an end-of-the-word signal such as a space character. The system enters said word, and the combined predicted word at this time may as a result include one word, which is the word "need". If the second word is also correct, the user may provide an additional end-of-the-word signal (e.g. a space character) to enter the second word too. According to one embodiment, a predefined interaction may confirm and enter all of the words of the combined predicted word. With continuous description of the current example, if the second word is not the desired word, the user may proceed to the correction procedure.

2^nd exemplary possibility - If the word "we" is not the desired word, the user may proceed to the correction procedure. The system relates said procedure to the first word of the combined word, and, for example, proposes the word "he", and provides the combined word "he also". If the combined word is correct, the user confirms it as described above. If only the word "he" is correct the user confirms and enters it. Then he either proceeds to correcting the second word "also", or he proceeds to entering a new additional word, which in this case the system may understand the word "also" is not a desired word and may change it to or replace it by the word "held". By providing the input information corresponding to the new word, the system may consider the word "held" as the first word of the new combined word, which in this case may be considered/regarded as a correct word based on the combined words of Table 1.

With continuous description of the current example, if the input information of the new word is the key value 21, then the new possibilities according to the Table 2 may be as follows:

Table 2 - key value comb, word frequency

2334 21 heicTat 1924 2334 21 keep_my 1831 2334 21 also_at 1148 2334 21 held_my 738

At this time, if the word "at" which has just be predicted (as part of the new combined predicted word "held at") is correct then the whole combined word "held at" is correct and the user confirms and enters it. On the other hand, if the word "at" is not what the user wants, then, the user may provide an interaction such as a gliding action on/from the space key towards left side (i.e. leftwards), to inform the system that the word proposed in not correct. Now, the system proposes a/the (only) combined word that corresponds to a combined word of the database of Table 2 that its first correct word "held" and its second word is a word other than the word "at". Such combined word is this example, the word "held my". The user then may confirm and enter said combined proposed/predicted word. Alternatively, after proposing the combined word "held at", the user may first press the space key to confirm the word "held", before proceeding to decision about the second word "at" of the combined predicted word.

According to a preferred embodiment of the invention, when a user provides the input information corresponding to a first word of the current combined word, the system may provide the word having the highest priority among the words of the corresponding 1-gram database. According to a preferred embodiment, before proposing said word to the user, if there is a previous word just before the current predicted word, the system may combine said two words to provide a temporary combined predicted word. The system then looks for that combined word within the corresponding group of 2-gram dictionary/database to see if it exists there. If such combined word does not exist in the database, the system may consider that the

frequency/priority value of such combined word as being lower than the lowest frequency value in that group, and according to one method, the system assigns the priority value "0" to the combined predicted word. On the other hand, if such combined word exists in the database, the system may consider its priority as the priority of the combined predictive word. In both cases, the system may compare the priority value of the combined predicted word with the priority value of the combined word within the 2-gram word database. If the frequency/priority value of the 2-gram combined word, having the highest priority relating to the other combined words of the group within the 2-gram word database, is significantly higher than the priority value of the combined predicted word, then the system may propose the second word of the combined word to the user, if not, the system may propose the second word of the combined predicted word to the user (i.e. the first word is already being entered). According to one method, a new word added to the dictionary may be assigned with a priority calculated by a method that may improve the prediction accuracy.

Another embodiment

Fig. 380B shows an exemplary diagram according to a preferred embodiment of the invention. In the step 380011, the user enters the input information corresponding to a desired word, which in this example is being the first word of a Possibly-Combined Predicted Word. The system may predict a word based on the received user's input information and, preferably, additional information related to a method of consideration of the previous word and/or the previous special character (e.g. a dot). A diagram of an exemplary such method is described in fig. 380A which will be described later. In the step 380012, the user may decide if the first word of the possibly-combined predicted word (which in the current instance has one individual word only) is correct or not.

Option 1 - If the first predicted word is correct, then in the step 380013, the user may confirm it and enters it. Now, as shown in step 380014, the combined current word has no words.

Option 2 - If the first predicted word is not correct, then in the step 380015, the user may reject. As a result, the system may propose another word instead as shown in step 380016. Now, as shown in step 380017, the system may inform the system if the new proposed word is correct. If YES, the user may proceed to step 380013, if NO, he may proceed to step 380018 by entering input information corresponding to the next (in this example, the second word of the combined word, which also informs the system that the new proposed word in a non-desired word. At this time, as shown in step 380019, the system may attempt to replace one or more of the already rejected words of the combined predicted word (if any), by other words of the corresponding locations within the group of corresponding N-gram word of the database used by the system. Then, as shown in step 380020, the user may be required to inform the system if at least the first individual word of the combined predicted word is correct. In this case, as an example:

Option A- If the answer is yes, then as shown in step 380023, the user may proceed to confirming/entering said word. In this case, as shown in the 340024, the second word of the combined predicted word may become the first word of it. The system/user now proceeds back to the step 380012.

Option B- If the answer is no, then as shown in step 380021, the user may proceed to changing said word by for example proceeding to the correction procedure of the invention, in which case the system may also change . After the system proposes the desired word, the user may proceed to confirming/entering said word, as shown in step 380022, the system may attempt to replace one or more of the already rejected words of the combined predicted word (if any), by other words of the corresponding locations within the group of corresponding N-gram word of the database used by the system. Now, the system/user may proceed to step 380023.

Regarding the exemplary diagram of fig. 380 A, as shown in step 380001, when the user enters the input information corresponding to a first word of a combined predictive word, then, as shown in step 380002, according to one method, the system may first consider the word having the highest priority in the corresponding group of the 1-gram word database used by the system. Then, in the step of 380003, the system may look for a previous word in the text (e.g. according to one method, if the last character before the cursor is an end-of-the word character, then the system considers that there is no word before the current word being entered). If there is no word, then as shown in step 380009, the system may propose the considered word, and exits. But, if there is a previous word, then, as shown in the step 380004, the system combines both words and considers them. Now, as shown in step 380005, the system looks to find out if the combined considered word exists in the corresponding group of a 2-gram word database used by the system:

Option A Step 380006 - If the answer is NO, the system may assign a very low (e.g. "0") frequency/priority value to said combined considered word.

Option B Step 380007- If the answer is YES, the system may verify to see if the combined considered word has a much lower priority/frequency than the 2-gram word having the highest priority in the corresponding 2-gram group of words of the 2-gram database used by the system. In this case:

Option A- if the answer is NO, then the system may propose/predict the individual word that it considered word, as per step 380009, and exits.

Option B - if the answer is YES, then the system may propose/predict the individual last word of said 2-gram word having the highest priority/frequency, as per step 380008, and exits.

Example 2

An example of another preferred embodiment of the invention is provided as follows:

By considering the keypad 379010 of fig. 379 A, as an example, for entering the first word of a sentence, if the user presses the keys 379011, and 379012, to provide the key press value 23, the system may propose the words "as" having the highest priority among the words corresponding to said key presses and may show it as an individual word in a front frame to the user.

If such a word is not the desired word, the user may provide a predefined interaction such as a predefined gliding action to inform the system that the corresponding predicted word is not correct/desired and the word is rejected.

As a result, the system may propose the word "he" which has the highest priority after the word "as" in its corresponding key press value (e.g. key presses) group of words of the database. (Alternatively, according to a first method, after the system proposes the word "as" to the user, proceeding to entering the next word may inform the system that the word "as" is rejected, while according to a second method, proceeding to entering the next word may inform the system that the word "as" is accepted.) he

After proposing the word "he", as an example:

Option 1 - if the word "he" is correct, the user may confirm/accept it and the system enters it. The user, then, may provide the key press value 2334 corresponding to his/her next desired word. In this example, the system may provide the word "also". also

If "also" is what the user wants, he may confirm/accept it. If not, he may:

- either reject it, and by considering the previous confirmed word "he", the system may

provide the combined predicted word "he held" as per corresponding group of 2-gram list as shown in Table 3, and preferably presents the second word of the combined predicted word as the new predicted word to the user. held or the user may (somehow) proceed to entering a next word, in which case and system may be designed to consider the word "also" as being rejected, and provides the combined predicted word "he held", which is the only corresponding combined word in Table 3.

Table 3 (e.g. 2-gram, key-value group 232334) key value comb, word frequency

232334 we need 28020

23 2334 he also 16729

23 2334 we also 7850

23 2334 he held 3879

23 2334 we keep 1995

23 2334 we help 504

Option 2 - The user does not confirm the predicted word "he" and may proceed to entering the key presses corresponding to the next word, in this example, the user provides the key press value 2334. By providing the new key presses, the system understands that the word "he" is not correct (i.e., because it was not confirmed, after replacing the word "as"), and may propose the combined word "we need" which has the highest priority among the words of its corresponding key-value group, as shown in table 3, and corresponding to the input information provided until then. we need

In the next step, if the desired first word is "we", the user accepts/confirms and enters it by for example providing a space character. Now, the combined predicted word may include only one word "need".

need

If the user desires said word, he may also confirm and enter it.

On the other hand, if the user does not want the proposed word "need", the user:

Option A - may either reject said word "need", and the system may propose a new word "also", also and now, he may again reject the new word "also" to get the word "keep", or he may proceed to correcting said word "need", by for example, providing a precise letter such as "h" in order to get the word "held",. Etc. It is understood that examples provided are based on considering Table 3. Option B - may proceed to entering a next word, for example, by providing the key presses value 14. In this case, the system may understand that the word "need" is not a desired word because it was not confirmed. (In a preferred embodiment, this option may not be permitted, and the user may first explicitly reject a proposed word if it is not desired, and then proceed to entering a next word).

Now, as an example, the system knows that the words "need" is already rejected by the user, and therefore, by considering the Table 4, after entering the new key presses, the system

automatically may propose the combined predicted word "help jr". Now, the user may either confirm the first one or two words of the combined predicted word, if it is his desired word, or he may for example proceed to entering a next new word by for example, using the FixLtr function to precise the first letter of the combined word "help jr", so that the system may provide another combined word of the group beginning with the same letter, which in this example is the combined word "held to". Table 4 (e.g. 2-gram) - key value comb, word frequency

2334 14 need to 83113

2334 14 helpjr 4526 2334 14 held to 1401

The embodiments described above combined with the data entry system of the invention may provide a simple word predictive data entry system with high accuracy of prediction in any language in general, and may be particularly beneficial in some languages where the accuracy may become acceptable. This system may also be beneficial in languages where the words are short, such as Arabic, Hindu, Japanese, Korean. It can also be beneficial for databases using databases of words that include SMS word. Accordingly, this system may also be very beneficial for searching contents, because the user may be permitted to enter any one or more words corresponding to a search field such as one or more, preferably consecutive, words of a title of a song/video, etc.

It must be noted that in some cases, some words such as "and" provided before a current word may not be considered by the predictive system when said current word is being entered.

According to one embodiment, when predicting the next words, in some cases, previous undesired words may not be considered by the system if there are some types of special characters (e.g. period/dot, comma, etc.) and/or some words such as "and", "or". An additional embodiment

Based on principles of accepting or rejecting and considering N-Gram word databases, to predict words, many other methods and embodiments may be considered. Diagram of fig. 381, is another example of such methods, (e.g. This diagram is self-explanatory). In this embodiment, when the user enters the input information corresponding to a second word and the system predicts a word, the user can/may accept or reject it before accepting the first (e.g. single) proposed word of said combined predicted word.

The system may permit one or more rejection actions for a proposed individual word of a (potentially) combined word, or for the whole combined word. According to a first method, the system may not permit more than one rejection for an individual word of a, or for the whole, proposed combined word, and the user may either proceed to the correction procedure of the invention, or he/she may proceed to entering the next word and the system may change the previous word/s, while according to a second method, the system may permit more than one rejection for the aforementioned. An exemplary N-Word predicted word diagram

Fig. 382 shows an exemplary system relating to a word predictive system predicting an N-gram combined predicted word, and the method of managing such combined word for entering text, (e.g., The diagram is self-explanatory). More than one accept/reject means to control more than one word of individual words of combined predicted word

According to one embodiment of the invention, one or more means may be used to enable a user to accept or reject any of the words of a combined word. As an example, a navigating means may be used to navigate within the words of a combined predicted word to select one or more of its individual words to accept or reject said individual words. According to another embodiment, the system may have more than one said means, each related to point-to/select one (or more) of the individual word of a combined predicted word, to accept or reject said individual word/s. As an example, a predefined gesture provided in a predefined direction on from the backspace key may correspond to rejecting the first word of a combined predicted word, and a predefined gesture provided in a predefined direction on/from the space key may correspond to rejecting the second word of a combined predicted word. Preferably, if a combined predicted word has two individual words, after both words are predicted correctly, providing a single end-of-the- word signal such a space character may confirm and/or enter both of them. Next word is combined

According to another embodiment, instead of or in-addition to considering the previous word, the system may be designed to apply to consider the next words to be entered.

Dynamic priorities

Priorities of the words relating to each other, within of a group of words corresponding to a sequence of key-values, may vary depending on the location of said words within a text such as within a sentence. For example, priority of a word within a group relating to other word of said group may vary of said word is at the beginning, in the middle, or at the end of a sentence. A sentence may be considered to begin at the beginning of a document, after and end-of-the-word- signal (e.g. dot, comma, "?", etc.). As and example, the word "him" may have higher priority over the "win" if it is used in the middle of a sentence. The same word "him" is not generally used at the beginning of a sentence, therefore, it may have lower priority relating the word "win", when it is used at the beginning of a word. Undesired single word, looks back to combine, then proposes 1

Preferably, if the user rejects a current predicted word having one predicted word, then the system considers the previous word within the text, and by comparing the combined information with the corresponding portion of the corresponding N-gram dictionary, the system may propose another word to the user. If the word being entered is the first word of the sentence, then, the system may propose another word accordingly (e.g. preferably, by considering the dynamic priority of the word to be proposed). Said proposed word may be replaced by another word, if either the user rejects it or he proceeds to entering the next word without accepting it.

Words beginning with the same letter as highest, being the first in the list of N-gram group - to lower the priority of words that begin with the same character

According to another embodiment, the combined words of a group of combined words relating to a same key values may be arranged such that to minimize the user's interaction to select them. As an example, in the example of Table 4, the words of the group of combined words may be arranged as follow:

2334 14 helpjr 83113 2334 14 needjo 4526 2334 14 heldjo 1401

Such arrangement permits quick selection of a desired word. For example, if the combined word "help jr" is presented to the user is an undesired combined word, then, rejecting the second word results in presenting the next word in the list "need to", or using the FixLtr function to precise the first letter of the combined word "help jr", results in that the system may provide another combined word of the group beginning with the same letter, which in this example is the combined word "held to".

Dictionary for combined separately being organized/solicited

According to a preferred embodiment, database of a same N-gram word may be a separate database from the database of 1-gram words. Next new word until getting the 1^st word is correct

According to another embodiment, each time the system predicts a new word based on user's input information relating to his/her desired word, the system may update the combined predicted word and accordingly may change any of the single words of the new combined predicted word. According to one method, the user may confirm (e.g. by providing a gliding action towards the right side) or reject (e.g. by providing a gliding action towards the left side) each new the proposed word. The user then, may proceeds to input a next word, unless one or more beginning words of the combined predicted word are the desired word, in this case the user may enter them by confirming them (e.g. providing several consecutive space characters corresponding to the number of the correct beginning words).

Considering predefined N words, but changing the 1^st word if more than N

According to another embodiment, the system may be designed to consider a predefined number of words for the combined predicted word, after which if the first word of the combined predicted word is not a correct word, the user may proceed to entering a next new word. In this case the system does not consider the first word of the combined predicted word, but it changes it to provide a new word other than those proposed relating to its (first) position within the combined predicted word. By doing so, at some point the system may correctly predict the desired word for said position, and the user may confirm it to enter it.

Options for the first word being entered;

When a first word of the current predicted word is being entered, there may be different methods to predict said word:

Method 1 - the system proposes the word having the highest priority regardless of the previous accepted word in the text.

Method 2 - the system combines the input information provided for the current word and the preceding confirmed word to provide a combined information, and compares the it with the combined words in the corresponding group of the N-gram database, and provides the corresponding combined word having the highest priority in the group. If there is no

corresponding combined word in the group, then the system behaves like in Method 1.

Method 3 - the system considers the word having the highest priority based on the input information provided for the current word and combines it with the preceding confirmed word/s to provide and considers a temporary combined word, if the temporary combined word has significantly lower priority relating to another combined word having the highest priority among the other combined words in the corresponding group of the N-gram database or does not exist within said group, the system may preferably propose said another combined word, if not, the system may propose said temporary combined word.

According to one embodiment of the invention, two more methods of inputting a word using the accept/reject methods may be considered

Option 1 - according to one embodiment, the system always combines the input information corresponding to the current word being entered with the previous accepted word/s and uses a corresponding N-gram database to predict a combined word. If there is not a previous word to combine, then the system predicts one word accordingly, and eventually, if the user provides a next word, the system may consider said two words to predict a combined predicted word accordingly.

Option 2 - according to one embodiment, the system may consider the previous accepted word only if the current word being entered as a single word (e.g. having the highest priority in its corresponding group) is not the desired word. According to one embodiment, the procedure of combining a word with a preceding word starts when the first rejection occurs. Said procedure may end when the there is no word in the current word. According to another embodiment, the procedure of combining a word with a preceding word starts when the first rejection occurs and a second word of the combined predictive word is being entered. Said procedure may end when the there is no word in the current word

According to one embodiment of the invention, if the current combined word has two words (or more), and both words are non-correct/desired, then:

Option 1 - correct-and-accept the 1^st, word, then eventually accept/reject the 2^nd word, and so on.

Option 2 - Reject or accept the 2^nd word. Preferably, the system may be designed such that to modify the 1^st word accordingly. Then, reject, modify and/or accept the first word.

Option 3 - Confirm the last word, then the one before, and so on. According to one embodiment, if the user corrects/accepts the I^s word of a combine word, the system may consider the 2^nd word of the combined predicted word to propose another combined predicted word. If there is no 2^nd word, then the system may predict the 1^st word based on the 1-gram, database.

According to one embodiment of the invention, if one of the single words (other then the first word) of a combined word is not correct, then, the rejecting action may be applied to the whole combined word, in such case, the system may propose another combined word from the corresponding group of the corresponding N-gram database. According to one method, by doing so, the system may even replace a correct single word of the previous combined word by a non- correct word of said another combined word.

The words of an N-gram database may have a value corresponding to the corresponding letter-key presses for entering them. According to one embodiment, the single words of the N- gram database may be separated by a value other than the letter key values, for example, by the value of the key to which the space character is assigned (e.g. value 6).

According to one embodiment of the invention, the combination of two single words each having the highest priority in their respective group within the database, wherein said combination may not be within the corresponding N-gram database, may be included within said group and a low value of frequency (e.g. 0 ) may be assigned to said combination.

According to one embodiment of the invention, when a user corrects a predicted word for example through the Correction procedure of the invention, the system considers said correction information and preferably also considers a corresponding 2-gram (e.g. or another

corresponding N-gram) word to preferably predict corresponding 2-gram (or N-gram) word.

Another embodiment

According to one embodiment of the invention, when a first word (e.g. of potentially, combined predicted word) is predicted, there may be two possibilities:

- if said word is the desired word, the user may accept it by for example providing a space character, and the system enters said word.

- if the proposed word is not the desired word, then, the user may reject it, and the system may propose a new word, and, preferably, provides a space character at the end of the new proposed word. Now, there may be two scenarios:

1- If the new proposed word is the desired word, the user may confirm it by for

example, providing a confirmation interaction such as, for example, providing a gliding action in a predefined direction. As an example, said gliding action may be a gesture provided on/from the space key rightward.

2- If the new proposed word is the desired word, the user may proceed to entering a next desired word, wherein the system considers the corresponding predicted word as the second predicted word of the combined predicted word. According to one embodiment, each time the user provides an input information such as a key interaction corresponding to one of the (e.g. additional) letters of the second predicted word, the system may propose a new word (e.g. preferably, a word other than the word(s) already proposed/predicted) corresponding to the input information relating to the first desired word, replacing the first predicted word of the combined predicted word. Preferably, (e.g. simultaneously), the system may also propose a second word corresponding to the provided input information related to the second word. According to one method, if during the entry of the second word the system does not succeed to predict the desired first word, the user then may proceed to entering a following word and so on.

After entering the information corresponding to entering the second word, if the system predicts the second word correctly, the user may accept said word by a providing a predefined interaction such as providing a space character. On the other hand, if the proposed second word in not the desired word, the user may reject it, by providing a predefined interaction such as that described earlier.

Preferably, each other interaction, such as accepting or rejecting interaction corresponding to a (e.g. second, third, etc.) predicted word, may cause the system to propose another first word (e.g. preferably, a word other than the word(s) already proposed/predicted) in that position (e.g. words already proposed as the first word) first word) within the combined predicted word. At any moment, when the system proposes/predicts the desired first word, the user may, preferably, confirm it (e.g. by providing a confirming action such as that described above). According to one embodiment, after entering all of the characters of the second word, if the first predicted word is still not the desired word, the user may proceed to entering the next desired word, and the system and/or the user may behave as described during the entry of the second word, and so on.

According to one embodiment, accepting a word other than the first word, may cause the system to show that word differently within the combined predicted word. This way the user may better see the rejected words of the combined current word. When the system proposes the first predicted word correctly, and the user confirms it, in addition to entering said first word of the current combined predicted word, the system may also enter the following consecutive accepted words of the current combined predicted word.

As mentioned before, preferably, in order to predict words, the system may preferably use preferably, 1-gram, and/or at least 2-gram word databases.

According to one method, if the system uses 1-gram words and 2-gram words databases, then preferably, the system at first may use the 2-gram database of words to predict words, and if it does not succeed to predict the desired word, then, the system may use the 1-gram database to predict a word such as the first word and/or the current word of the combined predicted word.

It must be noted that the system may use any N-gram word database to enhance prediction.

Preferably, a 2-gram word database may be used to provide any N-gram words. This is because, any N-gram word (except the 1-gram word) is generally comprised of several 2-gram words. With the continuous description of the current embodiment, an example is provided hereafter. Note that the term "combined predicted word" refers to one or more words being predicted by the system, wherein the first of said words is still not confirmed or accepted by the user as being the desired word of the user.

Fig. 384, shows as an example, various portions 384100, 384200 (including several portions 384201-384204), 384301, and 384400 (including several portions 384401-384410), of an exemplary 1-garm and 2-gram word databases used by the system. As an example, in order to enter the sequence of words "we held to", the user may first enter the input information corresponding to the word "we". The system may propose the word "as", as the first word of the combined predicted word, said word preferably having the highest priority in the portion 384100 (As described before, this is an exemplary database, and the frequency of use of the words relating to each other mat vary depending on the corpus used or the location of the words of the group within a phrase, etc.). This is not what the user desires, in this case the user rejects the proposed word. Now, the system proposes another word "he" preferably continued by automatically adding a space character. This is still not the word desired by the user. The user may proceed to providing the input information (e.g. pressing the appropriate key) for entering the first character of the next (e.g. second) word of the combined predicted word. Not confirming the first predicted word may result in rejecting said first predicted word of the combined predicted word. Now, in addition to predicting the first character of the second word of the combined predicted word, the system may also try to propose another word instead of the first word of the combined predicted word which was rejected.

By considering the appropriate portion 384201 of the database, the system may propose the combined predicted word "me a". In this example, the 2-gram "as a" having the highest priority is not proposed because the word "as" has already been rejected by the user as the first word of the combined predicted word. The word "me", now being the first word of the combined predicted word is not what the user want. The user may indirectly reject it by ignoring it (e.g. by not confirming it when it is shown to him) and proceed to providing the input information (e.g. pressing the appropriate key) for entering the next character of the second word of the combined predicted word.

By considering the appropriate portion 384202 of the database, the system may propose the combined predicted word "we as" because the first words (e.g. "as", and "me") of the 2-gram words "as he", "as we", "me as", and "me he", are words that are already being rejected (and/or ignored) by the user. The first word "we" of the new combined word "we as", is what the user wants. The user may confirm it by providing a predefined interaction such as that described above. The system enters said word.

Now, the second word "as" becomes the first predicted word of the combined predicted word. The user may proceed to entering the remaining input information (e.g. key presses) corresponding to the desired word "held". By considering the provided input information corresponding to the word "held" and the confirmed/entered word "we" and the appropriate portion 384204 of the database, the system may predict the word "need". This predicted word is not what the user wants. The user may reject the proposed/predicted word. The next 2-gram word of the portion 384204 that starts with the word "we" is the 2-gram word "we-also". The system proposes the word "also", preferably, followed by an automatically added space character. This is not what the user wants.

The user may then proceed to entering the input information corresponding to the first character "t" of the following word "to". By considering the combined input information, such as information relating to the rejected predicted words "also" and/or "need", the space character, and the input information corresponding to the first character of new second word of the new combined predicted word, and also considering the corresponding portion 384301 of the database, (and the fact that the word "also" has also already been rejected), the system may either wait until the user enters the input information corresponding to the next word of the second word of the current combined predicted word, or it may consider a 2-gram (or more) word of the database having longer length wherein its beginning portion corresponds to the input information provided by the user. Said considered 2-gram may be the corresponding 2- gram having the highest priority.

As an example, by considering the portion 384400 having several portions 384001-384010, the system may propose the word "keep" as the first predicted word of the combined predicted word, which is the first word of the 2-gram word "keep the" which has the highest priority after the 2-gram words ""need to", and "need for", which in this case are discarded by the system because the word "need" has been rejected by the user. In the current combined predicted word, preferably, the first word "keep" is continued by the characters " t" which corresponds to the corresponding continued characters of the 2-gram "keep the". The new word "keep" is still not the desired word. The user may reject it, by not confirming it and proceeding to providing the input information corresponding to the next character "o" of the second word "to" of the combined predicted word.

By considering the corresponding portion 384403 of the dictionary and the combined

information, the system may now propose the 2-gram word "help to". The current word (e.g. second word of the combined predicted word) "to" is correct. The user may accept it by for example providing a space character.

The first word "help" of the combined predicted word is not what the user wants, therefore according to one method, he may proceed to entering the third word of the combined predicted word, when the user provides the first character of said third word, the system may change the first word of the combined predicted word by replacing it by the word "held" which is the first word of the 2-gram word "held to" of the appropriate group 384401 of 2-gram words

corresponding to the appropriate input information (e.g. key presses and rejected words corresponding to the first word, and accepted word "to") provided by the user, and having the next highest priority in the corresponding group of 2-gram words. Now, the first word of the combined predicted word is what the user wants. He may confirm it. Confirming the first word as this example, results to entering said word "held" and the following accepted word "to". The first character of the last word of the combined predicted word, now becomes the first character of the first word (e.g. to be predicted) of the new combined predicted word. The example provided above, was specifically selected by the inventor to show that even the low priority words of a group of words having many words with the same key press information, may be selected automatically by the system during text entry. This permits to significantly reduce the number of correction procedure (e.g. or maybe eliminating the correction procedure) during text entry. In many cases the user may enter an entire text without providing correction (e.g. except maybe the last word in the text, that is not continued by an additional word).

As mentioned before, preferably, three (e.g. simple) different interactions may be required during the entry of a text:

1. An interaction (e.g. providing a space character) for accepting the last predicted word. 2. An interaction (e.g. providing a gliding action on/from the Space key leftward) for rejecting the last predicted word.

3. An interaction (e.g. providing a gliding action on/from the Space key rightward) for confirming a desired word replacing a rejected word (e.g. the first word of a combined predicted word).

Potentially, the accuracy of prediction may reach 100%, except for the words that are not included within the database(s) used by the system. According to one embodiment of the invention, the N-gram (e.g. 2-gram) database used by the system may include bi-gram words ignoring some of the (very) common words such as "the", "a", "of, etc. For example, the 3-gram phrase "to the mall" may be reduced to "to mall", or the 4-gram phrase "going to the mall" may be reduced to "going mall". Accordingly, for one method, during the entry of a word having at least one consecutive preceding common word(s) (e.g. preferably adjacent to the word being entered) such as "to the", the system may ignore the word "the" and/or "to the", during the search in the database. This may augment the accuracy of the system.

It must be noted that, preferably, a predicted word rejected (e.g. or not accepted) by the user, may not be considered during search and may not be re-proposed to the user, unless the user forces the system by, for example means of Correction.

Preferably, a correction procedure provided by the user may be related by the system to the first word of the combined predicted word.

It must be noted that the combined predicted word may include one or more predicted words.

It must be noted that during the entry of a word, the user may use the features of the invention such as Correction, Insertion, Next (e.g. confirming a letter of an undesired predicted word as being the desired letter in a specific position of the desired), etc., to help the system to better predict a word.

Preferably, a predicted word may be accepted or rejected before a special character is provided at the end of said word. If the predicted word is rejected, then the special character may also result to proposing/predicting a new word to the user by the system. In general, according to one embodiment of the invention, any interaction, except the accepting action, (e.g. entering a new letter, a special character, etc.) provided by the user during the entry of words, may permit the system to predict/propose another word instead of the preferably, the first rejected word of the combined predicted word. Said interaction may be a key interaction or another predefined types of interactions such as a command or mouse function.

It must be noted that, although the confirmation is described to preferably being related to the first rejected word replaced by a desired word, according to one embodiment of the invention, the user may confirm any of the replaced and/or rejected words as his discretion.

It must also be noted that, when the system replaces a rejected predicted word by a new predicted word, the system may also replace other rejected words of a combined predicted word accordingly.

According to one embodiment of the invention, if the system proposes a last choice relating to the user's input information from the database, and the user rejects it, the system may alert the system that his desired word is not in the dictionary, in which case, the user may enter his desired word by precise characters. In order to enhance the efficiency of the search, the system may use only a (first) portion of the database beyond a predefined rate of frequency of words. In this case, if the system does not find the user's word, in the portion of the database beyond the frequency rate, then the system alerts the system that his word is not in the system, and when the user begins to enter his desired word by entering precise character, the system may user the other portion (including words lower than the predefined frequency rate) of the dictionary to eventually predict the user's word without user's knowledge of such action.

It must be noted that, in the embodiment described, any to a new proposed word replacing a rejected word, a space character may be added automatically. According to one embodiment of the invention, a help window may be pointed to preferably the first rejected word. Note that, in the example of fig. 384, the words of each of the 2-gram words are being separated by the character "_" as an exemplary separating character. Preferably, said separating character may be a space character. Hereafter the principles of such a system are described. Fig. 385 shows an exemplary diagram demonstrating some of the general principles of this embodiment of the invention. Given a situation where the user enters an individual word and where there is no input information relating to a combined predicted word (e.g. the user enters a word, which may become the first word of a combined predicted word, or which may be considered to be the first word of a combined predicted word), and where a variable, for example NW, designates the position or index value of a word (possibly in a combined predicted word), and where another variable, for example NL, designates the position or index value of a character or letter in any word

(preferably not a combined predicted word). Given the described situation where NW=0 (no words in the combined predicted word) and NL=0, the system (i.e. and implementation of the embodiment) may be considered to be in the Start State (38501) of the embodiment.

After the user has entered (38502) the first word the system will set the value of NW to 1 (38503). Now the user may choose (38504) to (a) accept the first word and consequently, the word-state of the first word may be considered as entered such that the first word is not considered part of the combined predicted word (38522) and then the system will set the value of NW to 0 (38523) and finish at the exit state (38524), or (b) reject the first word and consequently, the word-state of the first word may be considered as rejected (38505). The system may now change the first word (38506), after which the user may choose (38507) to (a) confirm the first word and consequently, the word-state of the first word may be considered as entered such that the first word is not considered part of the combined predicted word (38520), after which and based on the value of NW (38521), the system will either (1), in the case that NW=1, then set NW to a value of 0 (38523) and finish at the exit state (38524), or (2), in the case that NW does not have a value of 1, the system will increment the value of NW and initialize the value of NL to 0 (38508) where the system may then continue to step 38509 in fig. 385, or (b) reject the first word in which case the system will increment the value of NW and initialize the value on NL to 0 (38508) where the system may then continue to step 38509 in fig. 385. Continuing from step 38508 (where the user will begin to type a new word) or from step 38513 (as the case may be) in fig. 385, the user may type the next letter (38509), where upon the system may increment NL by a value of 1, and such that this next letter forms part of the last (current) word of the combined predicted word. Now, the system may change (38510) the first word of the combined predicted word, after which the user may choose (38511) to (a) confirm the first word and consequently, the word-state of the first word and any words with accepted- state that follow consecutively from the first word may be considered as entered such that these entered words (where the quantity of these entered words may, for example be represented by a variable, N) are not considered part of the combined predicted word (38517), after which the next consecutive word with rejected-state, if any, or the current word (where no rejected word exists in the current combined predicted word), being typed, may be considered to be the first word of the current combined predicted word (38518), after which the system will may decrement the value of variable NW by the value of variable N (38519), after which the system will continue to step 38512 in fig. 385, or (b) reject the first word in which case the system the system will continue to step 38512 in fig. 385.

Continuing from step 38519 or from step 38511 in fig. 385, the user may (38512) either (a), in the case where the variable NW has value 1, type or continue to type the first word of the combined predicted word (38502), or (b), in the case where the variable NW does not have value equal to 1, consider the position of the last typed letter (38513). Where upon, the user may either (a), in the case that the last typed letter is not the last letter in the last word, continue to type the next letter (38509) where upon the system may increment NL by a value of 1, and such that this next letter forms part of the last (current) word of the combined predicted word and then the system continues through step 38510 and onwards as described above, or (b) choose (38514) to (1) accept the last word and consequently, the word-state of the last word may be considered as to have accepted-state (38515), or (2) reject the last word and consequently, the word-state of the last word may be considered as to have rejected-state (38516), such that in either case (i.e. rejected last word, or accepted last word) the system may now change the first word and possibly any words with rejected-state (38506), after which the user may choose (38507) to continue with the system as already described.

Note, that any word in a combined predicted word may preferably have one of two word-states, either accepted state or rejected state. - After rejecting a word, and accepting the new word (by pressing SP key), the system does not provide a new SP and may consider the accepted word for predicting the next word

As mentioned, when the user enters the information (e.g. including key presses) corresponding to the first word of a combined predicted word and the user rejects said first word, the system may propose a new word and preferably automatically adds a space character at the end of the new word. According to one embodiment of the invention, the user may accept said new word in the same manner (e.g. by pressing/tapping the space character) as a first word would have been initially accepted. If the accepting action is provided by pressing the space key, then preferably, said action may be considered as a confirming action relating to the new predicted word, preferably, without adding a space character to the space character (e.g. automatically) provided after rejection. Preferably, a word accepted this way may be considered in the prediction of the next word that may be entered. Do not confirm, does not affect prediction based on previous

According to one embodiment of the invention, special characters/functions (e.g. entered precisely) act as an end-of-the-word signal but do not confirm a predicted word. As an example, after the system predicts a word, and the user rejects said word and the system may predict a new word, the user may provide a special character or a function, and proceed to enter a next word. The system may use the entry of information (e.g. key presses) for entering one or more characters of the next word in order to propose one or more new words instead of the rejected word.

According to a first embodiment, if a special character/function is to be entered at the end of a predicted word, the user may first enter said special character/function regardless of said predicted word being the desired word or not, before accepting or rejecting said predicted word. If said predicted word is what the user wants, he may accept it, if not, he may reject it. The system may replace said predicted word by a new predicted word, and may automatically add a space character at the end of the new predicted word.

According to a second embodiment, if a special character/function is to be entered at the end of a predicted word, the user may first accept or reject said word, and then provide the special character/function. In this case, special character may be inserted before the space character that may be included at the end of the word as a result of rejection. According to one embodiment, after a word is rejected, as a user provides each of the special characters/functions of a sequence of special characters/functions, the system may replace a predicted word by a new word predicted. The system may insert said special character(s) at the end of the new predicted word(s). At any moment during the entry of said special

characters/functions, if the predicted word is the desired word, the user may preferably accept/confirm the predicted word, and may then continue to provide the remaining special characters/functions. According to one method, the special characters/functions may be of many kinds such as a dot, a comma, a "?", a CapsLock function, a Shift function, etc.

By using the embodiments just described, as an example, entering the first word of a new phrase/paragraph may result in prediction of one or more new words replacing a rejected previous word (e.g. the last rejected word of the previous phrase/paragraph).

Other enhancements may be considered to augment the accuracy of the system. As an example, the value of the frequency/priority of a word may vary depending on the word, the character, or the chain-of -characters preceding it. For example, the frequency of a word "mg" may be higher than any of the words "as", "we", or "he", (e.g. having the same key presses) if it is provided after a number (e.g. 240 mg).

According to one embodiment, different methods/features may be used such that the system better present the first word of a combined predicted word having several predicted words, to a user. As an example, the first word may be shown in another color relating to the other words of the combined predicted word, the first word may be printed in a larger-font relating to the other words of the combined predicted word, the first predicted word or all of the words of the combined predicted word may be spoken/read-aloud to the user by a text-to-speech system, etc., each time the system predicts a new first word. For example, according one method, each time the system predicts a new first word of a combined predicted word, the system may show it next to the finger of the user interacting with a zone such as key interaction relating to a character of the new word being entered, or in another predefined location. According to one embodiment of the invention, the embodiment described above may be restricted to a combined predicted word that preferably may have up to two predicted word. In such case, according to one method, the confirmation procedure and accepting procedure of a predicted word may both be provided by a same predefined interaction such as preferably a tapping/pressing action on the Space key. This may simplify the user's interactions with the screen and the system.

In the current embodiment, if the combined predicted word has a single word which is rejected and a special character such as a dot/period is entered after it, according to a first method, in order to avoid ambiguity, any additional special character including a space character provided after that and before beginning to enter a next word, may not result in predicting another word by the system instead of the rejected word. In this case, preferably, after the user enters the first character of the next word, the system proposes/predicts another word. According to a first method, each special character (e.g. excluding a tapping/pressing action on the space key) may result in predicting another word by the system instead of the rejected word. In this case, preferably, in order to confirm the desired word, the user may provide a predefined interaction such as a tapping/pressing action on the space key.

It must be noted that the interactions, features, embodiments, methods, and related drawings and flowcharts, etc., using to describe the linguistic rules described above, have been described and shown as examples to show the basis of the principles used to increase the accuracy of the word predictive data system of the invention. Said interactions, features, embodiments, methods, and related drawings and flowcharts, etc., may be combined partially or totally, or in some cases replace or swap each other by people skilled in the art. As an example, the interactions used for accepting and rejecting a word may be swapped. According to another example, in order to predict a word, instead of:

- referring to the words of the two-gram database wherein the length of said words correspond to the number of key presses provided by the user, then referring to the words of the two-gram database wherein their beginning characters correspond to the number of key presses provided by the user, and then referring to the words of the one-gram word database;

the system may:

- first refer to the words of the two-gram database wherein the length of said words correspond to the number of key presses provided by the user, then the system may refer to one-gram word database(s), and then again refer to the words of a two-gram database wherein their beginning characters corresponds to the number of key presses provided by the user.

Creating text and sending it to an applications

According to one embodiment of the invention an application can be designed that a user can create a content such as text in a unified window and send it to another application to be used by this other application. According to one method, the sending process may be initiate by an interaction such as a gesture interaction in a predefined direction relating to said other application. Preferably said gesture can be provided from anywhere (e.g. on a touch screen) in said direction, providing gesture from anywhere towards different predefined directions relating to symbols or functions have already been described throughout this and previous applications filed by this inventor. As an example, after the user creates the said content a gliding action towards a first direction may send said content to a Facebook application and a gliding action in a second direction may send said content to an email application. Note that sending a content such as text can be applied to any type of content, either created by the user sending the content or with reference to another source.

Capitalized Wordlist

The system may use and/or include additional features to enhance the accuracy of prediction. According to one embodiment of the invention, the system may include/use a wordlist wherein preferably each of the words of the wordlist includes at least one capitalized letter. The capitalized letter(s) may be any of the letters of the word. The words of the wordlist may be part of the words of the database of words of/used-by the system that includes words wherein none of their letters is capitalized (a "mixed database"), or preferably the words of said wordlist may form a separate database of words (for ease of explanation, the separate databases are herein referred to as "capitalized" and "non-capitalized" databases).

According to one method of the use of such database, when a user provides a predefined interaction during the entry of information corresponding to a desired word, the user may inform the system to refer to the words of the database that include capitalized letter(s) to search for a corresponding word based on the input information provided by the user relating to said desired word. As an example, by considering that the system may use separate databases as described above, when the user desires to enter the word, "John", the user may first provide information such as a "shift" function to inform that the next letter to be entered is a capitalized letter. The user may then provide the input information (e.g. key presses) corresponding to at least some of the letters of his desired word. In this case, where said "shift" function is provided, according to one method, the system may first refer to the words of the database that includes capitalized letters to search for a corresponding word and to predict a word. According to one method, if the desired word is not one of said words of that database or the system does not find a

corresponding word in said database, the system may refer to the words of the dataset that do not include capitalized letters. It must be noted that the "shift" function may be used for any of the characters of the desired word.

With continuous description of the current embodiment, according to one method, the system may include N-gram (e.g. 1-gram, and 2-gram, etc.) wordlists wherein at least one of the words of an N-gram word includes at least one capitalized letter.

It must be noted that, an identical word may be included in at least one of a non-capitalized- database and a capitalized-database, wherein in the non-capitalized database all of the letters said word are in lowercase, and in the capitalized database at least one of the letters of said word is capitalized. Accordingly, when a user and/or the system, adds a new word, wherein at least one of its letters is capitalized, to the capitalized database, the user may manually, or the system may automatically, may also add said word, wherein all of its letters are in the lowercase, to the non-capitalized database.

According to one method, the system may predict/present to the user a word of the capitalized database where said word includes a capitalized letter in a position within said word as it is defined in the database, even if the user provides capitalizing information corresponding to a letter in a different position within said word. As mentioned, providing an interaction such as a "shift" function may refer the system to the capitalized database wherein its words include at least one capitalized letter. According to one method, a single shifting action may be enough to search and present, as is, a corresponding word from a capitalized database, regardless of the number of capitalized letters in said word. - Apostrophe

According to one embodiment of the invention, in some languages such as French or Italian, a character such as an apostrophe is used between two words to, in general, connect them to form of a single word (i.e. "connected word"). Words known as "contracted form" words are an example of such words (e.g. "c'est") wherein by connecting such words to each other at least one of the words is contracted. Preferably, the apostrophe may be part of the first word (e.g. " ") of the contracted word. Preferably, a connected word may be a word of the 1-gram word database. Alternatively, the apostrophe may be part of the second word (e.g. " 'est ") of the connected word. Alternatively, such connected word may be a word of the 2-gram word database. Alternatively, the apostrophe may not be included in any of the first word or the second word of the connected word.

According to one embodiment, a connected word may be considered by the system as a 2-gram word. Accordingly, for entering a connected word, the system and/or the user may, preferably, follow the principles of entering a word using N-gram databases as described herein. As an example, a user may first enter the first word of a connected word by providing the input information (e.g. key presses) corresponding to the letters of said word, and preferably provide precisely (e.g. by an appropriate gliding action) an apostrophe. According to this embodiment, the system may search for a word corresponding to the combined input information (e.g. key press and an appropriate gliding action) provided by the user. By entering precisely the apostrophe the system may be informed that entering (e.g. the information corresponding to) the first word is completed. The system may search for a corresponding word (e.g. preferably, including the apostrophe) and if it finds such word, it may propose it to the user. At this time, two scenarios may be considered:

1) If the proposed first word is the desired word, the user may proceed to enter the second word of the connected word, preferably without providing a confirmation interaction such as pressing on the space key. Such action may inform the system that the proposed first word is confirmed, and the system preferably enters it. In the case that the system requires a confirmation interaction such as pressing on the space key, the system may preferably not provide/present a space character.

2) If the proposed first word is not the desired word, the user may reject it by providing a predefined interaction such as a gliding action as described before. According to one method, the system may propose another word corresponding to the information provided (e.g. including the apostrophe). There may be two scenarios:

a) If the new proposed first word is the desired word, the user may confirm it by providing a predefined interaction such as pressing on the space key. The system may enter said word, preferably, without including a space character at its end.

b) If the new proposed first word is not the desired word, the user may proceed to enter the second word. The system may propose a different first word for each interaction (e.g. tapping action on a letter key, or other interactions such as those described before, etc.) during the entry of the second word until the system proposes the desired first word, in which case, the user may confirm the first word as described before, and may continue to enter the information corresponding to the remaining characters of the second word (e.g. which now may eventually become the first word of a combined predicted word). In the case that both the first and second words of the connected word are not the desired words, according to a preferred method, the user may proceed to correct the first word, and confirm it.

Note that any database of words of/used by the system may include any symbol such as letters, special characters, or icons representing characters and/or functions, etc.

Auto spacing

According to one embodiment of the invention, during the entry of a combined word, in some cases an auto-spacing method may be implemented within the system. As an example, after rejecting a proposed word by providing a predefined interaction such as a gliding action in a predefined direction, preferably, on/from the space key, if the user proceeds to entering a next word, the system may automatically add a space character at the end of the last word of the combined predicted word.

It must be noted, that the principles that will be described hereafter, may be applied to any word predictions system, including those using an ambiguous keyboard (e.g. a key includes several letter/characters) , or those using a keyboard wherein generally each of its keys includes a single letter/character (e.g. QWERTY keyboard, AZERTY keyboard, etc.).

According to one embodiment of the invention, when the user enters input information (e.g. key interactions) corresponding to the first word of a chain f words such as a phrase, a sentence, etc., the system may first refer to an N-gram (e.g. 2-gram) word database to select and/or predict a corresponding word. If the predicted and/or selected word is not what the user desired to enter, then the system may operate according to at least one of the procedures (separately or combined), below: (a) the system may look for the next corresponding word(s) within the N-gram (e.g. 2-gram) word database, (b) the system may look for the corresponding word(s) within a 1-gram word database.

According to another embodiment, when the user enters input information (e.g. key interactions) corresponding to the first word of a chain of words such as a phrase, a sentence, etc., the system may first refer to a 1-gram word database to select and/or predict a corresponding word. If the predicted and/or selected word is not what the user desired to enter, then the system may operate according to at least one of the procedures (separately or combined), below: (a) the system may look for the next corresponding word(s) within a 1-gram word database, (b) the system may look for the corresponding word(s) within an N-gram (e.g. 2-gram) word database.

It must be noted that according to one embodiment, at any moment, during prediction and/or selection of a word corresponding to the input information provided by the user, the system may navigate between one or more different N-gram (e.g. 1-gram and 2-gram) databases. As an example, after the user provides the input information corresponding to the desired word, the system may predict and/or select the first word of one or more of the corresponding 2-gram entries in the corresponding 2-gram database. After the system proposes a number of the relevant words preferably all of the relevant words, where none of them are confirmed by the user (where the relevant words correspond to the input information from the 2- gram database) the system may refer to the 1-gram database to propose the next word(s) from the 1-gram database.

The embodiments just described may be used with any of the word predictive embodiments such as those using "Reject" and/or "Accept/Confirm" procedures.

Intentional Gliding Actions

As mentioned before, gliding actions explained throughout this application and the previous patent applications filed by this inventor, may be provided in the air. According to one embodiment of the invention, in order to differentiate between the gliding actions provided intentionally from those provided unintentionally, the user may provide (prior to, after or simultaneously with a gliding action) a predefined interaction to inform the system of at least an intentional gliding action. It must be noted that the gliding action provided by the user may be detected by a motion detection system integrated within a device used to provided the gliding action or without a device where, in this case, the gliding action may be detected by a (e.g. an external) system such as a camera. As an example, if the user uses a device such as a remote control where the motion of the device can be detected by a means (e.g. infrared fields, ultrasonic detection, accelerometers, etc.), the user may press a key preferably integrated within the device to inform the system of an intentional gliding action (e.g. prior to, after, or simultaneously), and provide a gliding action using the device.

As mentioned in this application, during the entry of a current word, considering the input information corresponding to the current word as a last gram of an N-gram word of the N- gram database, and considering one or more previous word(s) as the initial gram(s) of the N- gram word, may significantly improve the accuracy of predicting the current word. According to a preferred embodiment, when the system searches for an N-Gram in the corresponding N-Gram database (wherein N>0, preferably N>1), the improved accuracy of word/sequence prediction may be explained by the fact that the system processes a combination of ambiguous key presses and precise characters. According to one scenario, the higher the ratio of precise characters in the N-Gram, the more limited the number of interpretations of the remaining ambiguous key presses, thus limiting narrowing the number of matches in the N- Gram database, which may result in improving the prediction accuracy.

For the case of using an N-gram database to predict a word, when a user provides input information (e.g. wherein at least some of said information is considered to be ambiguous) corresponding to a desired word, the system preferably considers at least one of the preceding words/chain of characters in the document and/or text entry field, etc., as precise information (e.g. the characters of a word/chain of character already entered are considered as precise characters) relative to the word being entered, and combines said precise information with the information relating to the current word being entered and searches in the N-gram database for entries corresponding to the combined information.

According to one method, word(s) preceding a word being entered may be at least one adjacent word. According to another method, the adjacent word(s) preceding a word being entered may be word(s) that precede the current word by one or more predefined words between said adjacent word(s) and the current word in the document.

It must be noted that a preceding word may either be entered by the user or it may be part of a document not created by the user (e.g. opened by a user) and wherein after the user positions a cursor on/at the end of said word and begins to enter a current word, the system considers said word and/or a another previous word a as preceding word.

According to a preferred embodiment, the system may search within the N-gram (e.g. 2-gram) database for several types of words, such as below:

- at least one word of a first type of words, wherein all of its characters correspond to the input information provided by the user; and

- at least one word of a second type of words wherein its beginning characters correspond to the input information provided by the user;

According to one method, the system may preferably propose/present one word from the first type of words and/or one word from a second type of words to the user.

Correct after reject

According to one embodiment of the invention, a system may be designed based on the principles just described (e.g. the beginning one or more grams of an N-gram are considered as a (one or more) precise chain of characters/information by the system, and followed by the input information corresponding to the current gram/word being entered). When a user provides the input information (e.g. key presses) corresponding to a desired word and the system predicts a word and the user rejects the/said predicted word, the system may predict a new word (e.g. instead of the previous word) based on the principles of the data entry system such as the principles of using an N-gram dictionary (e.g. 2-gram dictionary). Furthermore, according to a first method, the system may preferably automatically switch to Precise Character Mode by switching the keys of the first keypad of the invention into their Precise (e.g. Letter) Mode instance (e.g. in Precise Mode, each of at least the letter keys of the first keypad may be considered as/replaced by a second keypad of the invention). At least two scenarios may be considered, (a) if the new predicted word is the desired word of the user, then, the user may accept/confirm said new predicted word (e.g. by providing an end-of-the-word signal such as pressing on the Space key), or (b) if the new predicted word is not the desired word of the user, the user may proceed to the Correction procedure by precisely entering at least one of the characters such as at least the first character of the desired word by interacting with the appropriate key (as described throughout applications filed by this inventor), for example, by tapping on the zone assigned to an appropriate identified character on a key of a first keypad in the Precise Mode (e.g. as mentioned before, in the Precise Mode, a key of the first keypad may be divided into a plurality of zone/keys each corresponding to an identified character. In the Precise Mode the plurality of divided zones/keys of the first keypad may be considered to form a corresponding second keypad. Therefore, tapping on a zone of the first keypad may be considered as tapping on the corresponding key of the corresponding second keypad.), or alternatively, by providing a gliding action on said key of the first keypad (i.e. when a gliding action begins on a first keypad of the invention, the system enters into the precise character mode, consequently, even if said gliding action is apparently provided on a first keypad (e.g. the system may not change the interface of the key), in reality it is provided on the corresponding second keypad) in the direction towards the appropriate identified character. By considering the input information provided until then (e.g. the key presses/interactions corresponding to the predicted word and the precisely entered at least first character) the system may predict a new word. According to one method, each time the user enters an additional precise character and/or, according to one method, interacts with the system, the system may predict a new word. The user may repeat the Correction procedure until the system presents the user's desired word. Accordingly, if the user's desired word is not in the dictionary, the user may repeat the

Correction procedure for all of the letters of the desired word. Alternatively, eventually the user may precisely enter (e.g. by the Correction procedure and/or the Insertion procedure) substantially all of the characters of the desired word.

As an example, by considering the keypad of fig. 386A, if the user desires to enter the word "we", he may press the keys 386002, and 386003. The system may propose/predict the word "he". The user then may provide the gliding action 386007 to reject said proposal. The system then, may propose a new word "as", and enters into the Precise Character/Letter Mode and changes the appearance of the keypad by entering it into the Precise Mode as shown in fig. 386B through the keypads 386101-386104. If the new predicted word is still not the desired word, then he user may tap on the specific key/zone of the second keypad corresponding to the letter "w" (e.g. the first letter of the desired word). By considering the combined input information provided until now, the system may propose the word "we", which is the desired word. The user may confirm the predicted/proposed word. Note that, in case the latest proposed word still is not the desired word, the user may precisely enter at least some of the remaining characters of the desired word by tapping on the specific keys/zones of the second keypad(s) (e.g. until the system provides the desired word).

As mentioned above, according to one method, one or more N-gram databases such as preferably at least a 1-gram and a 2-gram database may be used by the system to predict words. As mentioned before, when the user provides the input information corresponding to a desired word (e.g. the Current word), there may be at least two scenarios:

a) if there is a preceding/previous word(s) which is correctly entered/predicted (e.g. within the corresponding text), the system may preferably consider both the preceding previous word(s) preferably correctly entered/predicted combined with the input information

corresponding to the Current word, and may process (e.g. compare/refer, etc.) the combined input information with the corresponding N-gram (e.g. 2-gram) word(s) of the database used by the system, and may predict (e.g. base on the combined input information) a Current Predicted Word for/corresponding to said input information provided by the user.

b) if there is not a preceding/previous word (e.g. the provided input information may relate to the first word in a text/paragraph, or the provided input information is provided after an end-of-the-word signal such as a period/dot character, etc.), then according to one method, the system may preferably refer to or use another database such as an N-gram database (e.g. a 1-gram database, and/or a 2-gram, etc.) wherein their first word (e.g. the first gram of their entries) is considered as being the first word of, for example, a text field, and/or of a paragraph, and/or a sentence, and/or a specialized data entry method or entity, etc. Accordingly, the frequency of use and/or statistics of use of such N-gram words may be different to those used, for example, in the middle of a sentence.

According to one embodiment of the invention, an N-gram (e.g. 2-gram) word database may include an (e.g. one or more) N-gram (e.g. 2-gram) word(s) that may include a special characteristic such as having a (e.g. including one or more) special character. For example, a 2- gram word (e.g. of a 2-gram word database) may include a special character such as, a "." (e.g. dot), a "," (e.g. a comma), or a character or set of characters (e.g. "\", "\m" or CRLF) representing the "Enter" function/key of a keyboard, etc., as the 2-gram 's first and/or as the 2- gram's second word (e.g. gram). By considering the principles herein described, after a user provides the input information corresponding to a desired word, the system may consider such database to use to predict a word. This may greatly enhance prediction performance and accuracy when the system predicts a single word, depending on the word's location within a text.

With continuous description of the current embodiment, as an example, hereafter an exemplary portion of an exemplary 1-gram word database (portion 1.1) and an exemplary portion of an exemplary 2-gram word database (portion 2.1) are shown.

Key presses

1 224

, and 734731

i had 457245

i and 452011

. and 410234

Key presses . had 398021 224 . who 52056 and 5201500 , who 371 1 1 had 890123 , had 11789 who 87101 i who 5205

Portion 1.1 Portion 2.1

In the 1-gram database portion 1.1, the order of the words based on their frequency of use is as follows: "and", "had", "who". In the 2-gram database, if the first word of the 2-gram words (e.g. portion 2.1, note that the first words ".", and "," are represented by the key press value "1") is the word "i", then the order of the 2-gram words is as follows: "i had", "i and", and "i who". In this case, as an example, after entering the word "i" correctly, and providing the input information relating to the key press value "224", the system may first propose the word "had". If the user rejects said word, the system may propose the word "and". Accordingly, as an example, after entering the character "." (e.g. at the end of a sentence), and providing the input information relating to the key press value "224", then, by considering the previous word (e.g. in this example, the character "." as the first gram of a word of the 2-gram database used by the system) the system may first propose the word "and" (e.g. in this example the character dot may be considered as the first character/word/gram of three of the 2-gram words ". and", ". had", and ". who" of the exemplary portion 2.1). If the user rejects said proposed word, the system may propose the word "had". For entering the word "who", according to one method, the user may proceed to the correction procedure.

Accordingly, as an example, after entering the character/word "," (e.g. at the end of a word within a sentence), and providing the input information relating to the key press value "224", the system may first propose the word "and" (e.g. in this example the character comma may be considered as the first character/word of three of the 2-gram words of the exemplary portion 2.1). If the user rejects said word, the system may propose the word "who".

In the current embodiment, the N-gram (e.g. 2-gram) word database may include N-gram (e.g. 2-gram) words having a special character as one of their words/grams (e.g. preferably, as their first word/gram) representing the beginning of the text to be written in a text field, or beginning of a paragraph, etc.). For example, if the special character "\" represents the beginning of a paragraph and its key value is "1", then, the 2-gram words "\ and", "\ had", and "\ who", may be added to the portion 1.1 with a corresponding frequency of use of each of said words when they are written at the beginning of a paragraph. When a user enters one of said words, the system may know, for example, that the cursor is at the beginning of a paragraph (e.g. after a user interacts with the "Enter" key or provides a "Tab" function, the system may be informed that the cursor is at the beginning of a paragraph), and therefore, if the user provides the key presses corresponding to the key value "224", the system may refer to the 2-gram words of the 2-gram word database having the character "\" as their first word/gram, and wherein their second word/gram corresponds to the key press value "224".

By providing a word (e.g. a 2-gram) database based on the principles just described, the system may use only a single N-gram (e.g. 2-gram) word database to predict the desired word instead of using a group of word database(s) consisting of one or more word databases, such as, for example, a group consisting of a 1-gram and a 2-gram database. Several efficiencies and enhancements may now be incorporated into the system such that; for example, amongst others, memory usage, speed of search in the database, and simplicity of design may be enhanced.

According to one embodiment of the invention, an (N+X)-gram (e.g. 2-gram) word database may represent an (N+Y)-gram (e.g. 1-gram) word database where (N+X)>=(N+Y) and where X>Y. According to one method, preferably, the insertion of an (e.g. one or more, i.e. X number of) identified character(s)/word(s) may be inserted before and/or after a (e.g. one or more) database entry/entries of an N-gram (e.g. 1-gram) word database to form a (N+X)-gram (e.g. 2- gram, for X=l) word database. For example, an identified character/word (e.g. '-') or a predefined code such as for example an unused ASCII code, etc., may be inserted before each database entry in a 1-gram word database (e.g. such as Portion 1.1) to form a 2-gram word database. Preferably, the inserted character (e.g. in this case the first gram) may be delimited from the second gram by a delimiter such as a space character. In this example, the system may search the 2-gram word database to find a 2-gram entry where, preferably, the 2-gram database entry also represents a (e.g. one or more) 1-gram entry from the 1-gram database. Also, for example, the system may represent a 2-gram entry of a 2-gram word database as a 1-gram word alone (e.g. without a relationship to an adjacent word in a sequence of words such as a sentence or a paragraph).

Depending on characteristic(s) of a word, such as for example, the position of said word within a text, or which special character precedes said word, etc., said word may be considered as an N- gram (e.g. 2-gram) word. Accordingly, said word and its characteristic(s) may form an N-gram (e.g. 2-gram) word of a corresponding word database wherein, preferably, each of said characteristic(s) may be represented by one or more symbols as one or more grams in said Ingram. For example, given the insertion/entry by the user of an identified character/word/gram such as '?', T, "we" (e.g. as just described, in a bi-gram entry a (one or more) (special) character may be considered as a gram/word of said bi-gram word) followed by ambiguous input information corresponding to a desired word, the system may search for corresponding words (e.g. 2-gram words) in the database that include said identified character/word/gram and wherein their last gram/word corresponding to the input information provided by the user (e.g. it must be noted that some identified characters may represent the beginning of a sentence or a paragraph such as in the case of , ". want" or ", want" or "? want", "i want", and "we want") to predict said word.

According to one embodiment of the invention, preferably, for an (e.g. one or more) N-gram word database and input information (e.g. user input from the user's interaction with the system through tapping on zones/keys, sliding from zones/keys, rejecting, accepting or confirming words and/or system input information such as identified inserted characters/words or special characters of an entry in an N-gram word database) where the information may contain a (e.g. one or more) precise and an (e.g. one or more) ambiguous character(s), the system may predict the user's desired word(s).

According to this embodiment, for a scenario, where, for example, the input information may consist of a first set of characters (e.g. consisting of one or more characters) of an N-gram (e.g. 1-gram or 2-gram) that are precisely identified (e.g. ',', '#', 'a', Ί', 'that', 'that?) and a remaining set(s) of character(s) including/consisting-of at least one ambiguous character (e.g. wherein an ambiguous user interaction may be considered to represent any character in a set/group of character(s)), the system may search an N-gram word database to predict the user's desired word(s). Creating an N-gram (e.g. 2-gram) database (e.g. from a corpus of words) to be used by the system

Based on the principles described above, an exemplary N-gram database may be created (e.g. as a derivative of a corpus) to be used by the system is described below:

1. The database may include corpus-based N-Gram frequency list, wherein, preferably, N=2 (i.e. "bigrams"). The database may also include other N-Gram frequency lists, wherein N>0. For the purposes of this invention, a 1-Gram is either a word (token) a number (e.g. integer, fraction, etc.), a combination of characters and/or punctuation marks and/or special characters, or other symbol(s) or a sequence of symbols identified/counted as they may occur in a corpus (i.e.

without being divided into morphemes and/or in any other manner) which preferably are not space-delimited. Punctuation marks may also be identified as 1-grams. la. For the purposes of this invention, according to one method, "word" is any sequence that includes at least one character in an alphabet used by the language the frequency list refers to and preferably is not space-delimited. I.e. if a plurality of special characters/digits occur at any position in the word, and there is no space between the letters and the digits/special characters, such sequences are considered "words" and included into the count just as they occur in the corpus. According to a second method, "word" is any sequence of characters regardless of its characteristic (e.g. including letters, or excluding letters).

2. According to this principle, the contracted forms and possessive constructions ("I'm", "can't", "wanna", "cat's", "cats'", etc.) are preferably considered as single words/grams and may be referred to as undivided. Thus the frequency list contains entries like "cat's tail", "wanna do", "I'm here" etc.

3. By using the same principle, hyphenated words may preferaby remain undivided. E.g."socio- economic status" may preferably be considered as a 2-gram.

4. The count is preferably case sensitive. E.g. both "the south" and "the South" may occur in the list as separate entries. According to one method, it may not matter where in the word the upper case letter(s) occur(s), and how many upper case letters there are (e.g. words like "ASAP", "iPad" etc. may preferably be listed as they occur in the corpus).

5. According to one method, the bigrams may include punctuation mark characters, wherever they occur in the bigram-initial position.

E.g. sequence"Why? Because" may preferably produce bigram "? Because".

5a. According to one method, in addition, the punctuation mark characters may be considered as an integral part of the bigram-initial word and, as such, the bigram may be included into the bigram count.

E.g. "Hello! My", "really? We", etc.

5b. According to one method, in addition, the punctuation mark characters may be considered as an integral part of the bigram-final word and, as such, the bigram may be included into the bigram count.

E.g. "Hello My!", "really We?", etc. 5c. A sequence of two or more punctuation marks may be counted as it occurs in the corpus. E.g. "But why?! I think... Maybe..." may result in the following bigrams:

"But why"

"why?! I"

"?! I"

"I think"

"think... Maybe"

"... Maybe"

"... Maybe..."

6. The punctuation marks that have an opening and an ending or may be considered as pairs of symbols (e.g. ( ) [ ] "" { } < >) may preferably not be included into the count of the frequency list. Instead, they may function as boundaries that the count will not cross.

E.g. the sequence "some fruits (usually, apples) can cause" may result in the following bigrams: "some fruits"

"usually, apples"

", apples"

"can cause"

The following sequences, however, may preferably NOT constitute bigrams, and thus may not be included into the count of the frequency list:

"fruits ("

"( usually" [brackets preferably should not be included into the count]

"usually ," [comma appears in bigram final position]

"apples )"

"apples can"

"fruits usually"

7. The count may include numbers and special characters wherever they precede a word (i.e. are followed by a space, and then, by a word).

E.g. "5 o'clock", "24 years", "203 trees", "$20 US", "20$ in" etc. may be considered bi-grams. 7a. For this purpose, a "number" is a digit or any sequence of digits which are preferably not space-delimited, as well as any sequence of characters that is preferably not space-delimited and may consist of at least one digit and a special character(s)/punctuation mark(s), irrespective of where their position in this sequence.

E.g. "3", "20/8", "-1.5" ,"1,000,000", "-1", "35.7%", etc.

7b. According to one method, any numbers and/or special characters that occur after a word (i.e. after a space that follows a word) are preferably not included into the count of the frequency list E.g. "after 54 years" may result in the bigram "54 years".

Yet "after 54" may preferably NOT be counted as a bigram.

7c.According to another method, any numbers and/or special characters that occur after a word

(i.e. after a space that follows a word) are preferably included into the count

E.g. "after 54 years" may preferably result in the bigrams "54 years" and "after 54".

7d. According to one method, if there are consecutive chains of characters comprising digits/special characters separated by Space, they preferably are not counted as bigrams, unless a number/special character precedes a word.

E.g. the sequence "at 534 67 89 any time" will result in the following bigrams only:

"89 any"

"any time"

I.e. for the purposes of this invention, preferably sequences like "at 534", "534 67", "67 89" may not be considered as bigrams and may not be counted in the frequency list. 7e. According to another method, if there are consecutive chains of characters comprising digits/special characters separated by Space, they preferably are counted as bigrams.

E.g. the sequence "at 534 67 89 any time" may preferably result in the following bigrams:

"89 any"

"any time"

"at 534"

"534 67"

"67 89" In the cases of 7-7e, in an N-gram (e.g. 2-gram) word created as such, the number(s) may preferably be replaced by at least one predefined code/symbol or a special character (e.g. "#"), depending on their types (e.g. "23", "23.5", "-23", "23/5", "23%", etc.). As an example, the bigrams "89 any", "123 apples", and "5.8 kilometers", may be replaced, respectively, by the bigrams "# any", "# apples", and "* kilometers".Accordingly, as an example, if the system uses a bigram word database, when a user enters the input information corresponding a current (e.g. predicted) word being entered, wherein said word is preceded with a number, the system may search in the database for entries wherein their first gram is the relevant predefined code/symbol (e.g. special character) representing said type of number and wherein its second gram may correspond to the input information provided by the user.

8. "Return" ("Enter") may be treated as if it were a "Space". E.g. in the following text:

Have you seen her?

Yes, I have.

"? Yes" may preferably constitute a bigram.

9. The beginning of a document in the corpus may be marked by a predefined code/special character/symbol such as, for example, "^A". For example, this may result in bigrams such as "^Λ Hi", "^Λ Dear", etc. Accordingly, during the entry of the input information corresponding to a word, the system may check/identify if the word being entered is the first word of a document, and thereby search in the database for the entries wherein their first gram is the character (e.g. "^A") representing the beginning of a document, and the other corresponding gram (e.g., the second gram of bigrams) corresponds to the input information relating to the current word being entered

10. The database may be organized so that each entry includes an N-gram (e.g. 2-gram) word and is related to its frequency information. Thus the format of the database or frequency list may be, for example, as follows:

Frequency[TAB] Var 1 [TAB] Var2

E.g.:

235 same reason

220 ! However 200 7 books

198 us, they

11. According to one method, a gram other than the first gram of a an N-gram entry (e.g. the second gram of a 2-gram entry) may include or be constituted-of any type of character(s) such as letters and/or special characters (e.g. digit, punctuation mark, etc.) and/or a mixture of them.

12. According to one embodiment, a type of chain of characters excluding a letter may occur as the first and/or another gram of an N-gram entry of the database.

e.g.:

"2@# xwmy"

"code 34$%&"

"f34&* T54G" The current database may also include 1-gram words in form of N-gram (e.g. 2-gram) words. There may be many 1-gram words with high frequency in the corpus that constitute 2-gram words having low frequency of use. Such N-gram words may not be desired to be included in the N-gram database. In this case, such 1-gram words with high frequency may be modified to form N-gram (e.g. 2-gram) words wherein their first gram is a (one or more) special character (e.g. '"") or a code (e.g. ASCII code), and as such they may be included into the N-gram database. Preferably, to said special character a key value may also be assigned. According to one embodiment, after the user provides the input information corresponding to a desired word, the system may also consider the previous word relating to said input information and may search in the N-gram (e.g. 2-gram) database. If there is no corresponding N-gram word, the system may replace the previous word by the special character/code representing 1-gram words/chain of characters (e.g. including at least one character) of the N-gram database and search for N-grams wherein their first gram is said special character/code and their other gram (e.g. the second gram) corresponds to the input information provided by the user. If said database is a 2-gram word database, and the system finds a corresponding entry, it may propose the second gram of the entry as the predicted word.

As mentioned, the database created above, may preferably, be case sensitive. As such, during the entry of a word, if the user provides a predefined interaction such as providing a "shift" information/function relating to one or more characters of the word being entered, the system may consider said information and may look/search for the corresponding words within the database. According to one embodiment, if such a word with the specific case(s) is not found, the system may inform the user. According to another method, if such a word with the specific case(s) is not found, the system may look/search for a word corresponding to the input information provided by the user in the relevant database regardless of the said case(s).

In the embodiment just described, if the desired word is not in the dictionary, the user may proceed to the correction procedure. After entering such a word, the system may add it as an Ingram (e.g. 2-gram) word to the database wherein its first gram is said special character/code.

It must be noted, that the example of arrangement of an N-gram (e.g 2-gram) database described above is used to demonstrate the main principles of creating such a database. Changes to such an arrangement may be considered by people skilled in the art. For example, such an arrangement may exclude any of the exemplary cases described above, or it may include additional cases which are not described above.

It must be noted that an N-gram entry/word of an N-gram word database (e.g. where N>0) is considered, by people skilled in the art, as a single word have one or more grams. Therefore, when, for example, the system is using a 2-gram word database, the system may consider the word previous to the current word being entered as the first/initial gram of a 2-gram word. Based on this consideration, according to one method, during the entry of a word the system may create an input information consisting of two portions; a first portion including the precise character(s) of the previous word, and a second portion comprising the input information (e.g. ambiguous key interactions and/or precise characters provided by the user during the entry of the current word) corresponding to the current word being entered. According to one method, the system may also include/insert a predefined separator/delimiting-character, for example, a space character, between said portions of the information. The system then, may search for a 2- gram word of the database corresponding to the input information described above, and may accordingly predict/propose a Current word. Predicting/proposing such a word from an N-gram (e.g. 2-gram) word database has been described earlier in this application. As described before, the letters of a language may be grouped in four groups based on a common characteristic in their shape. Hereafter together they are called "original groups of letters" such as the following, for example:

Group 1: comprises letters that include a substantially closed circle in their shape (e.g., B, D, O, P, Q, R). According to one embodiment, this common shape has, preferably, higher priority over the other common shapes. Other letters, which may be divided into other common shapes, are divided in the following groups:

Group 2: letters that stand on one point: (e.g., F, I, J, T, V, Y)

Group 3: letters that stand on two points: (e.g., A, H, K, M, N, W, X)

Group 4: letters that stand on a wide base (e.g., C, E, G, L, S, U, Z)

As mentioned before, each of the above group(s) of character(s) may be assigned to a different input means (e.g. to a different key). In the arrangement above, some letters such as for example "R", and "P" may be confusing for a user, for example in classifying which group applies to each of the example's letter . For example, it may happen that a user may get confused and, for example, in order to enter the letter "R", the user may erroneously press the key corresponding to the letters that stand on two points (e.g. letter "R" has, both characteristics, it has a closed circle in its shape, and also stands on two points). To resolve such problems, according to one embodiment of the invention, confusing letters may be assigned to more than one group of characters. As such, as an example, the four following "enlarged groups of letters" may be considered by the system:

Group 1: comprises letters that include a substantially closed circle in their shape (e.g., B, D, O, P, Q, R). According to one embodiment, this common shape has preferably higher priority over the other common shapes. Other letters, which may be divided into other common shapes, are divided in the following groups:

Group 2: letters that stand on one point: (e.g., F, I, J, T, V, Y, and P)

Group 3: letters that stand on two points: (e.g., A, H, K, M, N, W, X, and R)

Group 4: letters that stand on a wide base (e.g., C, E, G, L, S, U, Z, and J)

Some of the letters of a group may be considered as the second degree letters in their respective groups. In the example of the enlarged groups of letters, each of the letters "P, R, and J", may be considered as the second degree letter in its respective group of letters. According to one method, the system may use one or more confusing set(s)/group(s) of characters corresponding to an ambiguous/confusing user interaction, to search an (e.g. one or more) N-gram (e.g. 1-gram or 2-gram) word database. The system may do this search using one or more ambiguous/confusing groups/sets

corresponding to a user interaction at any moment such as for example: (a) at the beginning of a new current word, (b) after a reject of the current word, (c) after the correction of the first character of the current word, and (d) preferably when, while (e.g. during or after) the user is entering the current word, a (e.g. on or more) predefined or calculated minimum frequency (e.g. 0 or greater, where 0 may represent that the word is not contained in the N-gram word database) threshold is achieved for the word corresponding to the current input information.

As an example, according to one embodiment, the system may use the enlarged groups of letters to predict a word. According to another embodiment, the system at first the may use the original groups of letters to predict a word, and at a redefined instance/circumstance/state such as for example, when the frequency of use of a corresponding word to be presented to the user is less than a predefined number, the system may begin to propose/predict words including (and/or through the use of) the second degree character(s) of any of the enlarged group(s) of letters (e.g. such as for example, those described above, or below) in the appropriate position(s) within the appropriate words to be predicted/proposed.

The example of grouping the letters as described above was based on their uppercase shape. Some people may erroneously interact with the letter keys based on the lowercase shape of a letter. In such case, according to one embodiment of the invention, the system may also consider the letters in the lowercase shape. An example (e.g. may be partial) of ambiguous/confusing groups/sets corresponding to a user interaction is available in the following Group Table 1.

Group Table 1:

Group 2 = P, 1, r

Group 3 = R, A, a

Group 4 = L, G

Group 1 = P, R, G/g, (eventually, also A)

According to Group Table 1, an identical character may be included in more than one group. For example, as stated above, the system may do a search using one or more

ambiguous/confusing groups/sets corresponding to a user interaction, such that in this example, if a user were to interact with Group 1 and Groups 5 and 6 (not shown in Group Table 1), and the system determines that the frequency of the current word found in the word database is below a minimum frequency threshold, then the system may use Group 4 instead of Group 1 to search for the user's desired N-gram word.

The use of ambiguous/confusing groups/sets, by the system, provides the user with greater freedom to interact with the system than a system without the use of ambiguous/confusing groups/sets. For example,

(a) some users may visualize the shapes of letters according to one paradigm (e.g. capital letters) while other users may visualize the shapes of letters according to another paradigm (e.g. lower case letters), or

(b) users may make a mistake while typing, and the system my use ambiguous/confusing groups while searching an (e.g. one or more) N-gram (e.g. 1-gram or 2-gram) word database to improve the accuracy of the system in finding the user's desired word, or

(c) when considering other scenarios where the use of ambiguous/confusing groups/sets that correspond to other user interactions and/or behavior may assist the user in finding the desired word or improve the accuracy of the system in finding the user's desired word, preferably, in the case where the system may not find a user's desired before the system uses one or more ambiguous groups/sets.

Furthermore, with regard to ambiguous groups/sets, a (e.g. one or more) character may be allocated to more than one group, and the system may be so designed such that such an allocation may be restricted to only one group or a smaller set of groups of all possible available groups (i.e. where a larger grouping may be devised). Such restriction of allocating letters to groups may be done in such a manner as to improve the search accuracy of the system in finding the user's desired word in an (e.g. one or more) N-gram (e.g. 1-gram or 2-gram) word database.

As an example, Group Table 2 below shows an allocation of letters into four groups based on their lowercase shape. They follow the principles of the original groups of characters (e.g. in this example, the letter "a" that could have been assigned to group 4, is exceptionally assigned to group 2). Group Table 2:

Group 1 = letters standing on one point: f, t, 1, r, v, y. (in their uppercase form, F, T, I, L, R, V, Y)

Group 2 = letters standing on two points: a, h, k, m, n, w, x (in their uppercase form, A, H, K, M, N, W, X)

Group 3 = letters standing on a large base: c, e, j, s, u, z (in their uppercase form, C, E, J, S, U, Z)

Group 4 = letters that include a substantially closed circle in their shape: b, d, g, o, p, q (in their uppercase form, B, D, G, O, P, Q). As described before, this group has preferably higher priority than the other groups.

Since each character is uniquely assigned to only one group in Group Table 2, this set of groups (Group Table 2) may be considered to be unambiguous/non-confusing. For examples, a single zone (as previously mentioned in this submission and several other submissions by the same author) may correspond to each group of Group Table 2. A letter may be assign to more than one group. In this case, the set of groups may be considered to be ambiguous. Furthermore, for the case where further groups are added to a Group Table, for example, an additional group 5, where a character may be contained in more than one group, such Group Table would also be considered to be an ambiguous set of groups. The system may use an unambiguous and/or ambiguous set of groups to enhance the accuracy of finding the user's desired word.

According to one embodiment, there may be considered a plurality of groups of letters including uppercase and lowercase letters to provide a "largest groups of letters" used by the word predictive system. An example of such groups is shown below:

Group 1: may preferably comprise letters that include a substantially closed circle in their shape (e.g., B/b, D/d, O/o, P/p, Q/q, R, and g). According to one embodiment, this common shape has preferably higher priority over the other common shapes. Other letters, which may be divided into other common shapes, are divided into the following groups:

Group 2: letters that stand on one point: (e.g., F/f, I/i, J/j, T/t, V/v, Y/y, and P/p, q, r)

Group 3: letters that stand on two points: (e.g., A/a, H/h, K/k, M/m, N/n, W/w, X/x, and R) Group 4: letters that stand on a wide base (e.g., C/c, E (e.g. exceptionally "e"), G, L, S/s, U/u, Z/z, and J/j). 2 000029

Grouping of characters may be applied to other alphabets or sets of symbols - here the set of Roman letters is used as an example. Upper-case and lower-case letters are mentioned as an example. Based on the principles described, according to one embodiment of the invention, at least during general-purpose text entry, when the user provides the input information corresponding to a desired word, the system may generally (e.g. preferably, substantially always) first (e.g.

according to one method, only) refer to an N-gram (e.g. 2-gram) word database (e.g. preferably, a 2-gram word database) wherein N is greater than one (i.e. N>1), regardless of the location of the word being entered within a text (e.g. even if the word is the first word of a sentence).

With continuous description of the current embodiment, during the Correction procedure/Mode, when the system presents the user's desired word, the user may confirm/accept said desired word by providing an end-of-the-word signal, for example, such as a space character, and the system may automatically activate the Predictive Mode of the system.

Alternatively, according to a second method, after the user provides the input information (e.g. key presses/interactions) corresponding to a desired word and the system predicts a word and the user rejects the word, the system may preferably enter into the Correction Mode and may (e.g. also) remain in Predictive Mode (e.g. the system may not activate the Precise Mode of the system), and preferably in order to provide a precise character for correction purposes the user may provide one or more appropriate interactions such as, for example, gliding action(s) to enter one or more precise characters. Other principles of this second method may be similar to those principles of the first method.

By considering the two portions (portion 1 and portion 2) of a 2-gram word database shown below, an example of entering a sentence "I like golf is provided hereafter:

Key presses

1 123 3431

fine grey 5400

like lost 3201

Key presses

nine lost 21 1 1

1 3123

I like 195000 like cost 181 1 I sing 10540 like golf 1551 I ing 5769 like grey 1280 029

Portion 1 (e.g. 2-gram) Portion 2 (e.g. 2-gram)

Figs. 386A to 386C, show an example of entering a phrase "I like golf, through the first method described above. In fig. 386A, after the user provides the input information (e.g. key pressing actions on the appropriate ambiguous keys 386001-386004) corresponding to the words "I" (e.g. by considering a 1-gram database, or by considering the second gram of a 2-gram word database (not shown)) and "like" (e.g. by considering the Portion 1 of a 2-gram database as shown above), the system predicts said words correctly, and the user accepts/confirms said words by providing a space character after each of said word are correctly predicted. Then, after the user provides the input information corresponding to the word "golf", and then by considering that the correct previous/preceding word was the word "like", the system may consider the 2-gram word "like lost" of the corresponding portion (e.g. Portion 2) of the database which, for example, has the highest priority among other words of said portion of the database that correspond to the combined input information, and may propose/predict the word "lost" 386008 which in this example is not the desired word. Note that in this example, the word "lost" is part of the 2-gram words "like lost" and "nine lost". The word "lost" by itself does not have a predefined frequency of use in an N-gram (e.g. 2-gram) word wherein N>1. It is selected because the 2-gram word "like lost" to which it belongs has the highest priority of use among all of the 2-gram words corresponding to the combined input information. The system proposes a single gram as the Current Predicted Word, from an N-gram (2-gram) word.

With continuous description of the current embodiment, at this time, the user may Reject said word by providing the gliding action 386007. Then as shown in fig. 386B, the system may propose a new word "cost" 386108 which still is not the desired word (i.e. "golf").

Simultaneously, the system may activate the Precise Character/Letter Mode of the system as described earlier in this application, wherein each of the letter keys in its precise instance may have/replaced-by a corresponding second keypad (e.g. together forming four second keypads 386101-386104) wherein each of the zones on a second keypad represents an identified character and wherein said second keypad is preferably also enlarged. At this time, the user may interact with (e.g. tap on) the zone 386107 to which the letter "G"/"g" is assigned. The system may replace the selected/highlighted (e.g. the first) letter 386110 of the Current word 386108 by the letter "g", and then, as shown in fig. 386C, by considering the combined input information (e.g. the key interaction on the first keypad, and the precise character now entered), the system may propose the word "golf" 386208. After the user confirms/accepts the predicted word "golf", 029 by for example providing an end-of-the-word signal such as pressing on the space key 386106, the system may enter said word, and preferably automatically activate (e.g. switch to) the Predictive Mode of the system. It must be noted that the example of entering a phrase (e.g. a collection/sequence of words) and the corresponding portions of the databases have been created and used to describe the current embodiment. The corresponding portions of another database may be different to those shown above. For example, portion 2 of the word database may not include the 2-grams, "like lost", and "like cost".

According to one embodiment of the invention, the system may be designed such that it may be switched into (or activate) the Precise Mode of the system after the user rejects a predefined N- th word such as the second word of the Current predicted word. According to one embodiment of the invention, the activation and/or use and/or switching of the Precise Mode and Predictive Mode of the system may be mutually exclusive, and the activation and/or use and/or switching of the Correction procedure/Mode and the Insertion

procedure/Mode of the system may be mutually exclusive. Precise/Predictive Mode

activation/use/switching preferably may not be mutually exclusive to the Correction/Insertion procedure/Mode of the system.

Gliding action before and after Rejection

According to one method, a predefined interaction, such as a predefined gliding action before or after rejecting the first predicted word, may correspond to different functions. As an example, according to one embodiment of the invention, during the entry of a desired word, a gliding action provided on or from a key in the direction of a letter/character on said key wherein said gliding action is provided before a Rejection action corresponding to rejecting the predicted word, said gliding action may correspond to an Insertion procedure/action of the invention (e.g. may correspond to inserting or adding an identified letter/character at the current cursor position of the Current predicted word) as described throughout this application and the previous applications filed by this inventor. Said/a gliding action, provided after said/a Rejection action, on or from said key in the direction of a letter/character on said key, may correspond to a Correction procedure/action of the invention (e.g. may correspond to replacing a

selected/highlighted letter/character in the current predicted word by the/said letter/character 2012/000029 corresponding to the direction in which the gliding action is provided) as described throughout this application and the previous applications filed by this inventor.

As an alternative method for providing an identified character, for example being used in the above mentioned embodiments and methods, for each highlighted/selected character of the Current word, the said second corresponding keypad of the invention may be located or displayed outside the letter keys. For example, if the system is using the first keypad on a first device such as a remote control for a media system (e.g. television, etc.), the second keypad may be displayed on a screen of a second device such as on the screen of a television.

Candidates

According to one embodiment of the invention, after the system predicts a word and the predicted word is rejected, the system may predict a new word and preferably propose a number of additional word choices, preferably equal to the number of letter keys, wherein the arrangement of word choices on the display, correspond to the arrangement of letter keys.

Preferably, each of said additional word choices may be have an indication (e.g. a box, background color, the font color, etc.) of color corresponding to the color of a corresponding letter key in the arrangement. At this time, according to one method, various scenarios may be considered:

(a) if said new predicted word is the desired word, the user may accept/confirm said predicted word;

(b) if one of the said proposed word choices is the desired word, according one method the user may interact with the letter key corresponding to that proposed word choice in order to enter said desired word and optionally a space character after said desired word; and

(c) if either, the said new predicted word or the said proposed word choices are not the desired word, the user may proceed to the correction procedure of the invention. Word completion

In addition to or instead of the Current predicted word, one or more words may be predicted by the system based on the input information provided by the user, said input information at least relating to some (e.g. one or more) of the characters, preferably the beginning characters, of a desired word having N gramms. Preferably, said one or more words may be words having the higher priority of use (e.g. as recorded in a database of words) relating to the other words corresponding to the provided input information. According to one embodiment of the invention, the selection of one or more words may be executed using at first an N-gram database/dictionary, preferably a 2-gram dictionary. If the system does not find an appropriate word, the system may then use other databases, for example, such as a 1-gram dictionary.

According to one embodiment of the invention, the word completion system may use an N-gram word database to complete a current word that is being entered. By considering the previous word (or several previous words) of a current word being entered and the input information (e.g. key interactions such as tapping or gliding actions) corresponding to the beginning one or more characters of the current word being entered, the system may more precisely and/or quickly complete/predict a word that is being entered. As an example, if the system uses at least a 2- gram word database, in this embodiment, the word predictive system may preferably first look for/search words within 2-gram database entries where an entry's first word/gram begins with said previously entered word (e.g. precise characters) and where the beginning characters of said entry's second word correspond to the input information (e.g. ambiguous input information) being entered for a current word.

As an example, if the user desires to enter the word "fish" after the word "freshwater" that is already entered, if the user presses the key (e.g. ambiguous letter key) corresponding to the first character "f" of the desired word (e.g. "fish"), and for example the only 2-gram entry

corresponding to the combined input information (e.g. that input information corresponding to the words "freshwater" and one or more of the beginning characters, in this example "f " of the word "fish") is the 2-gram entry "freshwater fish", then the system may propose/predict the word/gram "fish".

According to one method, if the system does not find a second word of an entry of for example, a 2-gram word database corresponding to the combined information, including the previous (e.g. correctly being entered) and for example the input information corresponding to one or more beginning letters of a desired word provided by the user at any stage/character while inputting the desired word, the system may then refer to a 1-gram word database to find a word corresponding to said input information corresponding to one or more beginning letters of a desired word provided by the user.

Phonetic characters As mentioned before, the data entry system and its features such as using an N-gram database to predict a word may also be applied, for example, to languages with a hieroglyphic system of writing wherein their input system may be based on the phonetic alphabet/transcription such as Chinese and Japanese. In fact, each of said languages are based on a set of characters, respectively, Roman characters and phonetic characters. In a Roman-based language a set of Roman characters constitute a word, in languages with a hieroglyphic system of writing (e.g. Chinese, Japanese) a set of phonetic characters constitutes a phonetic transcription system (e.g. Pinyin, Bopomofo, Romaji), or a phonetic alphabet (e.g. Japanese alphabet Kana). A Chinese or a Japanese word/word segment (e.g. represented by a hieroglyph(s)/ideogram symbol) written in Pinyin, Bopomofo, Romaji or Japanese alphabet Kana is hereafter referred to as "Phonetic

Word" and/or "phonetic word", etc. Accordingly, the principles of the data entry system using an N-gram (e.g. 2-gram) database may be applied to in the same manner to predict a word of a Roman-based language or a Phonetic Word. Note that, a Phonetic N-gram database may be a collection of N-gram Phonetic Words.

According to one embodiment of the invention, the N-gram principles of predicting a word as described throughout this application may be applied to the phonetic N-gram (e.g. 2- gram) databases such as Pinyin, Bopomofo, Romaji and Kana (e.g. including Hiragana,

Katakana) wherein, preferably, each gram of an N-gram entry of the database may be related to one or more ideogram symbols constituting a word (e.g. Chinese or Japanese characters).

Accordingly, after a phonetic set of characters (e.g. a word's phonetic transcription such as

Pinyin, Bopomofo, Romaji, or a word written in Japanese alphabet Kana, hereafter referred to as "phonetic word") corresponding to an ideogram symbol is correctly predicted by the system, the system/user may select the corresponding ideogram symbol.

For purposes of clarification of the above, as mentioned before, in addition to a language using Roman characters, the data entry system and its features, such as using an N-gram database to predict a word may also be applied, for example, to a language with a hieroglyphic system of writing (e.g. such as Chinese and Japanese) wherein their input system may be based on a phonetic alphabet/transcription. Therefore, the principles of the data entry system of the invention described throughout this application and the previous applications, filed by this inventor, may be applied to entering text in any language using an alphabetical input system such a as phonetic alphabetical system (e.g. such as Chinese and Japanese) and/or other types of alphabetical characters such as Japanese Kana.

In a Roman-based language a set of Roman characters constitute a written/printed word, while in a language using a hieroglyphic system of writing, a phonetic transcription may 2 000029 represent a written/printed hieroglyphic word, wherein said hieroglyphic word may include one or more hieroglyphs. A phonetic transcription representing a hieroglyphic word is hereafter referred to as "Phonetic Word". Accordingly, the principles of the data entry system using an Ingram (e.g. 2-gram) database may be applied to in the same manner to predict a word of a Roman-based language and/or a Phonetic Word. Note that, a Phonetic N-gram database may be a collection of N-gram Phonetic Words. Note that in many cases several hieroglyphic words may be represented by a same Phonetic Word.

According to one embodiment of the invention, the N-gram principles of predicting a word as described throughout this application may be applied to the phonetic N-gram (e.g. 2- gram) databases, for example, such as Pinyin, Bopomofo, Romaji and Kana (e.g. including

Hiragana, Katakana) wherein, preferably, each gram of an N-gram entry of the database may be related to one or more ideogram symbols constituting a word (e.g. Chinese or Japanese characters). Accordingly a first method, after a Phonetic Word, (e.g. including phonetic transcriptions) such as Pinyin, Bopomofo, Romaji, etc.) corresponding to a hieroglyphic word, is correctly predicted by the system, the system/user may select a corresponding hieroglyphic word. According to a second method, after the user inputs the information corresponding to some of the beginning characters of a Phonetic Word that its beginning characters correspond to said information, and/or alternatively the system may predict a hieroglyphic word corresponding to said information.

As an example, hereafter shows two portions of an exemplary phonetic (e.g. Pinyin) 2- gram phonetic word database and the exemplary corresponding Chinese words and key presses.

Keypress: 43231 43 Keypress: 1123 222

2-gram 2-gram 2-gram 2-gram

(phonetic) (hieroglyphic) (phonetic) (hieroglyphic) benci qu ting ama

renzi de jixu kan mm

oumei ru in timu nan

qunei bu ting wan 2-gram only, tree-based includes 1-gram

According to one embodiment of the invention, the database used by the system may be a single database (incorporating a mixture of 1-grams, 2-grams, etc.) organized as a tree. An example of a portion of such a database is shown below in Table 101.

Table 101

Interaction with the keys of second keypad (s)

It must be noted that when the system switches into the Precise (e.g. Letter and/or Special Character) Mode and shows one or more second keypads of the invention (e.g. four second keypads 369001-369004 as shown in fig. 369, each respectively including the letters of one of the four keys 370001-370004 of the first keypad of the invention, as shown in fig. 370), according to one method, a tapping action on a zone/key assigned to an identified character on one of said second keypads, or a gliding action provided on said key of said second keypad towards said identified character, may correspond to entering said character. In the case of said gliding action, according to one method, when the user touches anywhere on said key and starts the gliding action, the system relocates said second keypad under the user's finger such that the center of said keypad being the departing position of said gliding action.

Entering user's name 9

According to one embodiment of the invention, when the system is purchased by a user, he/she may be required to provide and/or may provide some information such as at least the user's personal information such as the user's, name, address, email address, credit card number (e.g. wherein at least some of the keys of the first keypad ambiguously correspond to one or more digits as described before), URLs that the user is mostly/generally using, etc. In this case, according to one method, the system may be designed such as to add the information to the corresponding word databases(s) used by the system. Usually, said information is provided by the user without using the data entry system of the invention before purchasing the system. This method of gathering the user's personal information may be extremely beneficial because the user's personal information may now be easily entered using the system, for example when the user is electronically shopping for goods and is required to provide the user's personal information. Preferably, each of the user's information (e.g. name and surname, address, email, etc.), may be considered as a single entry of the database. Said user's information is any example of any type of information that may be gathered and/or used by the system. The method of gathering and/using user's information or any other information may be done automatically while purchasing.

Highlighting the first letter after entering into Correction mode

According to one embodiment of the invention, the a character, preferably the first ambiguous character, of the Current predicted word may automatically be selected/highlighted after the system enters into the Correction Mode of the invention, for example after the user rejects a predicted word.

Quick glide for insertion of letters. Press and hold and glide for sp chars

According to one embodiment of the invention, a predefined type of gliding action such as a quick gliding action (i.e., a beginning a gliding action before a predefined period of time is expired after touching a corresponding key) from/on a letter key in the direction of a letter on said key may correspond to the Insertion procedure (e.g. inserting a letter at the cursor position within, for example, the current word being entered) of the invention as described before.

According to one method, pressing-and-holding a letter key for at least a predefined period of time may activate the Special Character Mode of the system. The user may then provide a desired gliding action (e.g. or vice versus). As an example, before providing a desired gliding action that may correspond to entering a special character, the user may press-and-hold an appropriate (e.g. letter) key for at least a period of time (e.g. to enter the system/key into the Special Character Mode, and preferably without removing the user's finger from said key, provide said desired gliding action to enter a desired special character. Preferably, after providing the gliding action, the system may automatically switch back to the Predictive Mode. Note that in the special character mode, the system may also include functions. As an example, a predefined interaction, such as a predefined gliding action may correspond to activating a corresponding function. According to one method, when the system enters into a mode instance, substantially all of the keys of the keypad may change/replace their functionalities/characters to/by new functionalities/characters.

With continuous description of the current embodiment, according to one method, pressing-and- holding a letter key for said predefined time period or for another, for example, longer time period and without the user's finger providing a desired gliding action as mentioned in the former part of the description of this method, removing the user's finger from the key may correspond to entering the special character which may be located in the center of the key, as described throughout this application and other applications filed by this inventor.

According to another method, (e.g., press-and-holding for a predefined laps of time, and) providing a back-and-forth gliding action may correspond to entering the special character assigned to the center of said key in the special character mode. Accordingly, press-and-holding a key for at least a predefined laps of time, and releasing said key (e.g. without providing a gliding action) may preferably ambiguous correspond to a special character assigned to said key. The principles of assignment of letters and special character to a key have already been described.

As mentioned before, providing a predefined interaction such as a gliding action in a predefined direction on/from a key, preferably, other than a letter key may activate the Precise Mode (e.g. Precise Letter Mode of the system). According to one embodiment of the invention, providing another predefined interaction such as a gliding action in another predefined direction on/from a key , preferably, other than a letter key may activate the Special Character Mode, preferably, in the Precise instance, as described earlier. In both embodiments, preferably, repeating any of said predefined interactions (e.g. without providing the other predefined interaction in between) or providing another predefined interaction, may activate the Predictive Mode (e.g. Predictive Letter Mode) of the system. According to one embodiment of the invention, when the system is in any of the Precise Modes (e.g. in the Precise Special Character Mode, or in the Precise Letter Mode, for example, after rejecting a predicted word) providing the press-and-hold action on a key may activate the Special Character Mode of the system. The user may then provide a gliding/tapping action as described above. In the current embodiment, when the system enters into the Special Character Mode, preferably, in both cases the appropriate interface corresponding to the gliding interactions for entering special characters/symbols may be presented to the user, for example, by displaying the appropriate interface on a corresponding screen, or by using a text to speech technology.

According to one embodiment of the invention, when the system is in the Precise Special Character Mode, providing a gliding action on the corresponding key may be related to

Insertion/Correction relating to entering a precise letter procedure of the invention.

It must be noted that a key may be a physical key or an imaginary/virtual key, and the interactions with a key, such as a tapping/gliding action on said key, may be provided on a physical key, a soft key, in the air, etc.

Languages with Hieroglyphic system of writing (e.g. Chinese/Japanese)

When the system is used for entering text in languages with hieroglyphic system of writing (e.g. Chinese, Japanese), preferably, in addition to the Phonetic N-gram database(s), the system may include an N-gram (e.g. 1-gram and/or 2-gram) database of hieroglyphic words. According to one embodiment of the invention, during the entry of the input information and/or after the system predicts correctly a first Phonetic Word (e.g. based on the input information such as at least one of, key presses, reject action, correction procedure, etc., as described before in this patent application), preferably, the system may propose a corresponding hieroglyphic word (e.g. as the best choice) for example the corresponding hieroglyphic word having the highest frequency of use, preferably within the corresponding words of an appropriate database of hieroglyphic words (e.g. preferably a 2-gram database, or a one-gram database, etc.)

corresponding to the instance/location of the desired hieroglyphic word within the text that the user is entering. Preferably after confirming the predicted Phonetic Word, at least two scenarios may be considered:

a) If the predicted hieroglyphic word is the desired word, the user may proceed to

entering a next desired Phonetic Word or another interaction (other than a rejection procedure that will be described in scenario (b) hereafter), and the system may enter said hieroglyphic word. b) If the predicted hieroglyphic word is not the desired word, then according to one method, the user may Reject the predicted hieroglyphic word by providing a predefined interaction, preferably the same interaction used for rejecting a Phonetic word. The system may propose another hieroglyphic word. Preferably, the system may also propose a selection of additional corresponding hieroglyphic words (e.g. if any). At this time at least two possibilities may be considered a. If said another predicted hieroglyphic word is what the user desires, then as described above, the system user may proceed to entering a next desired

Phonetic word or another interaction, and the system may enter said word. b. If said another predicted hieroglyphic word is not what the user desires, then, the user may select his desired hieroglyphic word among the proposed additional hieroglyphic words. The selection of additional corresponding hieroglyphic words may preferably be arranged in one or more (e.g. such as up to four) second keypads (e.g. preferably, having a three by three matrix of zones/keys) of the invention, as described and shown throughout the applications filed by this inventor. To select one of said additional hieroglyphic words, according to a first method, the user may tap on said word, wherein according to a second method, the user may touch the corresponding second keypad of the invention and provide a gliding action towards the desired hieroglyphic word. Providing a tapping or a gliding action to select a symbol on a three by three matrix of keys/zones has already been described in detail in the patent applications filed by this inventor. With continuous description of the current embodiment, according to one method, the hieroglyphic words of the selection may be assigned to the one or more second keypads of the invention based on for example their common characteristics such as their shape, meaning, sound (for example of the first pinyin), number of hieroglyphic symbols of a word, etc.

According to another embodiment of the invention, during the entry of the input information and/or after the system predicts correctly a first Phonetic Word (e.g. based on the input information such as at least one of, key presses, reject action, correction procedure, etc.), preferably, the system may propose a corresponding hieroglyphic word (e.g. as the best choice) for example the corresponding hieroglyphic word having the highest frequency of use, preferably within the corresponding words of an appropriate database of hieroglyphic words (e.g. preferably a 2-gram database, or a one-gram database, etc.) corresponding to the instance/location of the desired hieroglyphic word within the text that the user is entering.

Preferably after confirming the predicted Phonetic Word, at least two scenarios may be considered:

a) If the predicted phonetic and hieroglyphic words are correct, the user may press Space to confirm/enter the word. b) If the predicted phonetic word is not a desired word, then according to one method, the user may Reject the predicted phonetic word by providing a predefined interaction. The system may propose another phonetic word. At this time, at least the following possibilities may be considered: a. If said another predicted phonetic word is what the user desires, as well as the hieroglyphic word, then as just described above, the user may confirm both the Phonetic and the hieroglyphic words, and the system may preferably enter the hieroglyphic word. b. If said another phonetic word is not the desired word, then the use may

proceed to the correction procedure until the system predicts the correct word, and may proceed to the possibility (c) as described below. c. If said another phonetic word is the desired word but the another predicted hieroglyphic word is not what the user desires, then, the user may Reject the predicted hieroglyphic word by providing a predefined interaction, i.e. by providing a predefined interaction such as for example gliding from the Space key upwards. The system may predict another hieroglyphic word and may propose a selection of additional hieroglyphic words. The user may either confirm said predicted word or select his desired hieroglyphic word among the proposed additional hieroglyphic words.

The selection of additional corresponding hieroglyphic words may preferably be arranged in one or more (e.g. such as up to four) second keypads (e.g. preferably, having a three by three matrix of zones/keys) of the invention, as described and shown throughout the applications filed by this inventor. To select one of said additional hieroglyphic words, according to a first method, the user may tap on said word, wherein according to a second method, the user may touch the 9 corresponding second keypad of the invention and provide a gliding action towards the desired hieroglyphic word. Providing a tapping or a gliding action to select a symbol on a three by three matrix of keys/zones has already been described in detail in the patent applications filed by this inventor.

With continuous description of the current embodiment, according to one method, the hieroglyphic words of the selection may be assigned to the one or more second keypads of the invention based on for example their common characteristics such as their shape, meaning, sound (for example of the first pinyin), number of hieroglyphic symbols of a word, etc.

Bopomofo

According to one embodiment of the invention, Bopomofo (Zhuyin Fuyao) phonetic symbols are classified into four groups and assigned to the four (e.g. letter) keys of the first keypad of the invention and preferably may be used to enter Phonetic words preferably based on principles as described for other phonetic alphabets above. The said classification may be based for example on the traditional way of their writing. As an example, the symbols may be arranged into four groups as following:

The first group may preferably include the symbols that may be written with a single stroke:

* ¾ < V *7 £ - (e.g. this group may include two exceptions:

The second group may preferably include the symbols in which the first or the last stroke is a horizontal stroke

The third group may preferably include the symbols in which the first or the last stroke, is a vertical stroke

(e.g. this group may include two

exceptions: tl] ) 0029

The fourth group may preferably include the symbols which preferably may not be included within one of the first three groups, and wherein in their writing/shape they include a diagonal right-to-left stroke or a diagonal left-to-right stroke (e.g. of any kind such as long or short).

Words including special characters

According to one embodiment of the invention, in addition to the letters of a language (e.g. divided based on their common shapes and) assigned to a number of keys such as for example four keys, special characters may be divided into several groups (e.g. four groups, such as those shown and described throughout this application and other patent applications filed by this inventor) based on their common characteristics and each group being assigned to a separate key (e.g. to said four keys). They may be inputted as part of a current word being entered.

As an example; Fig. 387A shows the distribution of special characters on four keys of the first keypad of the invention. In this example, the punctuation mark characters are assigned to the key 387001 and are indicated by an icon 387011, the digits 0-9 are assigned to the key 387002 and are indicated by an icon 387012, the arithmetic characters are assigned to the key 387003 and are indicated by an icon 387013, and other (e.g. rarely used) characters are assigned to the key 387004 and are indicated by an icon 387014. It must be noted that some exceptions may be considered in this distribution. For example, the characters "#" and "@" are assigned to the key 387004. It must be noted that the common characteristics as just described is one of many other characteristics that may be considered by people skilled in the art.

According to one embodiment of the invention, a tapping action on a key may ambiguously correspond to any of letters and/or special characters assigned to said key. Accordingly, the words (e.g. entries) of a dictionary/database of words may include any of said special characters and/or other characters. As an example, in order to enter the word -sk2 ! C", the user may tap the on the keys 387003, 387002, 387002, 387001, and 387003. The system may at first predict the word "cnn's" 387008 which may have the highest priority (e.g., itself or its corresponding N-gram) among the corresponding database entries. The user may reject the predicted word, and as shown in fig. 387B, the system may propose another corresponding word for example the word "sam's" 387108 for example the word corresponding to said key presses and having for example the next highest priority. This word is still not the desired word, therefore the user may proceed to the correction procedure. As shown in fig. 387B, the user may first tap on the key 387111 relating to the letter "s", the system may propose the word "shave" (not shown). If this is still not the desired word, the user then, may tap on the key 387112 corresponding to the letter "k". The system may propose the word "skate" (not shown) which still is not the desired word. Now, as shown in fig, 387C, the user may (e.g. manually) enter the system into the Precise Special Character Mode, and tap on the key 387209. The system may propose the word

"Sk2 k" 387208.

Alternatively or in addition to the principles described above, according to one embodiment of the invention, providing a short pressing action (e.g. a tapping action, pressing a key for less than a predefined laps of time) on a key of the first keypad may ambiguously correspond to a letter assigned to said key, and providing a long pressing action (e.g., pressing a key for at least a predefined laps of time) on a/said key may ambiguously correspond to a special character assigned to said key. This may help the system to better predict a word that includes letters and special characters, because the user may inform the system which type of character (e.g. a letter or a special character) must be in a specific position within a word being entered by providing corresponding short and long pressing actions. It must be noted tat according to one method, the short gliding action may ambiguously correspond to both letters and special characters, and the ling pressing action may correspond to the special characters only (or vise versa).

According to one method, different key values may e assigned to the long pressing actions on the keys of the first keypad, and to the characters of the corresponding groups.

According to one embodiment, in addition to or instead of the special characters, a long pressing action on a/any (e.g. such as for example a letter) key may (e.g. also) ambiguously correspond to a letter assigned to said key, but may also inform the system to search within a specific group of entries in a database for a word being entered. According to one method, said specific group of entries may include words with specific characteristics such as words belonging to SMS vocabulary, medical vocabulary, etc. By doing so, the system may better and quicker predict a desired word. According to one example, if the user taps on the keys 387003, 387004, and long presses on the key 387003, the system may look for two types of words, a first type consisting of words corresponding to the key interactions wherein their last character is a special character, and a second type consisting of words belonging to the SMS vocabulary. In this case the word "LOL" may have the highest priority, and the system may propose said word. Otherwise, if the user provides (short) tapping actions on 387003, 387004, 387003, the system would have provided another word such as the word "los" having highest priority in the corresponding general entries of the/a database. Note that in the current embodiment, according to one method, said specific group of entries may constitute or be considered a separate database and/or be also included or be considered as being part of the non-specific (e.g. general) entries/words within/of a/the database.

According to one embodiment of the invention, a special character may (e.g. ambiguously) be assigned to more than one key of the keypad of the invention.

According to one embodiment, providing a first type of interaction with a key of the first keypad of the invention such as pressing said key for a predefined short period of time, and, preferably subsequently, providing a desired gliding action towards a letter on said key may correspond to entering said letter precisely (e.g. for insertion/correction). Accordingly, providing a second type of interaction with a key of the first keypad of the invention such as pressing said key for a predefined longer period of time and, preferably subsequently, providing a desired gliding action towards a special character on said key may correspond to entering said special character precisely (e.g. for insertion/correction). In the current embodiment, when the user provides the first type of interaction (e,g, with a key) the letters are shown on the appropriate key, and when he provides the second predefined interaction (e.g. with a key), preferably, instead of the letters the special characters may be printed/shown on said key. Fig. 387A shows, as an example, the letter of the alphabet being assigned to the keys 387001 to 387004 of the first keypad of the invention, and Fig. 387D shows, as an example, the special characters assigned to the same keys 387001 to 387004. When a (letter) key (e.g. 387001) is pressed and held for at least a predefined amount of time, the system may enter said key into its special character mode, and may enlarge said key. Accordingly, as an example, a special character may be entered precisely by press-and- holding an appropriate key and providing an appropriate gliding action. During the entry of a word, this procedure may be used and may be repeated if necessary until the system predicts the desired word. In this embodiment, according to another method, the first type of interaction may be assigned to entering special characters and the second type of interaction may be assigned to entering letters.

During the entry of a word, inserting precisely a special character (e.g. "!", ?, etc.) as part of a word being entered may be confusing for the system, because the system may not know if the user either intends to insert said character as being part (e.g. a character) of a word being entered or as a character separately from/regardless of the word being entered. For example, after the user enters the word "MP3", (e.g. by tapping action on the keys 387002, 387004, 387002, or, by tapping on the keys 387002, 387004, and long pressing action on the key 387002), the system may propose the word MP3) inserting the special character "?" precisely may provide the word/chain of characters "MP3?". The system may not know if the intended word is the word "MP3" or the word "MP3?". To resolve this problem, different embodiments may be considered. Some of such embodiments are described hereafter: According to a first embodiment, if at least one of said words/chain of characters is not in the dictionary/database of words used by the system (e.g., as an N-gram), the system may, automatically or by asking the user, add said at least one word to the database of words. This may augment the number of words of the dictionary/database significantly, and may reduce the accuracy of the prediction. This method may be time consuming and distracting for the user.

According to a second embodiment, if a special character is entered before providing an end-of-the-word signal, said special character may be considered as being part of the word being entered, and if said special character is entered after an end-of-the-word signal said special character may not be considered as being part of the word being entered.

According to a third (e.g. a preferred) embodiment, a gliding action provided (e.g.

before/after rejecting a predicted word) on any of the letter keys of the keypad of the invention, may correspond to inserting a corresponding character (e.g. letter, special character) precisely, and a tapping on a key of any of the second keypads of the invention in the precise mode (e.g. in Precise Letter Mode or Precise Special Character Mode) may correspond to

correcting replacing a corresponding ambiguous/precise character of the predicted/proposed word by the appropriate precise character being entered. If there is no ambiguous or precise character to correct (e.g. none of the characters of the word being entered is

selected/highlighted, or all of the characters of the current word are being fixed/modified as precise characters), the system may insert the precise character at the cursor position in (e.g. to the end of) the current predicted word. Preferably, the precise character provided through the Precise Letter Mode or through the Precise Special Character Mode, may be considered as part of the current word being entered. According to one method, the special character provided through a gliding action may preferably not be considered as part of the word being entered. As an example, after a word is being entered for example by tapping on the keys of the first keypad or after a correction procedure, inserting a precise special character through a gliding action may 029 insert said character at the end of said word, regardless of said word, and may be considered as an end-of-the-word signal. For example, after the system predicts the word "cost" without requiring a correction, or after the system predicts the word "golf through the correction procedure (e.g. by a tapping action on the letter "G" of the appropriate second keypad of the invention), inserting the character "?" through a gliding action on the corresponding letter key of the (e.g. first) keypad in its special character mode, may insert said character "?" at the end of any of said words. In the current method, preferably, a precise letter inserted through a gliding action may be considered as part of the word being entered. According to another embodiment of the invention, a chain of characters (e.g. preferably excluding a space character) wherein at least one of its special characters is entered precisely through a gliding action may be considered by the system as a word. Different scenarios regarding such a word may be considered:

1) After entering such a chain of characters, if the user provides an end-of-a-word

signal (e.g. space character), the system may consider said chain of characters as a word, and if said word is not in the appropriate word database (e.g. dictionary), the system may enter said word to the appropriate (e.g. 1-gram, and/or N-gram wherein N>1) word database.

2) After entering such a chain of characters, wherein at least one of its letters is

ambiguous, if the user provides a rejection interaction, the system may enter into the correction procedure of the invention as described earlier (e.g. the system may also simultaneously propose another corresponding word (e.g. if any exists) of the word database of the system.

In the scenarios described above, according to one method, if the user desires to enter a special character such as a "?" or "!", or etc. at the end of the chain of characters/word but not being as part of it, the user may first enter/confirm said word and then may provide said special character.

It is understood that other methods and embodiments to resolve the problem may be considered by people skilled in the art.

In the embodiments above, using an N-gram (e.g. 2-gram) word database including words including special characters may augment the accuracy of the prediction of a word including one or more special characters. 2 000029

According to one embodiment, by considering its previous word(s) in the sentence/text, if a word being entered is not related to (e.g. does not exist in) an N-gram (e.g. 2-gram) word database, then the system may combine and add said previous word(s) and the word being entered in an entry of said word database. According to one method, if said word alone is also not included in the one-gram word database, it may also be added to said word database. If the words of the 1-gram database are organized as N-gram (e.g. 2-gram) words wherein their first gram word is a special character relating to their status within a text/sentence, then said word being entered may also be entered as an N-gram (e.g. 2-gram) word accordingly.

The embodiment described above may permit to enter any type of words/chain-of-characters which are included in one or more databases of the system, easily and accurately. As an example, in order to enter the word "http://www", the user may simply tap on the corresponding keys 387002, 387001, 387001, 387004, 387001, 387003, 387003, 387002, 387002, 387002. The chances that the system predicts such a word without requiring correction are very high. If the system uses a word completion technology, tapping the beginning characters may be sufficient for the system to predict said word. After the system predicts the right word, the user may enter a "." through, for example, the space key. In the example above, alternatively, the key interaction corresponding to one or more of special characters of the word "http://www" may be long pressing actions.

By being able to enter ambiguously and precisely chain of letters and/or special characters, the data entry system of the invention may use a specific databases wherein its entries include words with such characters. For example, the system may use a software programming language database of words and a software programmer may use the data entry system of the invention to write a computer program.

According to one embodiment, the system may permit the entry of any type of word even those not including any letter, such as codes, numbers, etc. According to another embodiment, the system may not permit the entry of words that do not include at least one letter.

According to one embodiment of the invention, the correction procedure relating to a word including a special character being entered may be allowed to be applied only to the letters of said word, according to one method, after correcting, if there is more than one word in the database that corresponds to the input information relating to the desired word, then the system may propose said more than one word(s) and the user may select one of them.

According to one embodiment of the invention, when a new word which is not in the database is entered by the user, a predefined interaction outside the keys of the (e.g. first) keypad of the invention, may correspond to informing the system to add said word to the database (e.g. in form of an N-gram word wherein N>0, preferably in form of a 2-gram word). According to one method, interacting with any of the keys such as proceeding to for example to entering the next word may inform the system that the user does not desire to add said new word to the database.

Resuming Modes

It must be noted that throughout this application different mode instances (e.g. keypads) of the system have been described. Hereafter the main modes are being summarized for the sake of clarity:

Predictive/Ambiguous Mode

Mode 1: Predictive Letter and/or Special Character Mode (e.g. the first keypad of the invention) - a predefined interaction such as a pressing action on an ambiguous key may correspond to a any character of a group of characters assigned to said key. According to one method, a first type of interaction such as a short/soft pressing action (i.e. tapping) action may ambiguously correspond to a first group of characters (e.g. letters, letters and special-character), and a long/stron pressing action may ambiguously correspond to a second group of characters (e.g. special-characters, letters and special-character).. Precise Character Mode

Mode 2: Precise Letter Mode (e.g. activated from an ambiguous key)

a first predefined action on an ambiguous key, other than a tapping action such as a gliding action without first holding said key, may activate a corresponding second keypad relating to a group of precise characters. Such action may preferably correspond to an identified character (preferably a letter) assigned to an appropriate key/zone of the second keypad.

Mode 3: Precise Special Character Mode (e.g. activated from an ambiguous key)

a second predefined action on an ambiguous key, other than a tapping action such as a gliding action with first holding said key, may activate a corresponding second keypad relating to a group of precise characters. Such action may preferably correspond to an identified character (preferably a special character or function) assigned to an appropriate key/zone of the second keypad. Mode 4:

Precise Letter Mode

Each of the keys of the predictive keypad may be replaced by a corresponding plurality of specific zones/keys referred to as a second keypad of the invention, Each of at least some of said zones may represent an identified character, preferably a letter. A predefined action such as a tapping action on a specific zone/key of a second keypad may enter an appropriate

character/letter precisely.

A first predefined user's interaction such as a gliding action in a first direction on the backspace key, or a predefined response of the system to a user's input information/interaction (e.g.during the entry of a word) may activate this mode.

Precise Special Character Mode

Each of the keys of the predictive keypad may be replaced by a corresponding plurality of specific zones/keys referred to as a second keypad of the invention, Each of at least some of said zones may represent an identified character, preferably a special character/function. A predefined action such as a tapping action on a specific zone/key of a second keypad may enter an appropriate special character/function precisely.

A second predefined user's interaction such as a gliding action in a second direction on the backspace key, or a predefined response of the system to a user's input information/interaction (e.g. during the entry of a word) may activate this mode.

Correction Mode

During entry or edit of a word, the system may automatically or the user may manually enter the system into the Correction mode to replace at least one (e.g. generally an ambiguous) character of the current word proposed to the user. By considering said at least one precise character and the input information corresponding to at least some of the other characters of the desired word received by a processor of the system, during the entry/editing of the word, the system may more precisely predict the desired word. The correction mode procedure may be activated based on different factors such as for example, after a rejection interaction, when the word being entered is not in the database, or manually by the user for example by providing a gliding action on a key of the first keypad. Note that when a gliding action is provided on/from a key of the first keypad, said key enters into its precise mode instance, therefore said key may be considered as specific second keypad corresponding to/replacing said first keypad, even if said second keypad may have the same appearance as its related first keypad. According to another method, when the system enters into the precise mode, by for example providing a rejection action, each of the keys of the first keypad may enter into their precise character mode. In this case, a tapping action on a zone represented by a character may enter said character. Preferably, when the system into said mode, each of the keys of the first keypad are replaced by a different second keypad, even if said second keypads have the same appearance as the keys of the first keypad. In this example, tapping on a zone corresponding to a specific character on a key of the second keypad corresponds to a specific/precise character, preferably, when the zones corresponding to an specific character on the second keypad are marked by a grid. Each of said zones may be considered as a key of the corresponding second keypad.

Insertion mode

The Insertion mode resembles to the principles described for correction mode with the difference that in this mode the precise character(s) are added/appended at the cursor position within the current word or within the text. Compound enteries vs N-gram entries

Although the terms "N-Gram in which N>1" and "combined word" have been used alternately throughout this patent application; in some of the embodiments, they may not considered full synonyms for the purposes of this invention, while in some other embodiments relating to using linguistic rules the N-gram words are referred to as combined words,_they preferably may have different meaning depending on the particular in which they are referred to. In some cases both terms may refer to the same type of items in the database (e.g. an N-gram).

In some cases, the system may consider a multi-part entry of the database to provide output (e.g. to propose the word/s) at a high rate of accuracy. Regarding this invention, the difference between "N-Grams in which N>1" and "compound words" may have to do with the presentation of the predicted words to the user (i.e. with the output). Specifically, in case of a "compound word" the system may propose all its constituent parts, so that the user then may

confirm/correct/reject (some of) the parts by applying methods described in this patent application; whereas in the case of an "N-gram in which N>1" one ore more of the constituent parts/grams (preferably, the last part/gram) may be displayed each time the user enters information corresponding to a gram (e.g. to a single/isolated word), so that it is this particular constituent part to which the confirmation/correction/rejection procedures may be applied. It must be noted that in the embodiments wherein the system uses an n-gram database to predict/propose a word, according to one method the system may propose the whole entry corresponding to the input information provided by the user and after the user confirms his desired word which in general is the last gram of the proposed n-gram word, the system enters said gram of the proposed N-gram word.

Auto-correction

According to one embodiment of the invention, the system may include an auto-correction means such that for example if a user mistypes a combined word (e.g. slightly), the system may relate his/her input information to the right group of the N-gram database.

According to one embodiment of the invention, proceeding to entering a next word of a current combined predicted word without confirming the previous word, may result in

confirming/accepting the previous word. It must be noted that, although in many embodiments above, the principles of the combined predictive word are being described as relating to up to a 2-gram combined word, such principles may be applied to combined words with more than two individual words.

According to a preferred embodiment of the invention, Correction procedure including the FixLtr function, may preferably be applied to the first word of a combined predicted word.

According to one embodiment of the invention, in a predefined mode, each of some of the keys, such as the letter keys of the first keypad may be designed such that they correspond to a number of applications such as Facebook, Twitter, etc. As such, as an example, in an appropriate mode, providing gliding actions on/from a key in different appropriate directions may correspond to different appropriate applications that preferably may be activated by the system. According to one method, each key may correspond to a group of applications having common characteristics. 0029

A keypad of the invention may be positioned on a pop-up surface (or programmatic pop-up graphic and input mechanism) on the screen of a device. Accordingly, each of the portions of a split keypad may be positioned on a different pop-up. Said pop-up surfaces (or programmatic pop-up graphic and input mechanisms) may preferably have substantially the size of the keypad or its portions, respectively.

In the embodiments and examples described throughout this application, a first group of symbols (e.g. letters) to a key of the first keypad has been described to be selected by a first predefined type of interaction with a key (e.g. taping, double-taping, gliding on or from said key (if on screen key), press and holding, taping on said key and touching a zone of the screen outside said key, etc.) of the first keypad. Then, said selected symbols had been assigned to the keys of a second keypad of the invention. One of said symbols could be entered by a predefined interaction with (the appropriate key of) the second keypad. The same procedure was described for a second group of symbols (e.g. special characters such as digits 0-9) assigned to said key of the first keypad, wherein said second group of symbols could be selected by a second predefined type of interaction with a key of the first keypad, and one of said symbols could be entered as described foe the entry of the symbol of the first group as just described. Clearly, the examples of the embodiments have been provided to describe the procedure of the entry of a symbol through the data entry systems of the invention. It is understood that the order of the assignment of said groups of symbols to the key of the first keypad may be reversed (e.g. the digits group being selected by said first type of interaction, said letters group being selected by said second type of interaction) by people skilled in the art. Note that in all of the embodiments, the type of interaction described is exemplary and may be replaced by any other type of interaction such as those described known by people skilled in the art. For example, a press-and-holding action on a first key for at least a predefined of time to enter said key into a mode instance may be replaced an interaction such as pressing a second key and simultaneously interacting with said first key.

Additionally, many types of interactions with a key (of the first keypad) for assigning a different group of characters to each of them may be considered by people skilled in the art. For example, each of, touching a key and preferably touching another key, touching a key and touching a different predefined location on the screen, touching a key and touching any location on the screen, etc, may be considered as a different type of interaction with said key.

Thus, while there have been shown and described and pointed out fundamental novel features of the invention as applied to alternative embodiments thereof, it will be understood that various omissions and substitutions and changes in the form and details of the disclosed invention may be made by those skilled in the art without departing from the spirit of the invention. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto. It is to be understood that the drawings are not necessarily drawn to scale, but that they are merely conceptual in nature. It must be noted, that any of the systems, features, means, methods, etc., described in this patent application may be used separately or being combined with systems, features, means, methods, etc., of other different embodiments of the invention.

It must be noted that the number of keys, configuration of letters on said keys, key arrangements, type of keys, etc., are being described as examples. They can be designed differently by people skilled in the art by considering the principles of the invention. It must also be noted that a touch sensitive surface may be touchpad, a touch screen, etc.

It must be noted that although during this application there has been described that a user may touch/press on a key such as the center key of a keypad to provide a gliding action, as described in detail before, if the second keypad of the invention is a dynamic keypad, then the user may touch anywhere on the screen/surface or anywhere on a dedicated zone on the screen/surface (e.g. such as the zone not including one or more keys of the first keypad as described in some embodiments), and that touching point may predefined to correspond to the center of said keypad which, preferably, may dynamically form under the touching point or on another location on the screen. The user then may provide the desired corresponding gliding action.

It must be noted that although in some embodiments through this application for describing the principle of the data entry system of the invention, a touch screen has been used as an example, obviously, any type of touch sensitive surface may be considered for the same purpose.

It must be noted that the data entry system preferably in some cases uses more than one interaction with a key and/or zone simultaneously by for example press/glide and holding on/from a first key and pressing/gliding on/from another key/zone. Obviously, if the user desires to use one hand only with the system, then according to one method, the user may first press/glide on/from said first key and remove his finger. Then he may provide said pressing/gliding on/from said another key/zone.

It is noted that, although according the preferred principle/embodiment of the invention a full set of the letters of a language are distributively assigned to few (e.g. four) keys of a first keypad and another same full set of the letters of the same language are distributively assigned to few keys of a second keypad (e.g. to eight keys of a telephone-type) such that any key of said letter keys of said first keypad and any key of said letter keys of said second keypad have at most one common letter, however, minor modifications/variations may be considered/provided to this principle by people skilled in the art. For example, one (or more) of the (e.g. ambiguous) keys of the first keypad and one (e.g. ambiguous) key of the second keypad (e.g. as just mentioned above) (for some reason) may have more than one common character (although this may augment the ambiguity and slow down the recognition capacity of the system requiring more user's input information).

The input system of the embodiments of the present invention may be used with substantially any electronic and/or computerized device, such as cellular phones, GPS devices, remote controls, handheld devices, television settop boxes and music players. This input system is particularly useful for small sized and/or portable devices on which it is inconvenient to place an entire keyboard. In addition, these input devices may be used on appliances not primarily intended for data input, such as washing machines, refrigerators and the like, which conventionally do not include large keyboards.

It is noted in many paragraphs a first and a second keypad have been named to be used with the data entry system of the invention, obviously. The system may instead use a first and a second group of (different) input signals for the same purpose. Said input signals may be provided by any other input means.

It must be noted that although in this patent application "key" and "(predefined) type of interaction with key" has been used to demonstrate different embodiments and methods as described, obviously, any (predefined) input means (e.g. providing a first and a second group of input signals) and any (predefined) type of interaction provided through said input means to provide a (predefined/different) input signal may be used for the same purpose. For example, instead of distributively assigning the set of letters of a language to the keys of a first keypad, said letters may be assigned to a first predefined type of interaction (e.g. single-pressing action on said keys) with each of said keys of said keypad. Also, instead of distributively assigning another set of letters of said language to the keys of a second keypad, said letters may be assigned to a second predefined type of interaction (e.g. double-pressing action, pressing (e.g. and holding) a mode key and single-pressing, etc.) on/with each of said keys of said first keypad. Obviously, based on the principles of the data entry systems of the invention as described above, the distribution/assignment of the letters to said first and second type of interaction with said keys may be such that letters assigned to the first type of interaction with any on the keys of said keypad and letters assigned to the second type of interaction with said key of said keypad have at most one common character. It must be noted that the features such as keypads, key arrangements, number ot keys, assignment of a set of characters of a language to each of a first and second set of keys (e.g. or to each of a second and a second type of interactions with one set of keys), type of keys, type of interactions, etc., as described and shown in this patent application are shown as examples to describe different embodiments of the invention. Obviously, other type of such features can be considered by people skilled in the art.

It must be noted that the first and the second keypad of the invention can have any number of keys. For example, in some embodiments, the second keypad is shown to be a telephone-type keypad having nine keys and in some other embodiments it is shown to have twelve keys.

It must be noted that in addition-to or in replacing-of the correction procedure of the invention, the system may include a method to propose to the user words other that the one predicted by the system. For example, if the predicted word is not the desired word, the user may provide a predefined interaction for example on the screen and the system may show a second keypad wherein its keys include the other words (e.g. having less frequently of use) that correspond to the input information provided until that moment by the user for entering a desired word. Then if the user's desired word is one of said proposed word, the user then may for example provide the corresponding gliding action for selecting/entering his desired word. If any of the words proposed by the system is not what the user desires to enter, then, preferably while holding his finger on the screen, the user may touch another location on the screen and the system shows another second keypad with additional corresponding words (e.g. having less frequently use). And so on.

It must be noted that the procedure of entering a symbol of a group of symbols through the second keypad of the invention wherein the number of symbols of said group of symbols are more than the keys of the second keypad by providing multiple touches to provide multiple second keypads of the invention until the second keypad includes the user's desired symbol as described may be used for any symbol. For example it may he used for entering any punctuation mark character even if the number of the characters of the group is more than the keys of the second keypad, by using such procedure the there is no limit to the number of symbols that a each group may include. The same method may also be used for providing words during the word completion procedure. For example after the used enters into the world completion procedure, if he touches the screen the system may show the second keypad of the invention including some of the words beginning with the user's key presses. As an example if the desired word is not on a key of the presented second keypad, then preferably while holding that finger in 2012/000029 touching positing on the screen the user may touch with another finger a location on the screen and the system may open another second keypad with more words. The user may repeat this procedure if the system does not show his word on said another second keypad.

As mentioned, according to one method, in embodiments such as the media search and word completion, a gliding action in a direction may correspond to a list of words or elements. If the list of the words/items/stems in a gliding action direction is long, then according to one method, several gliding actions in said direction may be provided wherein each of said gliding actions may correspond/include some of said words/items/stems. The order of said words/items/stems may be based on a predefined principle such as frequency of use or alphabetical order.

According to one embodiment of the invention, one of the groups of symbols assigned to a predefined interaction with a key of the first keypad such as the backspace key may be called MENU to which menu functions such as User Guide, Language, Reset Settings, etc. may be assigned.

It is noted that the first and second groups of keys may include substantially any suitable number of keys which may be hard or soft keys or combinations of hard and soft keys. The keys may be organized in various configurations and the characters and symbols may be assigned to the keys in any suitable manner. The symbols of the system may be grouped in any number of symbol modes based of any (e.g. arbitrary) category. Optionally, the first group of keys has fewer keys than the second group, as mainly the first group of keys are intended to be used in entering text.

The term key is meant to include any device which identifies finger actuations including pressure sensors, thermal sensors, acceleration sensors, optical systems for tracking movements of the finger, finger caps and gloves with sensors. The sweeping gestures of embodiments of the invention including embodiments for dialing telephone numbers may be identified using various touch sensitive surfaces including internal or external touch screens and a mouse pad. Alternatively, sweeping gestures may be identified by an electronic stylus, acceleration sensors and/or other sensors for identifying user finger movements. The sensors may be mounted on a surface, on finger caps, on gloves and/or on any other suitable mount. The keys or touch screen may be included with processor 130 in the same housing or may be included partially or entirely in a separate unit connected through wires or wirelessly (e.g., using Bluetooth or Wi-Fi) to the unit including processor 130.

According to one embodiment of the invention the sweeping gestures may be provided in the air and are detected by suitable sensors such as a camera and/or acceleration sensors. 9

In some embodiments of the invention, the data entry systems are adapted to provide synthesized voice feedback on the letters or symbols entered and/or the current symbol mode such that the user need not look at the screen while entering data.

While the above description relates to the Latin alphabet, the principles of the invention may be implemented on other alphabets.

The special characters on the second keypad may be assigned mainly to a same key so that the user may easily remember their location. If the second keypad is a standard telephone- type keypad said special characters may be assigned to the key to which the digit "1" is assigned.

The data entry device in accordance with any of the above described embodiments may be included in a mobile phone, a PDA, a computer or any other device.

According to one embodiment of the invention, after the user provides the input information (e.g. key presses) ambiguously corresponding to several words (e.g. of the dictionary of words of the system), the system may show a first selection of at least some of said corresponding words on the screen. According to one method, said first selection of words may be shown on a (e.g. an imaginary) second keypad (model) of the invention such as a three-by- three matrix keypad as described throughout this application wherein, preferably, a key of said second keypad has at most one word of said first selection of words. According to one method, the user may select one of said selected words by interacting with the corresponding key of the second keypad using a method of interaction such as one of the methods described throughout this application (e.g. tapping/gliding, etc.). If the number of the words corresponding to the input information is more than the number of keys of the second keypad and the user's word is not within the keys of the second keypad, then, the user may provide a predefined interaction, such as a long gliding action, etc., and the system may show a second selection of the said corresponding words on the keys of said second keypad, and so on.

According to another embodiment, after the system shows said first selection of at least some of said corresponding words on the (e.g. an imaginary) second keypad (model) of the invention as described above, the user may select one of said words that begins with the first letter of his desired word. If said word is the only word corresponding to the user's input information, or if the user provides an end-of-the-word signal such as a space character, then, the system may consider said word as the user's desired word and enters it. If said word is not the only word corresponding to the user's input information, then, the system selects all of the words corresponding to the user's combined input information (e.g. the words corresponding to said input information and beginning with said first letter) and shows them on the keys of the 9 second keypad. At this time, the user may select one of said words presented on the second keypad if it is for example, his desired word. If his word is not on any of the keys of the second keypad and the system includes more words relating to said combined input information, then, the user may provide a predefined interaction such as a long gliding action, etc., and the system may show another selection of the words corresponding to the combined input information on the keys of said second keypad, and so on.

It must be noted, that a procedure such as a method of the data entry assigned to a first type of interactions with a key of the keypad of the system may be assigned to a second type of interaction with a key, and vise versus. For example, assignment of two different methods of entering a precise character within a word being entered, by relating said precise character to replacing an ambiguous character or inserting/adding it within the word being entered, to two different type of interaction with a letter key, may be reversed between said two types of interaction (e.g. the first type of interaction may be a gliding action provided on/from a letter key relating to a letter on said key, and the second type of interaction may be a tapping action on a letter/letter-zone on said key during the Precise Character Mode Instance.

It must be noted that the examples of interfaces such as a first and second keypads are used to describe the principles of the dat entry system of the invention. It must be noted that other types of interface may be used for the same purpose, for example, when a user rejects a predicted word, instead of or in additions to the second (e.g. precise) keyboard (e.g. the plurality of second keypads of the invention), the system may have available or may enable/present another interface such as a microphone and a speech recognition system so that the user being able to enter for precise characters (e.g. for the correction purpose).

It must be noted that the term "combined predicted word" is used in some embodiment, generally refers to an N-gram Current word.

Conclusion

It will be appreciated that the above described methods may be varied in many ways, including, changing the specific elements used and their layout. It should also be appreciated that the above described description of methods and apparatus are to be interpreted as including apparatus for carrying out the methods and methods of using the apparatus. The present invention has been described using non-limiting detailed descriptions of embodiments thereof that are provided by way of example and are not intended to limit the scope of the invention. Many specific implementation details may be used. The above embodiments may be implemented in hardware, software, firmware or combinations thereof. It should be understood that features and/or steps described with respect to one embodiment may be used with other embodiments and that not all embodiments of the invention have all of the features and/or steps shown in a particular figure or described with respect to one of the embodiments. Variations of embodiments described will occur to persons of the art. Furthermore, the terms "comprise," "include," "have" and their conjugates, shall mean, when used in the claims, "including but not necessarily limited to."

It is noted that some of the above described embodiments may describe the best mode contemplated by the inventors and therefore may include structure, acts or details of structures and acts that may not be essential to the invention and which are described as examples. Structure and acts described herein are replaceable by equivalents which perform the same function, even if the structure or acts are different, as known in the art. Therefore, the scope of the invention is limited only by the elements and limitations as used in the claims.

The titles used in this application and in the related applications may preferably not been considered as part of the specifications.

It must be noted that all of the interactions such as (key) pressing/tapping actions, gestures, movements, handwritings, drawings, etc., as described throughout this application may be provided in the air. A movement detecting means such as a camera, an accelerometer, etc., may be used to detect and/or perceive/capture those interactions.

Claims

1. A data entry system comprising:

A plurality of input signals to which substantially all of the letters of an alphabet are assigned; wherein the letters assigned to the input signals are divided in a plurality of groups of letters such that the letters of each of the groups have a similar characteristic in their shapes, and;

wherein the first group substantially includes the letters that have a substantially closed circle in their shape.

2. The system according to claim 1, wherein one of the other groups substantially includes the letters that stand on one point.

3. The system according to claim 1, wherein one of the other groups substantially includes the letters that stand on two points.

4. The system according to claim 1, wherein one of the other groups substantially includes the letters that stand on a wide base.

5. The system according to claim 1, wherein said plurality of groups of letters are four groups of letters.

6. The system according to claim 5, wherein each of said four input signals is provided by a different input means.

7. The system according to claim 5, wherein said input signals are provided by key interactions.

8. The system according to claim 1, wherein in the assignment of letters to the keys, the first group of letters has priority over the other groups.

9. The system according to claim 1, wherein said group includes the letters "B", "D", "O", "P", "Q", and "R".

10. The system according to claim 2, wherein said group includes at least the letters "F", "I", "T", "V", and "Y".

11. The system according to claim 3, wherein said group includes the letters "A", 'Ή", "K", "M", "N", "W", and "X".

12. The system according to claim 4, wherein said group includes the letters "C", "E", "G", "L", "S", "U", and "Z".

13. A word predictive data entry system comprising:

a plurality of input signals to which substantially all of the letters of a language are assigned, such that at least one of said input signals; and

a database of words; wherein upon providing at least a one input signal the system predicts a first word of the database, and wherein;

if the predicted word is not the desired word, the user provides a predefined interaction rejecting the predicted word, and the system presents an input interface to the user for entering identified characters.

14. The system according to claim 13, wherein upon receiving a precise character the systm predicts another word.

15. The system according to claim 13, wherein the system predicts a second word.

16. The system according to claim 13, wherein the letters assigned to the input signals are divided in a plurality of groups of letters such that the letters of each of the groups have a similar characteristic in their shapes.

17. The system according to claim 16, wherein the first group substantially includes the letters that have a substantially closed circle in their shape.

18. The system according to claim 16, wherein one of the other groups substantially includes the letters that stand on one point.

19. The system according to claim 16, wherein one of the other groups substantially includes the letters that stand on two points.

20. The system according to claim 16, wherein one of the other groups substantially includes the letters that stand on a wide base.

21. The system according to claim 16, wherein the letters are divided into four different groups, wherein the letters of each group have a common characteristic in their shapes.