KR101312875B1 - Data entry system - Google Patents

Abstract

As shown in FIG. 18, a data entry system for processing opaque data includes clear keys 14207 and 14208 and opaque keys 14203. The system predicts words from the ambiguous input. If you do not want a predicted word, clear input replaces the symbol within the word with the desired symbol. The second prediction is made based on the explicit input.

Data entry system. Keyboard, clear keys, indefinite keys, word prediction system.

Description

Data entry system

Related Applications

This application discloses PCT / US00 / 29647, filed Oct. 27, 2000, the earlier PCT patent applications, which are hereby incorporated by reference in their entirety; PCT / US02 / 22385, filed July 12, 2001; PCT / US04 / 12082, filed April 19, 2004; And PCT / US05 / 19582, filed June 3, 2005.

In addition, the present invention is directed to U.S. Provisional Patent Application No. 60 / 691,107 filed June 16, 2005, the entirety of which is incorporated herein by reference; United States Provisional Patent Application 60 / 693,176, filed June 23, 2005; United States Provisional Patent Application 60 / 695,840, filed June 29, 2005; United States Provisional Patent Application 60 / 695,841, filed June 30, 2005; United States Provisional Patent Application 60 / 697,394, filed July 6, 2005; United States Provisional Patent Application 60 / 704,243, filed July 29, 2005; United States Provisional Patent Application 60 / 704,531, filed August 1, 2005; United States Provisional Patent Application 60 / 710,352, filed August 22, 2005; United States Provisional Patent Application 60 / 718,949, filed September 20, 2005; United States Provisional Patent Application 60 / 724,368, filed October 6, 2005; United States Provisional Patent Application 60 / 727,158, filed October 14, 2005; United States Provisional Patent Application 60 / 732,829, filed November 1, 2005; United States Provisional Patent Application 60 / 737,813, filed November 17, 2005; United States Provisional Patent Application 60 / 741,556, filed November 30, 2005; United States Provisional Patent Application 60 / 786,835, filed March 28, 2006; And US Provisional Patent Application 60 / 797,405, filed May 2, 2005, which claims priority.

The present invention relates to data entry. More specifically, the present invention relates to a data input system characterized by inputting data using an improved method of prediction and key press / glide voice interaction.

Conventional systems and methods for electronically entering characters include the use of a standard keyboard, such as a QWERTY keyboard. However, as electronic devices have become smaller in recent years, new methods have been developed for inputting desired symbols.

In such a way, a number of alphanumeric symbols are assigned to the same key by using a multi-pressing system on the numeric keypad of a standard telephone. One drawback in such systems is that a single key must be pressed several times to enter certain symbols, thereby increasing the total number of key presses and slowing down the symbol input process.

The second way to facilitate symbol input on very small devices was to simply miniaturize the standard QWERTY keypad into the devices. However, such miniaturized keypads are often made clumsy and do not provide enough space between the keys, causing multiple keys to be pressed when trying to press only a single key.

Another way to facilitate symbol input on very small electronic devices is through the use of speech recognition software. Such methods have been used for some time, but have suffered numerous obstacles. Most notably, speech recognition software suffers from indistinguishable distinctions and often requires significant dictionary input for the system to recognize specific speakers and their spoken and spoken habits. In addition, speech recognition software, in mitigating these problems, has grown in size and requires a large amount of processing, making it particularly unsuitable for the limited energy and processing capacity of smaller electronic devices such as mobile phones and text pagers. not.

Accordingly, the present invention improves on the previously disclosed data entry systems by providing, among other devices, additional improved features for prediction and disambiguation using various input means, including prediction keypads, and prediction and speech techniques. It aims to do it. It is also an object of the present invention to provide key input / sliding and improved data input prediction using speech methods. The invention also aims to improve the correction capabilities in such data entry systems.

To this end, the present invention comprises, among other embodiments, a plurality of input receiving means for the input of at least arbitrary symbols, each receiving means being assigned to one or more symbols, the user being able to input the desired input. It is intended for a data entry system that makes use of an input means.

Second input receiving means for receiving a second input, the second input being an exact or definite symbol or signs from the desired input; The system provides a predictive output based on input from at least the plurality of input receiving means. If the predictive output is different from the desired input and the first character of the desired input is different from the first character of the predictive output, the user is predefined to provide a first sign of the desired input. Inputting a correction signal may enable the system to provide a second predictive output based on input previously provided from the user and the correction signal.

In the drawings, like numerals refer to like elements throughout the several views.

1A-1C illustrate exemplary keypads having various numbers of keys that can be used with a data entry system, in accordance with one embodiment of the present invention.

2 shows a database of four keys and corresponding words of an exemplary keypad for use with a data entry system.

3 illustrates a plurality of words proposed on a screen of an electronic device according to an embodiment of the present invention.

4 illustrates an example keypad that may be used with a data entry system, in accordance with an embodiment of the present invention.

5 illustrates an example keypad that may be used with a data entry system, in accordance with an embodiment of the present invention.

6A-6G illustrate exemplary assignments of characters for four keys of a keypad that can be used with data entry systems, in accordance with an embodiment of the present invention.

FIG. 7 shows a keyboard with clear character keys showing a method of predictive data entry system using this type of keyboard.

8 illustrates an example keypad that may be used with a data entry system, in accordance with an embodiment of the present invention.

9 illustrates an exemplary keypad and screen of the device, in accordance with an embodiment of the present invention.

10 illustrates a portion of a database of four keys and corresponding words of an exemplary keypad used with a data entry system, in accordance with an embodiment of the present invention.

11A shows a database of four keys and corresponding words of an exemplary split keypad on an electronic device used with a data entry system, in accordance with an embodiment of the present invention.

11B illustrates a database of four keys and corresponding words on an exemplary split keypad on an electronic device for use with a data entry system using handwriting and key presses, in accordance with an embodiment of the present invention.

12 illustrates an exemplary media player using the data entry system of the present invention for content retrieval, in accordance with an embodiment of the present invention.

Figure 13 illustrates an exemplary telephone using a data entry system, in accordance with an embodiment of the present invention.

14 shows a database of four keys and corresponding words of an exemplary keypad used with a data entry system, according to one embodiment of the invention.

FIG. 15 illustrates a portion of a database of four keys and corresponding words of an exemplary keypad used with a data entry system, in accordance with an embodiment of the present invention.

16A-16B illustrate exemplary keypads with touch sensitive keys that can be used with a data entry system, in accordance with one embodiment of the present invention.

17 illustrates an exemplary keypad for use with a data entry system in which the characters of a qwerty keyboard are assigned to six keys of the keypad, in accordance with an embodiment of the present invention.

18 illustrates an electronic device having unambiguous keys and clear keys for use with a data entry system, in accordance with an embodiment of the present invention.

19A illustrates four keys and a database of corresponding words and an output method of an exemplary keypad for use with a data entry system, in accordance with an embodiment of the present invention.

19B illustrates a database of words used with a data entry system and output provided to a user, in accordance with an embodiment of the present invention.

19C-19D illustrate outputs presented to a user, according to one embodiment of the invention.

20-20A illustrate exemplary media players having a circular sensitive keypad for use with a data entry system, in accordance with an embodiment of the present invention.

21 illustrates an example scalable electronic device using a data input system, in accordance with an embodiment of the present invention.

22A-22B illustrate exemplary telecommunication devices using a data input system, in accordance with an embodiment of the present invention.

Figure 23 illustrates an exemplary keypad of the present invention that can be used with data entry systems, in accordance with an embodiment of the present invention.

FIG. 23A illustrates an example keypad of FIG. With elliptical keys and another key arrangement, in which keys of a keypad are divided into two groups on an electronic device for use with a data entry system, in accordance with an embodiment of the present invention. Illustrated.

FIG. 24A illustrates key interactions of a user's fingers and keypad interacting with electronic devices that duplicate keys used with a data entry system, in accordance with an embodiment of the present invention.

FIG. 24B illustrates fingers of a user wearing finger caps, providing key interactions of a keypad and interactions that overlap keys used with a data entry system, in accordance with an embodiment of the present invention. .

The data entry system according to the present invention provides a system for inputting at least any symbols / text using a reduced number of keys (eg, a keypad having them) (eg, hard, soft, virtual dual, etc.). It starts. The data entry system may use a plurality of symbols including at least alphabet characters of at least one language, the symbols being distributedly assigned such that at least two characters are assigned to at least one key. Can be.

In the present systems (e.g., predefined) interaction, such as, for example, the act of pressing a key, may, unambiguously, be the key (in this patent application, the key may be called "unclear", and the interaction may be " Correspond to any symbols / characters assigned to "unclear" interactions). As described above, for example, to enter one of the symbols (e.g., symbols) assigned to a key, the user provides a predefined interaction with the key, such as pressing a key, and Of speech information (e.g., the name of a character (e.g., a character)) corresponding to the symbol for selecting the symbol (e.g., a character) from among the symbols (e.g., characters) assigned to the interaction. Speech) may be detected and analyzed based on at least one of the user's voice and / or movements of the user's mouth.

The number of symbols may include substantially all basic symbols, such as letters and numbers, punctuation marks, and instructions for entering or editing text or manipulating a computer. For example, to enter a punctuation symbol, the user can press the corresponding key and utter that symbol. Other symbols, such as macros, word-of-a-words, and the like, may also be assigned to a reduced number of keys.

Even when there is no utterance, it can be considered as utterance information, at least one of the symbols assigned to the key being input by providing interaction with a predefined key when there is no utterance, and the lack of utterance being detected by the system. Can be. For example, to enter a predefined punctuation symbol assigned to a key, the user may use the predefined pressing behavior of the key corresponding to the symbol without speaking.

According to one method, the reduced number of keys can duplicate the telephone-type keypad. The letter assignment may resemble the letter assignment of a telephone-type keypad.

According to another method, the letters of the Jamo can be assigned to four keys of the reduced number of keys.

According to one method, symbols such as characters that have a predefined utterance that are indistinctly similar and are difficult to distinguish from each other through corresponding utterances may be assigned to different keys separately from one another.

In addition, other related input technologies such as at-least-a-portion-of-a-word (e.g., corresponding speech information (e. G. (portion-by-portion data entry system using a press / glide action on one or more corresponding keys), which may be predefine). Here, other data entry methods such as, for example, complete (eg, hard, soft / on-screen) PC keyboards and / or handwriting recognition systems may be used to enter the text / data. Can be combined / used with (eg, pressed and spoken) data entry systems.

Data entry systems (eg, providing correct or unambiguous symbols) may use (eg, very) a small number of keys (eg, here at least some of the keys are unclear keys) (eg and a user's voice or candidate) Word prediction system (e.g., symbols / words of words) to provide a very accurate or clear data / text input system by minimizing user interactions such as minimizing user interaction to select one word among words. The principles of word prediction systems are known to those skilled in the art) based on unclear key press actions corresponding to letters and using a database of words and the corresponding key press information of each word. . Using the press-and-speak data entry system of the present invention, a user can provide at least one correct or definite symbol of a word or press (unclear) keys corresponding to the remaining symbols of the word. You can enter a word. By considering at least one correct or definite symbol of a word and considering key presses corresponding to the remaining symbols of the word, the system can predict the word more accurately or clearly.

The push / slide and voice information (eg with / without voice) data entry systems of the present invention are combined with other data entry systems, such as a word prediction system based on key-presses provided on a keypad (with unknown keys). Can be. The principles of word prediction systems based on key presses (eg based only on key presses) are known to those skilled in the art. The word of the language is entered correctly or clearly at least one of (eg preferably at the beginning) letters of the word (eg or symbols (words can also include symbols other than letters)) It may be entered using the press / slide and speech data input systems of the present invention to provide only key presses (without utterance) corresponding to other characters (eg or symbols). For example, using a telephone keypad, to enter the word "ball", the user presses the first letter of a word (such as "b"), presses the key corresponding to that letter, and provides a speech corresponding to that letter. Thereby pressing (e.g. after) speech and corresponding keys to other characters of the word (e.g., "all" which is a chain of characters in this example). In this example, providing only a sequence of key presses (eg without speech) for all the letters of the word "ball" (including the letter "b") (e.g., corresponding thereto) The word "call" may also correspond. However, in the above example, since the initial letter "b" is entered correctly (or clearly), the system does not consider the word "call". According to another example, to enter the word "home" using the same keypad, the user may first press the keys corresponding to the initial characters "ho" without utterance. He can then press the key corresponding to the next letter "m" and speak the letter (eg, entering it correctly or correctly). The user can then press the key corresponding to the last letter "e" without speaking. The chain key presses may correspond to the word "gone", but the word "home" will be selected / entered because the system correctly recognized that the third letter of the desired word was "m".

At least provide (e.g., the first) character of the word accurately (e.g., taking into account the first type of information consisting of the provision) and the remaining key presses (e.g. key presses corresponding to the remaining symbols of the word) (Eg, in combination with at least considering a second type of information consisting of the provision), in most cases the system itself is the desired corresponding within the dictionary of words that match the combined information. You can find words. In this example, by correctly (clearly) entering the letter "b", the system selects the word "ball" without requiring additional user interaction.

According to one method, the user may provide key presses corresponding to the letters of the word and may speak any of the letters during the key presses, even if the utterance is not synchronized (eg provided simultaneously) with the corresponding key press (es). You must understand that you can.

According to an embodiment of the present invention, the user may provide a key press corresponding to the letters of the word and utter the word.

Word prediction systems based on key press information use at least a word dictionary and a corresponding key press database.

According to one embodiment of the invention, during input of the text, the user provides key presses corresponding to the letters / symbols of the word and provides speech information corresponding to the word (eg letters of the word). (E.g., words that exist in a dictionary, out-of-dictionary words, chain symbols that do not exist in a word dictionary, etc.). On the other hand, if the user desires, providing only the key presses in the dictionary's word database by inputting (e.g. providing) the corresponding key presses (e.g., corresponding to the letters of the word) is a large number in the system's word database. May correspond to words of (eg, selections). For example, for higher accuracy (clarity) and a reduction in the number of choices, the initial (eg and / or intermediate, and / or last) key (s) of the word (such as initial and / or intermediate, and / or Or while pressing the key (s) corresponding to the last symbol (s), the user may utter the first (eg and / or intermediate, and / or last) letter (s) of the word (eg, tactical above). As one example, providing key presses and utterances corresponding to at least one letter of a word may accurately (clearly) enter the at least one letter). Thus, in view of the at least one correct (clear) letter and the key presses corresponding to the remaining letters of the word, the number of selections (eg candidate words) can be significantly reduced. Throughout this application, such systems may be referred to as "combined data entry systems" or rather simply data entry systems, and such terms are used interchangeably in the context of the present invention unless specifically stated otherwise. .

The basic premise of such joint data input systems has already been described as in the references contained herein, filed by the inventors, and embodiments of the present invention constitute systems and methods for enhancing such joint data input systems. do.

Pressing of the present invention to allow input of text by combining at-least-a-portion-of a word input methods (e.g., referred to as the partial-part data entry system of the present invention) -And-vocal data entry systems have been described in detail in the references contained herein, filed by the inventors. According to one embodiment of the word prediction data entry systems of the present invention, instead of entering a single symbol, the user presses / slides and speaks data of the present invention to accurately (clearly) enter at least a portion of the word. By using an input system it is possible to enter a word-part and provide only key presses (without speech) corresponding to other letters of the word. For example, to enter the word "predefined", the user may enter the "pre" portion by pressing / sliding on the corresponding key (s) of the keypad, without speech, the remaining characters of the word. The keys corresponding to "defined" can then be pressed. Obviously, during key press operations corresponding to the remaining symbols of the word, if desired, the user may provide a utterance corresponding to at least some symbols (e.g., utterances of the symbols) corresponding to the key presses to display the at least some symbols. You can enter it correctly.

As discussed above, using the present (eg, combined) system, a user can enter a portion of text based only on key inputs (no speech). For example, the user may know in advance that the word he would enter may be recognized by the combining system based only on the key inputs. In this case, the user may not provide any speech during word input. For example, by using a telephone keypad, the word "array" can be recognized based on only key inputs corresponding to those characters (no need to speak any characters).

It should be understood that the above-mentioned system may be used with a predefined keypad, such as a telephone-type keypad or other keypads such as those described in the references contained herein, filed by the inventor. . The keypad may have any predefined number of keys. 1A-1C show three types of keypads having at least 12, 6, and 4 keys, respectively, as shown and described in the preceding patent applications. 1A shows a telephone-type keypad. 1B shows a keypad with at least six keys, wherein the six keys represent at least letters of the alphabet's alphabet. On the one hand, the letters are preferably assigned to an arrangement / keys on the keys such that letters with utterances that are unambiguously resembling each other are / preferably assigned to different keys (eg described above). On the other hand, the characters are minimal if the key presses (eg values) corresponding to words in the (eg English) dictionary of the word database of the system have the same (eg order) key presses (eg values). Arranged to provide (eg corresponding to word groups) word groups (e.g. to reduce confusion). 1C shows a keypad with at least four keys, where the four keys represent at least characters of the alphabet's alphabet. On the one hand, the letters are preferably assigned to an arrangement / keys on the keys such that the letters with utterances that resemble indefinitely with respect to one another / preferably are assigned to different keys (eg described above). On the other hand, the characters are minimal (eg confusing) in which key presses (eg values) corresponding to words in the system's word database (English) dictionary have the same (eg order) key presses (eg values). May be arranged / arranged to provide (eg, correspond to word groups) word groups.

Using more keys that assign (eg, represent) a smaller (eg, fewer) number of language characters to each key is the key press (eg, spoken, if provided) information provided by the user for entry of a word. It is to be understood that the number of selected word candidates corresponding to the number will be minimized.

According to one embodiment of the invention, after entering key presses corresponding to a word and providing a utterance corresponding to at least one (e.g., first) sign of the word, an end-of-a-word ) Signal may be provided. The signal may also press the "space" key so that a space symbol is entered at the end of a word. According to another method, another word-ending signal may be provided so as not to include a space symbol at the end of the word. For example, entering a (defined) PC keyboard command or function or a (defined) punctuation symbol after the word may also be considered as a word-end signal.

According to another method, a user enters the first letter / part of a word using the push / slide and speech data entry systems of the present invention, and at the end of the key presses (without providing a word-end signal), It is possible to provide key presses corresponding to the last symbols of the word without utterance, and the user can continue by inputting the first letter / site of the second word using the press / slide and speech data entry systems of the present invention. The next word can be entered, and the system can be predefined to understand that the second utterance corresponds to at least the first letters of the next word, so that the first key press corresponding to the utterance (eg from that first key press) Starting key presses) can belong to the next word. This can be regarded as the word-end signal of the first word. Then, according to one predefined method, the system may enter a spacing symbol between the words (eg, an automatic spacing procedure). According to another predefined method, the system does not enter a space symbol between the words (eg to glue the words together).

According to another method, the idle time of at least the defined laps between the input of two words can be regarded as the word-end signal of the first word of the words.

As mentioned above, using the word prediction system of the present invention for entering a word, after the user provides key press information, if the user's information, at least one of at least three types of feedback from the system is Can exist:

1- The system can associate only one word with the user information. The word can be entered as the final result.

2- The system may associate a selection of at least two words to the provided information of the user. The system may present (eg, highlight) the word with the highest priority in relation to other words within the selection. If the word is what the user wants to enter, according to one method, the user can continue to enter the next portion of the text (eg a word) and the system automatically enters the word. If the entered word is not what the user wishes to enter, according to one method, the user can use a selection / modification means such as a selection / modification key, which for example presses (for example and holds) the key and the ( First) to utter at least some of the letters so that the system recognizes the words of the selected words. It is understood that the user may not retype key presses corresponding to the word while uttering at least some of the (first) characters of the desired word. For the modification method, the system can use the original key press information and the speech information for the modification procedure. When the correction speech information is provided, the user can release the correction key. Pressing (e.g. and holding) the key may inform the system of the beginning (e.g. and end) of the fertilization procedure. In order to avoid unnecessary longer corrected utterances, it is understood that the system can present the entire words of the selection to the user so that the user can stop uttering, if unnecessary, before uttering all the characters of the selection. It is understood that other means such as double pressing, sliding, and the like may be used instead of the pressing (eg and holding) operation. According to another method, instead of using the correct / select key, the user may point to a word and provide speech information using a pointing device such as a stylus, a mouse pointer, or the like. After the modification procedure is over, the user can finish the pointing procedure. Indicating (eg, and maintaining) the pointing device at the pointing position can inform the system of the beginning (eg and end) of the fertility procedure. Instead of speaking the letters of the desired word among the candidate words, the user may speak the word.

According to another method of selection, if the system associates / selects at least two words for the user's information and the words presented to the user (eg, highlighted) are not what the user wishes to enter, The means, such as a multi-directional key, can be used to navigate within the selected words to select the desired word. Demonstrating and selection methods using navigation / switching keys to select a word among a number of selected words are known to those skilled in the art (e.g., in pop-up lists, pop-up menus). Navigation up and down, etc.).

3-The system may match no words to the user's information provided for input of the desired word (eg, non-dictionary), and, according to one method, then the user may, for example, enter a correction key. By means of pressing (e.g. and holding), a corrective means can be used, and the symbols (e.g. letters) of the word can be uttered without providing the corresponding key presses again. The system may use the original key press information and the speech information during a modification procedure for entering a word. It is understood that the words can be added to the dictionary (eg, automatically or manually) if desired.

Priorities given / assigned to words in the dictionary (eg based on frequency of use) may be an important issue for better acceptance of the combined data entry systems of the present invention by the user. The priorities may be assigned to minimize the user's involvement in selecting a word among a group of words selected by the system with respect to key press (eg, and spoken) information provided by the user. As an example, FIG. 2 shows one page of a dictionary of the word database 12600. Column 12601 shows some words of the database. Column 12602 shows the frequency of the use values of the words associated with each other. The frequency of use values may be predefined to be considered as priority values assigned to words. Column 12603 shows key press values assigned to words based on key values (e.g., 1, 2, 3, 4) assigned to keys 12605, 12606, 12607, 12608, respectively, of keypad 12609. .

According to one method, for example, to enter the word "doing", the user may press keys 12606, 12608, 12606, 12606, 12608 (eg, key press value 24224) without utterance. In this example, four words 12604 corresponding to the key presses may be selected by the system. The system may automatically enter the word with the highest priority (eg "doing"), for example, unless:

According to one method, if the user wishes to enter another word such as the word "coins", the user can use correction / selection means as described above. For example, the user may press and hold the select / modify key and speak the letter “c”. The system can understand that the user intends to enter a word of choice beginning with "c". The word is the word "coins". According to another method, the user can navigate and select the word "coins" within a popup list provided by the system and including the four words. Furthermore, according to another method, the system may first show the word "doing" which has the highest priority first. The user can press a predefined modify / select key. The system can show "dying", the second highest priority word of the four words. The user can press the predefined modify / select key again. The system can show the word "pupil" which is the third highest priority word out of four words. And finally, the user can press the predefined modification / selection key again. The system can show the word "coins", which is the fourth highest priority word of the four words. The user can accept the word, for example by continuing to enter the next word.

-If the user wishes to enter the word "dying", the user can use the correction / selection means as described above. For example, the user can press-hold the Select / Modify key and utter at least the letter "dy" (one or more letters because the two words, "doing" and "dying" begin with the same letter "d"). It may be necessary to speak). The system can understand that the user intends to enter a word of choice starting with the letter "dy". The word is the word "coins".

According to another method, for example, to enter the word "coins", the user presses keys 12606, 12608, 12606, 12606, 12608 (e.g., key press value 24224) and at least presses the letter "c". You can speak. Only one word (eg, "coins") in the database corresponds to the key presses and utterances. The system can automatically enter the word (eg "coins").

According to one method, for example, to enter the word "clive", the user presses keys 12606, 12608, 12606, 12608, 12607 (e.g., key press value 24243) and utters the letter "c". can do. Two words (eg, "cycle" and "clive") in the database may correspond to the key presses and utterances. The system may automatically enter a word having the highest priority among the two words (eg, “cycle”). The user can scroll to the next word using the transition method as mentioned. The system then shows the word "clive", which is the second word corresponding to the key press and utterance information (e.g., although the word "noise" has the highest priority corresponding to the key entries, the word is presented to the user). Do not begin with the letter "c" (the system ignores the word).

The following is an exemplary overview of one of the conversion / navigation selection embodiments:

Providing one word key presses without utterance selects all those words:

   ■ The word with the highest priority, regardless of the initial letter (s)

      ● the next higher priority word, regardless of the initial letter (s)

         ○ etc.

Providing key presses and vocalization of at least one (first) character of a word selects the group of words that begin with that letter (s):

   ■ the highest priority word in the group

      The next highest priority word in the group

         ○ etc.

According to one embodiment of the invention, the first word of a group of words having the same key presses (eg, the same key press value) may be predefined to be selected by providing the corresponding key presses without utterance. Another word in the group having the same key press value and having the same initial character of the first word and having a second higher priority (eg lower than the first word and higher than other words in the group having the symbols) If desired, the user can provide key presses and utter (at least) a first letter. For example, to enter the character "doing", the user can provide the corresponding key presses without utterance. Thus, in order to enter the word "dying", the user can provide the corresponding key press leagues and enter the letter "d". According to one embodiment of the invention, the highest priority word is assigned to the provision of the corresponding key presses without providing the spoken information, the user provides the key presses and utters the at least (first head) letter of the word and the systems When selecting another word that corresponds to information provided by the user (eg, key presses and utterances), the system may select the word if desired by the user, including the word corresponding to only the key presses within the proposed selection. You can do that.

After becoming familiar with the system, the user can remember a number of words that can be entered based on only those corresponding key presses (eg without speech). A user assigns the words to (e.g., key) interactions only, such that a word he has the same key press values and starts with the same letter consists of the provision of the key presses and the utterance of the (first) letter (s) of the word. Can be entered (eg, "doing" without a utterance, "dying" with at least the utterance of the letter "d").

According to one embodiment of the invention, the user can add his desired words to the system's word database. He can also change the priorities of words (eg based on the frequency of use of the user).

The word dictionaries of the system and the priorities of the words therein may be tailored to the needs of market segments (eg, medicine, insurance, etc.). The system may also "learn" words that the user uses and assign priorities to those words accordingly.

During data entry such as text, the user may consider several parameters to minimize some user interactions, such as utterance, modification / selection procedures, and the like. For example, a user may remember that many common words, many long words, and the like may often be entered based on only those key presses.

As mentioned in the above-mentioned patent applications, in the combined data input systems of the present invention (e.g., including predictive data input systems), for reasons such as ease of use, they are also assigned to / represented by keys that also represent characters. The punctuation or commands that are to be input may be input by different predefined interactions regarding the interaction used for the input of the characters. For example, a punctuation or command may be assigned to the pressing of two keys, preferably simultaneously (eg by pressing and holding on the "shift" key and pressing the key corresponding to the punctuation or command). They may be assigned to another type of interaction, such as a slide action on that key or a longer single-press action (eg, shorter press actions can be used to enter characters or other symbols contained within a word). . These things have already been described in detail in the references contained herein, filed by the inventors.

According to another method, after entering a word (e.g., by providing a word-end signal), the user presses a key that represents a few characters of the language and at least a symbol, such as a punctuation / command, without a voice to perform a speechless push operation. You can enter assigned predefined punctuation or commands. These things have already been described in detail in the references contained herein, filed by the inventors.

Based on the correction / selection principles as described, instead of correcting / selecting words immediately after their entry, the user may then (eg after entering the entire document or taking notes during a meeting) After you make notes) and then edit / modify them, you can continue to modify / select them.

According to one embodiment of the invention, the system stores at least one of the key presses and their corresponding utterances, provided by the user for the entry of a word (e.g., the system is a precise (clear) letter (s) And ambiguous key presses). As shown in FIG. 3, if a user wishes to change a word (eg, “codes”) suggested by the system based on the user information, to the word he wants, he will be on the screen of the device 12700. Text 12701 may first be visualized and then directed 12705 towards the unwanted word “codes” 12702 (eg, using a mouse or stylus to indicate or select a word, for example). The system can store key press information previously provided by the user (e.g., and utterance information. As mentioned in previous applications, not speaking is considered by the system as utterance information) and stored in the user information. Other word options 12703 and 12704 may be proposed. The user may use a pointer, such as a mouse or a stylus, for example, to select a desired word from among the words proposed by the system, using a means such as a direction / navigation key (eg, within a pop-up menu). select "ducks" 12704, or utter at least some of the letters of the desired word to allow the system to recognize the word among other proposed / candidate words (e.g. in this case refer to the current example and It may be sufficient to utter the letter "du" to compare the word "ducks" with another word "docks", so that the user may not utter all the letters of the word "ducks". The word may be recognized and printed before finishing (for example, as also described in the present case, the system may store and use key press information). Standing, so the user can not provide the key presses for that word again). In this example, the user originally entered only key presses, so the system selected the word "codes" with the highest priority with respect to the key presses. After indicating the word, the user can utter at least the letters "du" so that the system recognizes and selects the desired word "ducks" among other proposed words (e.g., the word "docks" matches "d"). As a second character, beginning with the letter "o").

The combined data entry systems of the present invention require key presses corresponding to words and may require a small amount of utterance for input of in-dictionary words. As mentioned, for non-dictionary words the user can enter words one by one, for example by pressing keys and uttering symbols.

According to an embodiment of the present invention, in order to enter a word (or more words), such as a non-dictionary word of a document, the user can provide key presses corresponding to the word without utterance and continue the document. You can enter other words in. The system may match the information corresponding to the (non-dictionary) word with an existing word to provide an unwanted word, or may not match a word in a dictionary with the information. In the present case, instead of the word, the system may have been pressed (e.g., if the user has also provided a utterance corresponding to, for example, one or more letters of the non-dictionary word), the system may use letters and key press values or other characters in the word. Can print other predefined symbols about the symbols) (e.g., or define predefined symbols such as, for example, "*" or "X" symbols), such as the values of the keys. have. The user can then take out the document and indicate (as described above) the wrong / unwanted word (s). The user can then speak the letters corresponding to the word without again providing the key presses corresponding to the word (eg, the system can remember the key presses). If the user forgets the word utterance, he can hear the utterance of the word he has already provided during the input of the original key presses corresponding to the word. By indicating a wrong / unwanted word, the system may play a voice corresponding to a word that was provided by the user during word entry. For example, to enter the word "Cannes" (such as French words) that may not be included in the English word dictionary database, the user can press the corresponding keys and utter a word (such as "kan" when pronounced in French). ). The system may not find a word corresponding to key presses and utterances (eg, the utterance may not correspond to an English (eg word) utterance having key press information). In this case the system can store the speech information and the key press information and on the other hand can provide chain symbols corresponding to the values of the assigned keys (eg "212234") pressed for the input of words. For example, later, the user may indicate a chain symbol, and if necessary, the system may provide a utterance provided by the user to indicate that he is in the middle of entering a word. The user can then speak a word letter by letter (eg, without again providing key presses). The system provides the non-dictionary word "Cannes" taking into account the original key presses and the user's recent speech.

The embodiments described above can be very useful in some circumstances. For example, during a meeting or in a meeting room, a user may take / enter a note by providing key presses corresponding to the word he is typing (eg, the utterance of the word), and when needed, a very short utterance he enters ( Letters, syllables, (short) words). Also, for example, after a meeting ends, the user retrieves the document and continues to edit / edit the memo as described above.

According to another method, the user may enter an abbreviation for faster data entry. According to yet another method, the user presses one or more keys (eg random or predefined, non-performing keys) to provide (randomly) concatenated characters that correspond to words to be entered correctly (clearly) in the future. You can speak words. Then he can take out the document, point out the chain symbols, listen to his utterance, retype those key presses correctly, and provide the utterance of multiple characters for correct entry of the word if necessary. .

According to another embodiment, for entering a word, such as a non-dictionary word, the user may only provide corresponding key presses (eg, without providing utterance). If the user then points to a word, he may remember the word (eg, the utterance of the word) based on key presses and correct the word as described above.

Other selection / modification methods, such as navigating a pop-up list, can also be used for the same purpose.

To conserve memory, the system may include means such as keys to allow the user to remember only voice (eg, his voice) for a brief period, such as when the user is in a meeting or conference room. The user may memorize the voice only during the input of words-to-be-corrected / entered-later.

According to another method, subsequent-correction / to-enter-words may be entered by handwriting (eg words may be recognized immediately or later by a handwriting recognition system. The user may enter them in text and as described above). Can be entered later by modification methods).

Many methods and means can be used with the present system to enhance the ease of use of the combined (eg, including prediction) data entry systems of the present invention.

According to one embodiment of the invention, the word dictionary of the system may include a word-part (eg of that language) for use with the data entry system (eg including prediction) of the present invention. Depending on the configuration of the system, the word-part may be included in a word dictionary or may constitute a separate database.

In addition, to remind the user that a word can be entered based only on key presses, after entering the word, the system may highlight the word prominently on the screen of the device, for example by printing in bold font or , Printing in different colors, flickering, etc.

If the word does not exist in the database, but can be present in the database by deleting a portion, such as its end, the user can delete the portion and then enter the deleted portion at the end of the word. For example, the word "prescription" may be included in a database but the word "prescriptions" is not. For example, after providing information corresponding to the word "prescriptions", if the system does not select any words (e.g., but the user knows that "prescription" exists in the database of the system), The space (BkSp) "key can be used to delete the last key-press corresponding to the last character" s ". The system selects the word "prescription" and the user can add the letter "s" to the end of the word.

Several predefined single characters (e.g. "a", "e", "i" in English) can be assigned to different keys and can be used for interactions such as pressing on those keys without providing vocalization. Can be input in advance.

During the entry of a word, the user can enter a portion of the word using the partial-part data entry systems of the present invention. For example, a user may start pressing keys corresponding to the (start) letters of a word without utterance, press / slide the key corresponding to the middle or final part of the word and utter the portion (or vice versa). , Etc.) This has several advantages. For example, correctly (clearly) entered portions can help make the whole word easier to recognize. It also helps to enter words faster (eg fewer key presses). For example, to enter the word "recognition", the user can first press the keys corresponding to the letters "recogni" without utterance, and then correspond to the letter "t" (first letter of the "tion" part). You can press the key and utter that part.

As mentioned above, other data entry enhancements, such as using language models / rules, word completion systems, automatic correction systems, and the like, may also be used with the data entry systems of the present invention. As already mentioned, fingers, finger caps, etc. may be used in place of or in addition to the keys of the data entry system of the present invention. In addition to or in place of the user's voice, movements of human parts such as the user's lips may also be considered. These have already been described in detail in the references contained herein, filed by the inventors.

According to one embodiment of the invention, other data entry methods, such as multi-tap or handwriting, can be used to accurately (clearly) enter at least one of the symbols, such as the first letter (s) of a word. have. Information corresponding to the remainder of the word may be entered by providing (eg single) key presses corresponding to the remaining symbols of the word with / without utterance.

Press to accurately (clearly) enter at least one of the symbols of a word and to provide (eg, indefinite) key press (es) corresponding to at least one (eg preferably all) of the other symbols of the word Using a press / glide-and-speak and / or handwriting input and recognition system, and / or a complete (eg pc) keyboard is fast (eg handwriting alone is slow) and accurate (clarity) A word prediction system based on only key presses can provide one (eg, disappointing, inaccurate, and with other limitations known to those skilled in the art) and may also require less utterance or handwriting. The system also requires a small amount of memory and processing power / speed. For example, a speed recognition system (eg, combined with the corresponding key press (es)) may exclude the utterance of words (eg, generally, character input may be sufficient for word input). The handwriting recognition system can also exclude word level input and can generally operate based on character input. This is because the use of the present invention's predictive data entry system employs accurate (clear) character (s) input methods combined with (eg, unclear) key presses (e.g., reduced corresponding keypad). And / or significantly minimize the use of a writing system.

Providing key presses and uttering at least some of the (e.g., initial) letters of the word may provide words that provide key presses and may not be very accurate (clear) and use words in a language that uses a large amount of memory and processing power. Compared to speech, it is possible (eg, coupled data input) to allow the system to provide better accuracy (clarity) and use a small amount of memory and processing power. Thus, at least for a device with limited memory and processing power, a letter-by letter that requires a very small amount of terminology (eg letters, numbers, some punctuation, some commands). It is recommended to use the (letter) level approach.

Obviously, it should be noted that in order to select / indicate a word to be modified / changed, the pointing position may be predefined such as to indicate the word, place it before the word, place the word after it, or the like. For example, to indicate a word, the cursor can be placed after the last letter of the word or between letters of the word.

According to one embodiment of the invention, if a word does not exist in the dictionary but contains or excludes a prefix (such as "re") or a suffix (such as "tion") in the system database (of a word), the system is a prefix. You can suggest words with or without the suffix. The user can then add or delete other parts from / to words. For example, if a user attempts to enter the word "understanding" and the system does not find such a word by recognizing that the word may include the suffix "ing", the system will search for the word "understand". (E.g. by not considering the last three key presses provided for the word). If the system finds the word "understand", the system suggests it to the user and the user can acknowledge the word, for example by entering the remaining letters "ing" of the word. According to another method (eg, by analyzing provided key presses), the system can automatically add the prefix or suffix of the word.

As mentioned above, preferably, the keypad used with the combined data entry systems of the present invention may be one that maximizes the ease of use and accuracy (clarity) of the present system. Parameters such as the number of keys, the arrangement of characters on the keys, and the like may be considered different with each keypad. 4 shows, for example, a keypad 12800 in which four keys 1281-12804 correspond substantially to all letters of the English alphabet (eg, letters are assigned to four keys in a distributed manner). The keypad can be designed for use with the combined (eg, including prediction) data entry systems of the present invention. For example, on the one hand, for example, characters of character groups such as "m, l, n", "p, b", "d, e", "j, k", "t, d", etc. Unclearly resembling letters such as those can be assigned to different keys separately to maximize the accuracy (clarity) of the push / slide-and-voice data entry systems of the present invention, while "e" character groups such as, o, i, a ", and" s, d, r ", etc. are assigned to different keys to minimize the number of words having similar key presses when using the (e.g., combined) word prediction systems of the present invention. Can be assigned to each other separately. For example, by doing so, words such as "drives, drive", or "forces, forced, forcer", etc. may have different corresponding key presses. It is understood that the methods discussed above can be applied to any keypad with a few keys, such as a keypad with twelve keys (eg, a keypad with a telephone-type keypad key arrangement).

As mentioned above, when entering a word by the combined (including predictive) data entry systems of the present invention (e.g., as described above, based on key presses corresponding to the letters of the word, speech is included). Or by providing utterances of at least one (eg, initial) character of the word), the user preferably enters at least one of the initial characters of the word into the push / slide-and-utterance data entry method of the present invention. By typing (e.g., accurately) and pressing the keys corresponding to the remaining characters of the word. According to one embodiment of the invention, if the user wishes to provide other symbols at the end of a word, such as, for example, at least one word, at least part of a word (eg a suffix), etc., the user presses / slides of the invention It is possible to provide at least one of the initial characters of the word / part by the speech data entry method, for example, to continue pressing keys corresponding to the remaining characters of the word. This allows the generation and entry of combined words even if they do not exist in the system's word dictionary database. For example, in order to enter the word "friendship" in two parts "friend-ship" (e.g. the word friendship is a system), taking into account the words "friend" and "ship" that exist in the method and system's word dictionary database. May be present in the database of the word prediction data input systems of the present invention as described above (eg, corresponding to the letters of the word without speech). Provide key presses, or provide key presses corresponding to at least the (e.g., the first) letter of the word (e.g. "f") and utter the letters, and the remaining symbols of the word (e.g. "riend") without speech. To provide key presses for "). The user then provides a key press corresponding to at least the (eg first) letter (eg "s") of the word and utters the letter, corresponding to the remaining symbols (eg "hip") of the word without speech. By providing key presses, one can continue to enter the next word "ship". The key press and utterance corresponding to the first symbol of the second word can tell the system the beginning of the second word and the end of the immediately preceding word.

As mentioned, the user may enter at least one word-part, such as a suffix at the end of the word, using the same method. For example, to enter the word "management" as a two-part "manage-ment" (eg, the word "manage" and the suffix "ment", considering that the word and the word-part exist in the system's word dictionary database) ), The user can enter a word by the method described above.

In addition to the words above, it should be noted again that the system's word dictionary database may also include word-parts corresponding to at least one language. Even words that include from multiple sites or words, such as the word "workmanship" (eg, "work-man-ship") that includes several short words, can be input through the same method.

According to another method, chain symbols (e.g., words in a database, non-dictionary words, arbitrary chain symbols, etc.) are for example character-by-character presses of the present invention. It can be entered / attached at the end of a word using / slide-and-voice data entry systems. For example, in order to enter the word "management" as a two-part "manage-ment", taking into account the word "manage" present in the system's word dictionary database, the user first enters the word "manage" as described above. (Eg, provide key presses corresponding to the letters of the word without utterance, or provide key presses corresponding to, for example, at least one (e.g., initial) character (eg, "m") of the word). And uttering the letter and providing key presses corresponding to the remaining symbols of the word (eg, "anage") without utterance. A user may enter the chain symbols "ment" at the end of a word using symbol-one-one push / slide-and-utterance data entry systems of the present invention.

According to one embodiment, a word-end signal may be needed at the end of the word input, such as pressing a predefined key so that the screen does not shake when the user continues to enter the next word (eg, attaches it to the first word).

The above embodiments may enable the use of a word database with a significantly reduced number of words with the combined data entry systems of the present invention. For example, words ending with common suffixes (eg, "ed", "s", "ing", "ment", "ture", etc.) may not be included in the database. Also, words that consist of multiple shorter words (eg, shorter words exist in the database) may not be included in the dictionary. The system may also include an auto-correction system to automatically correct the combined words, for example. For example, if the user enters the word "memorize" and adds the word "ing" at the end, the system automatically modifies the combined word "memorizing" to provide the word "memorizing".

As mentioned above, if a word is entered using the word prediction system of the combined data entry systems of the present invention, the system may, for example, suggest a word that the user does not intend to enter. Also, for example, the system may not match the user's (eg key presses, and voice if present) information with any words in the database (as mentioned above, the system may in this case have the user's key). The system may print predefined or arbitrary chain symbols, such as, for example, predefined characters corresponding to the presses, or the system may print numerical values corresponding to the key presses of the user, or the like. It is possible). According to one method, for example, if a user wishes to correct a non-desired word (eg or chain symbols) or to select a desired word, then he indicates the wrong word (eg as described above) and then (eg The wrong word provided based on the corresponding key presses initially entered), the system can recognize the user's key presses provided for entry of the word by considering the chain symbols (e.g., obviously, the system can easily print the printed characters). To the corresponding key). For example, a user may indicate a word and utter at least some of the letters corresponding to the word (eg, if the word does not exist in the database, or for convenience, etc.). (E.g., instead, he may speak words). At the end of the utterance, the user may provide an end-of-the-speech signal, such as moving the pointer away from the word (e.g., the user may use other methods to inform the system of the beginning and end of the utterance). May be). For example, as described above, the user can press-hold a predefined key to inform the system of the start of utterance corresponding to the modification / selection, and release the key to notify the system of the end of utterance when the utterance is finished. Can be. Methods of correction / selection of words are already described in the references contained herein filed by the inventors.

In order to maximize the ease of use of the coupling systems by the user, a method of using punctuation symbols and instructions in the binding data input systems of the present invention may be considered. As described in the references contained herein, filed by the inventor, each of the most special symbols, such as punctuation symbols or numbers, interacts with its corresponding key and provides utterances corresponding to that special symbol. Can be entered. In doing so, any such punctuation or command can be entered during entry of text by the combined data entry systems of the present invention.

As mentioned in previous patent applications, in order to make the data entry systems of the present invention more user-friendly, some special symbols (eg, "." Provided at the end of a word) can be entered by pressing the corresponding keys without speech. Symbols) may exist. Some of the special symbols may be assigned to keys other than those to which letters are assigned. However, preferably the system uses a few keys, so some of the special symbols may be assigned to keys representing characters. Problems may arise when using the latter-mentioned symbols in a word (such as "?" At the end of a word when entering a word through word prediction systems of the combined word input systems of the present invention). For example, a user may use the word prediction systems of the combined word input systems of the present invention to provide key presses corresponding to the letters of a word without utterance, or at least (eg, the initial letter (eg, "r") of the word). The word "ready?" (E.g., by providing a key press corresponding to and voicing the letter, and by providing key presses corresponding to the remaining symbols of the word containing the special symbol (e.g. "eady?") Without the utterance (e.g., (The system may not know that the entered word ends with a special symbol, so the system may look for words with both five and six symbols). To solve the problem, the following describes a method of inputting special symbols (e.g., including instructions and functions).

According to one embodiment of the present invention, while using the combined word input system of the present invention, other symbols such as words, letters, and special symbols may be input mixed in the same chain symbols. The user can enter one or more subsequent words, chain letters, chain special symbols, etc., which press the keys corresponding to the words, chain letters, chain special symbols and press the words, chain letters. And by uttering at least the first character of the chain of special symbols. The first symbol of the chain symbols is assigned to the same key to which at least some of the characters are assigned, but different from the key interaction (e.g. single press operation) required to enter characters on the key. User is a "non-pronounced" symbol (e.g., entered by pressing the corresponding key without speech) that is input by an interaction with the user (e.g., a double press operation, pressing a key and another key simultaneously, etc.) May input a special symbol by providing a corresponding tour operation without speech at the beginning of the chain special symbols. For example, considering the keypad 12800, to enter the URL http://www.textentry.com, the user may do the following:

For example, using the symbol-one-one push-and-voice data input method of the present invention, the chain symbols " http " are entered (e.g., by pressing the keys corresponding to the characters and pressing the characters). ), After that;

A user may press a defined interaction or a key 12805 (e.g., as described in previous patent applications, for example simultaneously with other keys, such as a double-pressing operation on a key 12802 without a voice) key) and other predefined interactions such as pressing 12802) to enter a special symbol ":" (e.g., the symbol may be predefined to be assigned to a key interaction without speech). And then may enter special symbols “//” by providing two (eg, predefined) single press operations on key 12803 without speech. Entering the first special symbol at the beginning of the special chain symbols may tell the system that subsequent symbols entered are special symbols until the character is entered (eg, a key is pressed and the character is uttered);

The user can then enter the chain symbols "www" using the symbol-one-one push-and-voice data input method of the invention as described;

The user can then enter the symbol ".", For example, by interacting with the key 12806 (eg, a single press) without utterance, or by, for example, uttering a "dot" (e.g. All may be assigned to talk)

The user can then enter the chain symbols "textentry", for example using the symbol-one-one push-and-voice data input method of the present invention or using two parts (e.g. words in the system). This is done by typing in the two words "text" and "entry" that exist in the dictionary database. To do so, the user first presses a key 12802 corresponding to the letter "t" and utters that letter and subsequently presses the keys corresponding to the remaining letters "ext" of the word "text". The user can then press the key 12803 corresponding to the letter "e" and utter that letter and continue to press the keys corresponding to the remaining letters "ntry" of the word "entry";

The user can then enter the symbol "." By pressing the key 12805 without utterance or by, for example, uttering "dot" (eg, symbols can be assigned to both utterance and non-talk);

-(E.g. finally,) the user can enter the concatenation characters "com" using, for example, the symbol-one-one push-and-voice data input method of the present invention.

The chain symbols "http", "www", and "com" may be input by the word prediction system of the combined word input systems of the present invention (e.g., if the word is present in the word dictionary database of the present invention). Should be noted. For that purpose the user can do the following:

Enter the word "http", which provides for example a key press corresponding to the (eg at least) (eg first) letter (eg "h") of the word and uttering the letter and the remaining symbols of the word By providing key presses corresponding to the " ttp ";

Enter the word "www", which provides for example a key press corresponding to the (eg at least) (eg first) letter (eg "w") of the word and uttering the letter and remaining symbols of the word Provide key presses corresponding to the " ww ";

Enter the word "com", which provides for example a key press corresponding to the (eg at least) (eg first) letter (eg "c") of the word and uttering the letter and the remaining symbols of the word Key presses corresponding to the " om "

In general, each time the user switches between the types of symbols to be entered, the first symbol is uttered, or is entered via a key other than the keys representing the characters if it is a non-voiced symbol, or a non-pronouncing symbol is assigned at least a character. Is assigned to the same key, the symbol is a predefined interaction (e.g. double press, two keys pressed simultaneously) that is different from the interaction required to enter a character on the key assigned to that key (eg single press). , Etc.). It should also be noted that the word at the beginning of the mixed chain symbols can be input by the word prediction system of the combined data input systems of the present invention without providing a utterance corresponding to the first letter of the word. On the other hand, if the chain symbols begin with a special symbol, at least the first symbol can follow the methods.

According to another method, if the system is a sensitive surface for interaction (e.g. touch, pressure), one of the non-interaction interactions is a key, for example a single press action assigned to the characters on the key. It may be a sliding action for. These have already been described in detail.

Any chain symbols, such as the above-mentioned mixed chain symbols, may be symbols-one-one push-and-voice data input of the present invention as described in the references contained herein, filed by the inventors. It is understood that it can be entered using systems (eg, allowing the use of non-pronounced punctuation symbols and commands within the same key and the same type of key interaction to which the letters are assigned). According to one embodiment of the invention, switching means such as a mode key may be provided for switching between the push / slide- and voice data input systems of the invention and the combined data input systems of the invention. .

As already mentioned, space symbols, punctuation (eg, ",", ".", "!", "?", Etc.), functions, or commands (eg "enter") provided after providing information for entry of a word special symbols such as " enter ", " tabs, " In addition, some of the words provided at the end of a word (such as combinations of symbols that exclude or include special symbols, such as punctuation, such as, for example, "s", "n't", etc.) are the word-end signals of the word. Can be considered. For example, the word "doesn't" can be entered by first entering the word "does" and pressing the key corresponding to the "n't" part at the end and uttering a predefined utterance corresponding to that part. Can be. Obviously, such words that include punctuation symbols (eg, "", "-", "_") may be words in a dictionary database, and for example, symbols of the word (eg, with or without speech) (eg, Can be entered by pressing keys corresponding to a punctuation mark).

According to one embodiment of the combined (eg, including word prediction) data entry systems of the present invention, one of the letters assigned to a predefined interaction with the corresponding key is provided with and / or without corresponding speech. It can be input by the interaction of. For example, as shown in Figure 4, each letter "a" and "I" having a meaning in English may be assigned to different keys (e.g., different interactions having the same key) and that corresponding. It can be entered by pressing a key and activating the character and / or pressing the corresponding key without uttering.

The advantages of using a keypad with a few keys, such as four keys, have been described and demonstrated in the references contained herein filed by the inventor. The keypad may be, for example, a virtual / soft (eg fixed, dynamic) keypad used for the (eg touch) screen of an electronic device and used with the data entry systems of the present invention, such as the combined data entry systems of the present invention. have. For example, in addition to different types of tapping operations, for example, interactions that derive from, for example, sliding operations on a key of a keypad (eg, single slide, dual-slide, tap-and-slide, etc.) Other interactions, such as the ones of the present invention, on the one hand allow for the use of very few keys to better distinguish symbols from each other and on the other hand for complete easy and fast data entry such as text entry. It may be considered to assign at least a portion of the symbols of the data entry systems.
5 shows such a keypad 12900 as an example with four keys. For example, to enter a word (eg "driver") through the word-prediction data entry systems of the present invention and to enter a "space" symbol (eg immediately) after the word, the user is without speech. Providing key presses corresponding to letters of the word, or providing key presses corresponding to (eg, at least) (eg leading) letters (eg, "d") of the word and uttering the letters; In addition, key presses corresponding to the remaining symbols (eg, “river”) of the word may be provided. After providing the key presses, the user can glide on key 12902 to enter a space symbol (eg, immediately) after the word (eg, a "space" symbol glides on the key 12902 without speech). May be assigned to an action). This is by way of example only, with different symbol arrangements for different keys and different interactions with the keys, and also a different number of keys may be considered. For example, a letter on a key may correspond to sliding actions on the key (e.g., and if unintentionally entered by a pressing action on the key without utterance during word entry through the word-prediction system of the present invention). Or by voice), the “space” symbol may correspond to a press action on a key (eg, without utterance). These have already been described in detail in the present patent application and the references contained herein, filed by the inventors.

As mentioned above, the word dictionary database of the present invention may include words (eg and portions of words) of one or more languages. According to another embodiment, after entering at least one (first head) symbol of a word by pressing at least the first key and uttering at least one letter and providing key presses corresponding to the remaining letters of the word, the system If hesitates between two letters on at least a first key that has an unambiguously similar pronunciation with respect to each other, the system can, for example, suggest a word corresponding to key presses starting with the letters.

Another parameter to consider when assigning symbols, such as symbols, to keys on the keypad, is to harmonize (eg, finger (s) of the same hand) with the fingers of two hands, such as using two thumbs. Frequency of use of each key by the user's hand (to avoid a number of subsequent interactions). 6A-6B show four of a keypad that can be used, for example, with the combined data input systems of the present invention, taking into account the above-mentioned principles and the principles of the combined data input systems of the present invention as described above. Another example of an arrangement of characters for keys is shown.

As mentioned above, according to one embodiment the push / slide and voice data input systems of the present invention may be used together with other data input systems but separately. For example, the system may / are not based on the electronic device and only available key presses (e.g., without speech, such as a word-prediction system based on key presses without speech) / regarding other data entry systems, but examples For example, when providing information such as key presses corresponding to a portion of text such as a word, the information (eg key presses only) is different if the corresponding speech is not provided (e.g., the electronic device does not detect the speech). Based on the principles of the input system, it can operate independently of other systems so that it can be interpreted by other systems capable of entering portions of text (eg words).

On the other hand, as mentioned above, if associated with (eg, during) the input of key presses corresponding to a portion of text, such as a word, the user provides a speech corresponding to at least some of the symbols of the portion of the text or of the portion of the text. (E.g., if the electronic device detects the presence of speech) information corresponding to a portion of text / word (e.g., key presses and speech) is interpreted by the press / slide-and-voice data input systems of the present invention. It is contemplated that the push / slide-and-voice data entry systems are based on text (eg, based on the push / slide-and-voice data entry systems of the present invention (eg, including the word prediction systems of the present invention described above). Part of a word).

According to another embodiment of the present invention, even if a word is entered by another data entry system, the correction / selection of the entered word may be provided by the above-described push / slide-and-utterance systems of the present invention. have. For example, a user may indicate an incorrect word (e.g., select it using a mouse / arrow key, etc.) and (e.g. with / without the corresponding key presses (e.g. letters of the wrong / unwanted word) may cause the corresponding key presses. You can tell the system)) can utter the letters corresponding to the desired word.

One of the most confusing and frustrating issues of entering a word by the word prediction system based on key presses alone (eg without speech) is that if the word suggested by the system is not the word the user wants to enter, To choose. Based on the system, for example, if the desired word is one of the least frequently used words corresponding to the key presses provided by the user, several "" words are selected until the word is selected by the user. It should be noted that selection "key presses and concentration of many users' displays may be required. Key manipulations and long-lasting display focus are combined data entry systems of the present invention, such as a push / slide-and-utterance and word prediction data entry system based on consideration of at least some key presses provided for entry of words. S) can be reduced or eliminated using the data entry and modification systems of the present invention as described above.

As mentioned above, when using the combined data input systems of the present invention, in connection with the input of key presses corresponding to a portion of text, such as a word (for example, a way), the user may at least some of the portion of the text or the portion of the text. Speech can be provided corresponding to symbols. The system can detect whether vocalization is present or absent. Based on the principles of the combined data input systems of the present invention as described above, different systems of the combined data input systems of the present invention may be considered as described below, each of which may have its own advantages. have.

■ Example 1:

According to one embodiment of the present invention

In advance that only one word of the system's word dictionary database may correspond to chain key presses (e.g. corresponding to the letters of the word) provided by the user for inputting the word without providing the corresponding utterance. Defined and;

It is defined that other language words corresponding to the same chain key presses may not be included in the database. If the user wishes to enter one of the other words, he may, for example, pre-register as described above using the character-by-character press-and-voice data input systems of the present invention. Can be entered as a non-word. For example, if the user knows in advance that the word does not exist in the dictionary, he may provide key presses and utterances (eg substantially simultaneously) of the letters. If the user, for example, does not know in advance whether the word is in a dictionary and has noticed after input of key presses without a utterance, the user may, for example, speak the words of the word or utter the word as described above. have.

For example, taking into account the keypad 13200 of FIG. 8, only one word (eg, “doing”) in the system's word dictionary database corresponds to key presses 13202, 13204, 13202, 13202, 13204. It is appointed. If the user wishes to enter another word, such as "coins" with the same key press information, in addition to the key presses, the user may speak a letter of the word or speak the word (for example, as described above). ).

The system described above allows the words to be entered without ambiguity. Each sequence of key strokes for a key on a keypad without speech for word entry may correspond to only one word in the dictionary. The word may be a predefined word such as a commonly used word or a word determined by a user, or may be used in a predefined environment. The user may not be disappointed with selecting another word by manipulating the keys of the keypad. If the entered word is not what the user wants, he may simply speak the word or speak the letters of the word (eg without providing the corresponding key presses again). This is quick and easy.

■ Example 2:

According to one embodiment of the present invention

-It is predefined that only one word of the system's word dictionary database can correspond to the chain key presses provided by the user without providing the corresponding utterance to enter the word, and;

In most cases, only one word of the word database may correspond to information provided by the user, the information being the same chain key presses provided by the user and at least one of the words provided by the user (eg It is predefined to include the utterance of the first symbols);

It is defined that other words of the word database of the system with the same chain key presses may not be included in the database.

For example, taking into account the keypad 13200 of FIG. 8, only one word (eg, "doing") within the system's word dictionary database is generated by key presses 13202, 13204, 13202, 13202, 13204 without speech. It is defined to correspond. If the user wishes to enter another word such as "coins" or "dying" or "pupil" with the same key press information, in addition to the key presses, the user may speak or recall the (eg first) character of the word The word may be spoken (eg, as described above). As shown, each word in the database (eg, "coins" or "dying" or "pupil") that has the same key press information and needs utterances for the words entered, starts with a different letter, so that the user For example, when providing a word and uttering the first letter of the word, the user knows that the system will only provide one word corresponding to the information he has provided. Note that the word "doing" has the same key press information and starts with the word "d", but the word is assigned to key press operations without utterance.

The system described above allows the words to be entered without ambiguity. Each chain key stroke for a key of a non-voiding keypad for word entry may correspond to only one word in the dictionary. The word may be a predefined word such as a commonly used word or a word determined by a user, or may be used in a predefined environment. While each chain of identical key strokes on the keys of a keypad combined with the provision of the corresponding utterance for word entry may correspond to only one word in the word dictionary database. The word may be a predefined word such as a word which has the chain key presses and is most commonly used among words corresponding to the speech information. If the word entered is not what the user wants (for example without providing those key presses again)

The user may not be disappointed with selecting another word by manipulating the keys of the keypad. If the entered word is not what the user wants, he may simply speak the word or speak the letters of the word (eg without providing the corresponding key presses again). This is quick and easy.

■ Example 3:

According to another embodiment of the present invention:

In advance that only one word of the system's word dictionary database may correspond to chain key presses (e.g. corresponding to the letters of the word) provided by the user for inputting the word without providing the corresponding utterance. Defined and;

Other words in the word database corresponding to the same chain key presses (eg provided by the user) provide key presses and correspond to the utterance of at least one of the letters (eg at the beginning) of the word by the user. It is defined that it can be input by providing vocalization.

The system described above may allow input of predefined words without ambiguity. Each chain key stroke on the keys of a keypad without speech for word entry may correspond to only one word in the dictionary. The word may be a predefined word such as a commonly used word or a word determined by a user, or may be used in a predefined environment. On the other hand, each of the same chain keystrokes on the keys of the keypad combined with the provision of sufficient corresponding speech (e.g. at least necessary to distinguish a word from other words that correspond to the same key presses and require that speech to be input). as much as). If the word entered is not what the user wants, he may simply speak the word or at least many letters or all the letters of the word (eg without again providing the corresponding key presses). This is quick and easy.

For example, considering the keypad 13200 of FIG. 8, only one word (eg, "noise") in the system's word dictionary database corresponds to key presses 13202, 13204, 13202, 13202, 13203 without utterance. It is defined to be. If the user wishes to enter another word such as "cycles" or "clive" with the same key press information, in addition to the key presses, the user may speak at least one or more (e.g., first) characters of the word or May be spoken (eg, as described above). As shown, each of the words in the database (e.g., "cycles" or "clive") that have the same key press information and require utterances for the words entered, start with the same letter to distinguish the words from each other. The user may, for example, speak one or more letters of the word to distinguish it from other words of choice. It should be noted that utterances corresponding to the letters (eg or words) may be provided at predefined time intervals with respect to when at least one key press is provided. For example, the vocalization may be provided substantially simultaneously with providing the corresponding key presses or may exist mostly after a predefined time interval from before or after the key presses. During input of key presses, the user does not provide sufficient speech information (eg, only the letter "c" in this example) and if the system provides an unwanted word (eg "cycle"), the desired word (eg "clive"). In order to enter) the user can continue to the modification / selection procedure as described above.

According to one method, the user preferably enters at least the first symbol (eg a letter) of the word, for example by pressing the corresponding key and uttering the corresponding letter (eg or part / word), and said word without speech After providing the remaining key presses corresponding to the remaining symbols of the system, the system may attempt to find candidate words starting with the letter, wherein the remaining symbols (eg, letters) of the candidates are provided by the user without speech. Note that these correspond to key presses. It should also be noted that utterances corresponding to letters (eg or words) may be provided at least at a predefined time with respect to when at least one key press is provided. For example, the vocalization may be provided substantially simultaneously with the corresponding key press, or mostly, after predefined time intervals from before or after the key press.

As mentioned above, according to one embodiment of the present invention, a mode of interaction with a character key other than that of interacting with a key other than the character key, or used to enter a character on the key is used. Providing may be considered as a word-end signal by the system. For example, entering a word-part by providing information corresponding to the input of at least the first letters of the word and providing a sliding action or a double-press action for a key corresponding to the word-some first letter, such as a word suffix. Then, according to one method, the system may consider the word-part as the last part of the word and select at least one word corresponding to all the information based on all the information provided by the user. According to another method, the system may regard the word-part as an attachment to a word corresponding to the preceding information provided by the user and after selecting at least one word based on the preceding information, the system selects the word. A word-part can be attached to the at least one word. As mentioned above, according to one method, substantially all word-parts interact with at least one type of key other than the interaction required for the input of a symbol, such as a letter on a key (eg assigned to a key). Can be assigned to As noted in the references incorporated herein, filed by the Applicant, according to one method, at least substantially each word-part of each of the databases is assigned a key corresponding to the first letter of the word-part Can be.

As mentioned above, the modification procedures may be provided substantially immediately after the end of word entry or later. The user saves the text, opens it later to visualize it on the screen of the electronic device, instructs the word to change (for example, the instruction method can select the wrong / unwanted whole word), and the input of the word necessary for the entry of the word. At least one (eg, first) character may be spoken. By providing a predefined indication signal such as selecting an entire word, the system can understand that the user wants to change the entire word. Then, according to another method, in order to modify one (or one or more) symbols of the word, the user can point to the symbol (eg, indicate a position immediately after the symbol) and make a modification.

It is understood that other methods of selecting a word among candidate words, such as indicating / selecting one word, may be used separately or together with the utterance of the word (eg, at least some symbols of the word).

As mentioned above, the data entry systems of the present invention may use a keyboard, such as an on-screen (eg QWERTY) keyboard on the touch sensitive screen of the electronic device. As mentioned above, when a user types on the screen, he can tap (eg by mistake) an area (eg of a key) on the keyboard. The system may regard at least some of the neighboring keys as a pressed area (eg a key). This may each provide a set of keys having at least one symbol, such as a character to be considered with each user's pressing action during entry of a word. Each of the number of keys selected by each pressing operation on the keyboard may be associated with at least one symbol, such as a character, to simulate a key of a keypad having a plurality of symbols, such as a character. Based on the principle, the system can then use the method and keypad by the different data entry systems of the present invention, such as the push / slide-and-voice data entry systems of the present invention or the combined data entry systems of the present invention. have. For example, by using the on-screen keyboard 13100, as shown in FIG. 7, in order to enter the word "have", the user can press areas near 13101-13104 near the corresponding keys of the word. have. For each of the impact zones, the system may consider letters on the keys surrounding the zone, for example. For example, in this example, the system may consider at least some of the letters “h, j, y, u” for the impact zone 13101, and the system may consider at least characters for the impact zone 13102. at least some of the letters “e, r, s, d, v” may be considered for the impact zone 13103. It is understood that symbols corresponding to push operations as described above can be used with the data entry systems of the present invention. For example, for better accuracy (clarity), while pressing the keys, at least a utterance corresponding to at least one of the key presses may be provided to better recognize the at least one character. For example, a user may speak the letter "h" while providing the key presses. The system may realize a word starting with that letter and, for example, by considering the letters corresponding to the rest of the pressing actions provided by the user, the system may provide the word "have". These have already been described in detail. Instead of a QWERTY keyboard, any other keyboard can be used with any number of keys, any symbol arrangement, and the like.

Although the keys are shown and described as input means of the electronic device in many embodiments, the keys provided with the electronic device, such as a user's fingers, finger caps, stylus, or electronic device (eg, tilt) operations, etc. And / or any other input means that simulates interaction with the keys may be used for the same purpose. These have already been described in detail above.

Also in accordance with another embodiment of the present invention, after selecting a word based on the user's input information, the system may speak the word (e.g., a text-to-speech system) to allow the user to display the screen. You can decide if you want to change the word without looking at it, or you can approve it. For example, after listening to the system, he may continue to enter new symbols / characters, etc. if the user wishes to approve it. Also, for example, after listening to the system, if the user wants to change the word, he may, for example, speak at least one of the letters of the word, speak the word, or the like. Changes or recognition of words have already been described in detail above.

As mentioned above, speech information provided for the data entry system of the present invention may be based on at least one of user's lip movements or voice. When using the lip reading recognition system of the present invention (e.g., the letters of a language are assigned to different keys), sputtering, uttering, such as letters, with a lap of idle time between each uttered sign is said symbol. Can significantly improve their recognition accuracy. As mentioned in the references contained herein, filed by the inventor, symbols having ambiguities that resemble indefinite with respect to one another may be assigned to different keys.

According to one embodiment of the present invention, the lip read recognition system can be used with the combined data input systems of the present invention (eg including word prediction systems as described). For example, as mentioned above, in order to enter a word, a user presses at least one of those symbols, such as its initials (e.g. letters), accurately and clearly (presses) the corresponding key and speaks the letter. You can input by doing. The user can then provide a key press corresponding to the remaining letters of the word. Speaking a letter corresponding to a word and providing the remaining key presses without utterance may provide a time lap to pause utterance after uttering the word. If the user provides consecutive words, there may be a pause (eg most) between the utterances of the (eg first) letters of the consecutive words provided. To recognize the spoken (eg first) character, a lip reading recognition system can be used. This may add more discernment to the data entry systems of the present invention.

If the system does not recognize / select the desired word based on the information provided by the user through this embodiment, the user may continue to speak the above-mentioned words, for example, by letter-by-word or by the word. Modification / selection procedures as described above may be carried out. Speech provided for the correction / selection procedure may preferably be recognized based on the user's voice (eg, obviously, speech recognition may be based on lip reading, especially when the user speaks one or more characters separately). ). This is because current speech recognition systems are more sophisticated than lip reading systems in recognizing complex utterances (eg, utterances of words, faster utterances, utterances of letters without pausing). If there is an elaborate lip reading system, it is understood that it can be used to recognize speech based on user's lip movements.

If the user does not provide a pause when providing a utterance related to the first letter of the word (eg, uttering one or more letters), the lip reading system may substantially reduce the first part of the user's utterance (eg, the first letter of the word). Only corresponding)

As mentioned above, words in the dictionary database may be entered by providing corresponding key presses without utterance. If the system uses a lip reading system to accurately (clearly) recognize characters such as the first letter of a word, the system cannot recognize whether or not vocalization is provided to the begging of a word by the user (eg, the user's lips). May not be able to determine whether they have moved). In this case, the voice recognition system of the present invention may be used to know whether there was vocalization. Thus, it may be helpful if the user also provides utterance if he wants to speak a letter of the word by providing only lip movements (e.g. even very separately, the microphone detects what is happening by the user so that the system can Remind them to use the lip reading system to recognize speech).

For better accuracy (clarity) of speech recognition, it is understood that both speech and lip reading perceptions can be used (eg combined).

The word prediction of the present invention when symbols, such as letters with ambiguity that resemble unclearly in relation to each other, are assigned to the same key (eg and to the same type of interaction with the key) of the keypad (such as a telephone-type keypad). The data entry system may consider words that have the same key presses and start with any obscure symbols. For example, each utterance (eg, voice or lip movements) of a pair of letters “d, e”, or a pair of letters “j, k”, where each pair is on the same key of the telephone-type keypad. May be unclear even by using a lip reading system or even a speech recognition system. For example, by using a telephone keypad used with the combined data entry systems of the present invention, in order to enter the word "keep", the user can press the key corresponding to the letter 'k' and speak the word. have. The user can then continue to press the remaining keys without speaking. The system may also consider (eg, predefined) words that begin with the letter "J" and have the same remaining key presses (eg, "jeep").

Different means, such as a microphone and / or a camera (e.g., implemented in the electronic device or used as a peripheral device / attached device connected to the electronic device) may be considered for the detection of the user's speech. The detected and analyzed lip movements images of the user may be 3D (eg three dimensional). As mentioned above, other parts of the user's face (eg teeth, jaw, etc.) may also be considered when speaking. These have already been described in detail.

With continued reference to the touch sensitive keyboard of FIG. 7 and its corresponding embodiments, preferably, the user is able to accurately (clear) at least one (eg first) key press corresponding to the (eg first) letter (s) of the word. Can be provided (e.g. accurately / clearly) (e.g. without speech), and as described above in the present embodiments, the remaining symbols of the word are unambiguous (e.g., the unclear keys of the reduced keypad of the present invention). Key presses). This may provide enough information for the selection of words (eg without the need for utterance). It is understood that on the basis of the principles of the data entry systems of the present invention it may be considered to provide corresponding speech information as well. As mentioned above, the correct (clear) symbol (s) may be input by the handwriting input and recognition system.

As mentioned above, by using the word prediction data entry system of the present invention, it is possible to accurately / clarify at least one letter of the word, for example by providing a predefined pressing / sliding action on a key corresponding to the word. To provide the voice information corresponding to the character (e.g., to speak the character), and to perform a press / slide on keys corresponding to at least one (e.g., preferably all) of the remaining characters of the word. The word can be input by doing so. Clearly, in any embodiment of the (e.g., predictive) data entry systems of the present invention, the press / slide action for that key and the corresponding voice for inputting at least one correct / clear character of the word, the handwriting input and recognition system It may be replaced by another accurate / clear data (e.g. character) input system, such as a complete (e.g. pc) keyboard, etc. In addition, in the modification / selection procedures of the present invention, instead of using the user's speech, the user may employ another accurate / clear data (e.g. character) input system such as a handwriting input and recognition system, a complete (e.g. pc) keyboard, or the like. Can be used to provide one or more necessary characters to select / modify / enter a word.

With continued reference to the combined data entry systems of the present invention, the (eg complete) word dictionary of at least one language database and the word-some dictionary database of at least one language used with the system may be separated from each other. The system may be predefined to be informed that the provision of key presses may correspond to the input of a word of the language, for example the provision of a slide action may correspond to a word-part input of the language. The system may search for the word in the dictionary database.

It is understood that the combined data entry system of the present invention may include any database separately or include both. These have already been described in detail.

Although the "combined" data entry systems of the present invention have been mentioned in various forms in the present application and the references contained herein, the system includes, but is not limited to, the (eg first) symbol (s) of the word ( Combined with word prediction data entry systems, for example based on the provision of at least some (eg preferably all) of key presses corresponding to the letters of a word with or without uttering at least one of the letter (s) It should again be noted that reference is made to the data entry systems of the present invention, which combine the push / slide-and-voice data entry systems of the present invention.

Searching for words in a word database based only on key presses, for example provided on a telephone keypad, has been known for many years. As mentioned, the chain key press provided by the user may correspond to one or more words in the database, where the word with the highest frequency of use is usually suggested to the user as the first choice. Other methods have been proposed to accept the word or to select another word if the word is not what the user wishes to enter. According to the most famous word prediction system (e.g. T9) used in Nokia phones, entering the word "pda" requires seven additional key presses that include a word-termination signal such as a space symbol. Much attention should be paid to the display to identify each next word that this suggests. Even if the user knows in advance that the word "pda" is one of the lowest priority words among the candidate words, he must follow the same steps to select it. The worst case is when the user navigates between words and then finds that none of the words suggested by the system are desired. In this regard, the user must switch to another method, such as a multi-tap system, to enter a word and generally provide a plurality of tabs for each key corresponding to each symbol of the word. A large portion of the text entered in real-time situations includes non-dictionary words (eg, names, addresses, URLs, numbers, punctuation marks, words in other languages, abbreviations, expressions, etc.). For that reason, word prediction systems are not used for general data entry on computers such as tablet PCs or PDAs that really lack a small, quick and easy data entry system. Even in mobile phones, although most people are slow but still using a multi-tap system, the user knows that "what is obtained is typing," entering text symbol-by-one.

As mentioned above, the combined data input systems of the present invention provide information corresponding to a word in a dictionary (eg key presses corresponding to letters of a word and utterance information corresponding to a word, wherein the utterance information is at least one of the letters of the word). May be other types of information, such as uttering at least a portion of a word, such as uttering some, or uttering a word, or not providing utterances, wherein the utterance information is detected and analyzed by the system. These may minimize and even eliminate additional user interaction after entering the application filed by the Applicant, which has already been described in detail herein (eg, the system may provide keys for input of the desired word). Can eliminate additional interaction with the user, which means that experienced users can That the required, and eventually, it is possible to know the quantity or type of the utterance that require advance minimize the interaction with the display unit and even eliminate).

Another great advantage of the system is the ability to input out-of-dictionary chains of symbols such as words. As mentioned above, for example, if the system does not find chain symbols (eg words) in the dictionary database after input of information provided by the user, the user does not press additional keys and symbols corresponding to the chain symbols. You can enter chain symbols by uttering (eg, letters). As mentioned above, the system may consider original key presses.

Another great advantage of the system is that it is possible to generate special chain symbols (e.g., proper names, numbers, punctuation marks, etc.), preferably by pressing at most one key per symbol or pressing two keys at the same time. Chain symbols, etc.) and the ability to enter quickly and easily. For example, if the user knows in advance that the chain symbols that he wants to enter (such as a word or URL in another language) are any chain symbols that do not exist in the database, he is the push / slide-and-voice data of the present invention. An input system can be used (e.g., for symbol-to-one-by-one input of chain symbols) (e.g. a symbol that presses keys corresponding to the symbols and provides utterance information corresponding to the symbols, for example). By using one-by-one data entry system).

Using a few keys, such as at least four keys, with word prediction data entry systems based only on key presses (e.g., where the four keys of the keys are substantially assigned all characters of the language distributively). ), Cumbersome and even unrealistic because of the large number of corresponding words for chain key presses. The combined data entry systems of the present invention enable the use of a few keys and a quick and easy data entry system. Using a few keys with the data entry systems of the present invention (e.g., four of them correspond to characters in at least one language, e.g. the keys each for example of a tablet PC) The advantages of being divided into two groups located on one side) have already been described in detail.

In addition, data entry systems based only on key presses (eg, not taking advantage of providing at least the minimum amount of corresponding speech information) may be subject to large numbers of chain key presses if the dictionary database includes words in multiple languages. It would be unrealistic because of the corresponding words for. The combined data entry systems of the present invention make it possible to use multiple languages simultaneously and quickly and easily with a data entry system.

Also, as detailed above, all special symbols such as space symbols, numbers, punctuation symbols, and commands (eg, the "Enter" function) generally provide at most one press operation on a key. Or by pressing two keys simultaneously (eg within a chain of symbols such as words or independently). In addition to its primary purpose, a clear special symbol after the entry of information corresponding to a word is finished may inform the system of the word-end. This may be another great advantage of the combined data input systems of the present invention. After inputting information corresponding to a word, a user may input a desired symbol such as at least one special symbol. Special symbols may be considered as word-end signals by the system. In addition, the system may insert or execute a symbol after a word (eg, some of the special symbols may be used (eg may be used) within a word and may be part of an opaque symbol assigned to a key).

According to one embodiment of the invention, when a user enters a word by using the word prediction systems of the combined data entry systems of the present invention, the system is a system such as a text-to-speech system to provide the user with the selected word. Can be used. In most cases, the word is a desired word, and the user may inform the system of the acceptance of the selected word, for example by continuing to advance to the next symbol or input of the next word. If the selected word is not the word the user wishes to enter, the user may proceed to the selection / modification procedures as described continually (eg by speaking at least some of the word or symbols of the word). This embodiment can provide a complete eye free rapid data entry and correction system without additional key press operations.

As mentioned above, according to one embodiment of the present invention, the word prediction systems of the combined data entry systems of the present invention may utilize a word dictionary database, where the words are a chain key press provided by the user without speech. Only one word in the field is arranged. Several ways of organizing a database and searching within it can be considered, for example, as follows:

According to one method of database arrangement, the system may include two different word databases: each of the words may correspond to different chain key presses provided by the user without speech (eg each word is Corresponding to different key press chains) and the first word database, and each of the words may correspond to chain key presses provided by the user and the user may also provide the corresponding utterance as described above (eg, at least one The words may correspond to different key press chains) Second word database. If the user provides chain key presses and does not provide the corresponding speech, the system may detect that the user is not speaking. The system can then search for the word in the first database. At most one corresponding word may exist in the first database. The words can be entered and printed on the corresponding display displaying the text the user enters. If the word is what the user wants to enter, the user can go on to the next task, such as to continue typing the next word. If the word is not the word the user wishes to enter, the user can proceed to the modification / selection procedure as described above. For faster data entry, the user may not advance to the correction / selection procedure immediately after entering a word. He may move on to the next task, such as entering the next part of the text (eg the next words), and so forth, and later on with the modification / selection as described above.

If the user provides chain key presses and provides the corresponding utterance, the system can detect the user's utterance and retrieve the words in the second database. There may be one or more words corresponding to key presses and utterances of the user in the second database. For example, a user presses at least one key corresponding to at least one leading letter (such as a letter) of the word and utters the at least one letter, and then continues with at least some of the remaining symbols of the word ( For example, by pressing keys corresponding to all), according to one search method, the system can recognize the at least one character, and within the second word dictionary database, it starts with the at least one recognized character The remaining symbols of the words may search for words corresponding to the remaining key presses provided by the user. According to another search method, the system further comprises in the second word dictionary database that the symbols of the words correspond to a chain of key presses provided by the user and the initial letter (s) correspond to the at least one recognized character. You can search for words. Such search methods are understood to be described as examples. Other search methods can be considered by one skilled in the art.

In the continuing description of the current embodiment, there may be one or more corresponding candidate words in the second database. If only one word exists, the system can enter it. If there is more than one candidate word, according to one method, a predefined word among the candidates may be input. According to another method, one of the higher priority candidate words with respect to others may be automatically entered (each of the words in the second word database may be assigned a priority value based on its frequency of use, for example). Can have). The word may be printed on a corresponding display displaying text input by the user. If the word is what the user wants to enter, he may move on to the next task, such as entering the next word. If the selected word is not the word the user wishes to enter, according to another method, the user can provide more speech information, such as uttering at least some of the remaining letters of the word, the user being described above. You can move on to the process of modification / selection. If the user desires, according to another method, the user uses a switching means such as a select key to select a candidate word (e.g., presented one by one based on the frequency of use) until the desired word is found. You can try to navigate.

As mentioned above, for faster data entry (eg during a meeting), the user may not immediately proceed to the correction / selection procedure after receiving an unwanted word. He may continue to the next task, such as entering the next part of the text (eg the next words), etc., and he may later proceed to the modification / selection procedures as described above.

According to another method, the system may comprise a single word database comprising two above-mentioned databases.

According to another method of database arrangement, the system may comprise a single word database in which each word has a priority value, for example based on its frequency of use. When the user provides key presses corresponding to the desired word without utterances, the system selects the word with the highest frequency of use with respect to other words that correspond to the key presses (e.g., if present) that correspond to the same key presses. You can choose. In this method, in addition to key presses, if the user provides utterances such as uttering one or more initial characters of a word, the system is based on the association information and is most likely associated with other words corresponding to the association information. You can select the word with higher priority. If the word is not the desired word, the user can move on to the selection / modification methods as described above.

According to another method, the system further comprises additional information (e.g., a flag) where each word in the database indicates its spoken situation (e.g., needs or is spoken (e.g. "1") or not (e.g. "0")). ), A single word database that may include, or may be associated with. When providing a chain of key presses, if the system detects utterances, it can only consider words that indicate that the utterance indicator "needs no utterance," or vice versa. According to one method, for each chain of key presses that do not provide speech, there may be at most one word in the database. The procedures of input and selection / modification may be similar to those described above.

As mentioned above, chain key presses provided by a user may enter punctuation symbols, numeric symbols, commands (eg, "Tab", "Enter"), a space symbol, a symbol such as a letter at the end of a word, and And may include a word-end signal, such as one or more special symbols, such as uttering a symbol. Since the symbols are generally clearly input, according to one method, the system may not include key presses corresponding to special symbols among chain key presses for retrieval purposes. After entering the desired word, the system can enter special symbols. The user can move on to the next task, such as entering the next word without additional key presses.

As mentioned above, another type of word-end signal may be one that provides key presses corresponding to the next word and utters the utterance of the next word (eg, at the beginning / symbols). According to one method, the system can automatically provide a space symbol between words.

Different modification / selection methods of the combined data input systems of the present invention have already been described in detail.

As mentioned above, in order to change a word that has already been entered, the user can indicate (eg, first) the word. Directing a word to replace a word with another word may be defined so as not to conflict with the directive action provided for other purposes, such as inserting a character behind or within a word. For example, a user may select a word, for example by double clicking on the word, or he may place the cursor after the last symbol of a word that includes or excludes that special symbol, such as a space symbol or punctuation, etc. have. According to another method, a user may position and press-hold a predefined key at any position in the word (first, inside, last) and to select / enter another chain symbol (eg word). Can provide the desired vocalization.

In order to provide the minimum amount of utterance for clear word entry, the user may preferably speak the first letter of the word. If there is one or more words that match the information (eg key presses and utterances), the user may speak the next letter until the system provides the desired word. According to one method, each time a user speaks a character, the system needs to make a clear input of any remaining words corresponding to the information (eg, presses and utterances) already provided by the user (e.g., of the next character). The maximum number of points can be pointed out.

According to another method, in order to select a word in a document, the user may, for example, speak a word or at least speak at least one of the initial letters of the word, or at least one of the initial letters / portions of the word. Perform a press / slide and utterance on one, or provide one or more pressing actions corresponding to one or more (eg, initial) characters of the word without utterance. The system may compare the information provided by the user to find the desired word in the document with information or words corresponding to the words in the document. This is very fast and accurate without the need for navigational means such as a mouse. For example, referring to FIG. 9, and by considering the keypad 13300 to replace the word "doing" with the word "coins", the user may speak a word or select some of the initial characters of the word (eg, "do"). ") Can utter. The system may compare the utterance with the initial letters of the words in the document to find the word "doing". The system can find one or more words. The user can navigate between the words to reach the word he is looking for. See also FIG. 9, for example, and by considering the keypad 13300 to replace the word “doing” with the word “coins”, the user may, for example, begin with the initial characters (eg “do”) of the word. ), Two keys 13304 and 13303 may be pressed. The system may compare the key press actions with key presses corresponding to the initial letters of the words in the document to find the word “doing”. The system can find one or more words. The user can navigate between the words to reach the word he is looking for.

According to one embodiment of the invention, during the input of a chain key press corresponding to a word (e.g. and eventually providing a utterance of at least a portion of the word) (e.g. before providing a word-end signal), the system is each time The user may suggest candidate words that provide a new key press. According to one method, when the user enters the last key press, he can move on to the acceptance or modification / selection methods of the present invention as described above. According to another method, each time a chain of new key presses is entered, the system may suggest words with more symbols than key presses entered up to that moment. The user can select a word without providing the remaining key presses corresponding to that word (eg, a word completion system).

While pressing the keys corresponding to the word (eg, and, if provided, spoken information), the word completion system may consider (eg and suggest) words that begin with the key presses (eg, and speak). The word with the highest priority may be printed in the text that the user enters (eg and the cursor may be placed at the end of the suggested word). If the user continues to type new key presses (e.g. and utterances) and the key presses (e.g. and utterances) still correspond to the next letter of the word, the word (e.g. and the cursor) may remain in the same position. . If the next key press (e.g. and speech) does not correspond to the word and corresponds to another word (or other words) with the same start (e.g. key press) information, the word with the highest priority among them is entered by the user. Can be entered in the text (eg and the cursor can be placed at the end of the proposed word). At any moment of key press (e.g., vocalization) actions, if the suggested / printed word is what the user wants, he can provide a word-end signal such as a special symbol as described above and move on to the next task. Can be.

Although keys have been used to describe different embodiments of the present invention, any other input means (eg user's finger interactions, electronic device), as mentioned several times earlier in other patent applications by the inventor It should be noted that the motion movements of the < RTI ID = 0.0 >, etc. < / RTI >

As described, the word prediction systems of the present invention allow for a quick and easy word input system. According to one embodiment, no words may have a priority assigned to them. For each chain press operation that does not provide speech, there may be at most one corresponding word. Any other words having the same chain key press can be distinguished from each other by the user's speech or the speech of the word corresponding to at least one of the (eg, beginning) symbols of the word. Even if each of the words has a degree of priority (eg a value) in relation to each other, by using this system any word can be accessed (eg directly) and entered quickly.

As mentioned, other candidate words that have the same (above) chain key presses are determined by the user's utterance corresponding to the word (as described above) (eg, additionally by providing the user's utterance for key presses), Or by providing different traditional means or user utterances and other traditional means. To provide information to the user, other candidate words may be presented to the user in a list, or one by one, on the display unit. As mentioned, according to another embodiment other candidate words may not be presented to the user. If the user wishes to enter one of the other words, he simply provides the corresponding utterance without further interaction with the display unit (as described, for example).

As mentioned above, the data entry systems of the present invention can be used in many applications, such as call directories. For example, as described in other PCT patent applications, such as PCT / US2002 / 22385, filed by the inventor, the (eg text) information provided through one or more suitable embodiments of the data entry systems of the present invention may be applied to the corresponding application. May be used (eg having a database of words or phrases) to provide tasks (eg predefined). For example, the system can be used to:

Creating telephone calls;

-Provide input to the calling directory;

Providing input to interactive call centers;

Provide information about the system providing predefined tasks based on (eg textual) input of the user;

- Etc.

According to one embodiment of the invention, for example, a method of inputting data such as text within such an application may be considered. The method may, for example, comprise at least one of the following steps:

1) the user can provide key presses corresponding to textual information such as words without speech;

2) If the system recognizes the text (eg a word), go to step 4.

If not;

3) The system may require more information from the user, such as:

   o uttering text (eg words); or

   o Speak at least some of the letters that correspond to key presses provided by the user. The minimum number of characters to be spoken to distinguish the text (eg a word) from other texts (eg words) having the same chain key presses may be dynamically defined by the system.

If one or more texts (eg words) in the database correspond to the information provided by the user, the system asks the user to provide more information. For example, if the system is used to provide a telephone call based on the name of an entity such as a person or company to be telephone-called, the system may be telephone-called to the user, for example, postal code, office / home information, etc. More information may be requested, such as entering at least part of the address of the presence.

4) After successfully recognizing the user's input, the system can therefore move on to the next predefined task.

For example, as mentioned above, the system can be used with a calling directory to provide a telephone call based on, for example, at least a telephone-name to be called, or to provide a caller with a telephone number to be called. have.

According to one embodiment of the invention, the user presses each of the keys corresponding to the letters of the name, for example, and simultaneously utters the corresponding letter with each pressing operation, thereby the symbol-one-one data entry system of the present invention. You can use to enter the name to be called.

According to another embodiment of the present invention, the user can call-call using the symbol-one-one data entry systems of the present invention, for example, by pressing keys corresponding to the letters of the name to be called. Can be entered and (if necessary or required by the system), then can speak at least a letter of the name or can speak the name. Providing key information and speech information in different time wraps may be useful in situations such as when using a touch toned telephone. When using a touch tone telephone, pressing keys and speaking at the same time can combine the key tone with the corresponding speech, thus confusing the recognition system.

For example, if a user wants to call a company called "TheCompany" using, for example, a telephone keypad, according to one method, the user may first provide key presses corresponding to the characters of the name. have. If the keypress chain only corresponds to one existence (eg name) in the database, the system can request confirmation before proceeding to the next predefined task or to the next predefined task.

On the other hand, if the user wishes to call a company called "TheMarkers", for example using a telephone keypad, the user may first provide key presses corresponding to the letters of the name. The key presses may correspond to one or more beings in the database. For example, another entity in the database could be the name of another company called "TheMaskers." To distinguish the names from each other, the system may ask the user to provide utterance information, such as uttering at least six initial characters (eg, "themas") of the desired name, to determine which name the user wishes to enter. have. According to another method, the system may ask the user to speak a name, such as speech information.

It should be noted that the processing of the information may be performed locally (eg in a telephone) or the system may be installed in a server (eg in calling directory agreements) and processed remotely. In this case key press and speech information is sent to the server for processing and results or additional requests can be sent by the server to the user.

According to one embodiment of the present invention, telephone calls can be delivered internationally. According to one method for this, the system may process the information locally, including databases of entities such as people or companies in at least some countries to be telephone-called, to direct international calls to the destination, According to another method, the system may use some of the information provided by the user to direct at least some of the information provided by the user to a server in the telephone directory of the corresponding country for processing there.

For example, the user may enter the text "Benjamin Ghassabian NY USA". The system can direct information such as "Benjamin Ghassabian NY" to the appropriate telephone directory in the United States for processing there. According to another method, the system can direct information such as "Benjamin Ghassabian" to the appropriate telephone directory in the state of NY, USA, where it can be processed. The order of the words in the text may be predefined or “intelligent” to allow the system to freely provide the order of the words in the text.

According to one embodiment of the invention, the purpose of key presses (e.g., and speech) provided by the user with a predefined signal, such as pressing the "#" key before or after entering a telephone-calling entity, e.g. Type). For example, pressing a predefined key, such as the "#" key, before a chain key press, for example, the key presses provided after pressing the "#" key would correspond to the entry of the name of the party to be called. And after the name is entered, the system can inform the system that the other party can be called automatically. Thus, for example, pressing a predefined key such as the "*" key before (e.g. or after) a sequence of key presses, for example, the key presses provided after pressing the "*" key are Corresponds to the entry of the name of the party to be called, for example, after the entry of the swap name, the system can inform the caller that the other party's telephone number (eg or other information such as its address, etc.) can be provided to the caller. have. Or vice versa. Moreover, according to one embodiment of the present invention, it is not possible to provide the system with a predefined signal before or after chain key presses, for example, that the key presses provided by the user correspond to the dialing of the telephone number. I can tell you.

Note that one or more words (eg, first and last name) of an entry in the database may be included. According to one method, the user presses the corresponding keys of a word to achieve the entry of each word and, if necessary, the word-entry by providing the corresponding utterance for entering the word before proceeding to the entry of the next word. While one-by-one can be entered, in another way, the user may first provide key presses corresponding to all the words in the entry and then provide the spoken information needed to enter each word.

The data entry systems of the present invention can enhance or even replace interactive call centers based on voice only. For example, when a user calls a company, instead of listening to all the choices that the company's automatic voice system suggests to the user (which is time consuming and confusing in most cases), any of the data input systems of the present invention can be used. By having the ability to enter the text of the user, the user can skip at least some steps suggested by the calling systems and directly enter the information he wishes to enter. For example, after calling the company's automated call center, the user may request specific support by entering his wishes using the data entry systems of the present invention. Also, for example, if no one is available, the user may enter his contact information, such as his name, and / or his address, and / or his phone number, etc. (eg, later) to be contacted by the company. have.

It should be noted that the embodiments, methods, and examples provided in this patent application are used only to illustrate the concepts of use of the data entry systems of the present invention in various applications. It is understood that other embodiments, methods, and examples based on the principles of data input systems of the present invention, their use in different applications, and the like, may be considered by those skilled in the art.

It should be noted that while the embodiment described above requires less vocalization, it requires more attention / interaction of the user on that display. As mentioned in the other embodiments, if the first proposed word is not the word the user wishes to enter, the user indicates the word (eg by selecting it in the word processing window) and the symbols of the word (eg It is understood that the focus on the display can be minimized by providing utterance corresponding to at least some of the symbols of the word or uttering the word.

The functions of applications such as call directories, call centers, answering centers, and the like are known to those skilled in the art. Combining the above applications with the data entry systems of the present invention provides faster interaction with the present systems to achieve the goal and allows for the easy and quick entry of complex text.

Different procedures for the selection / modification of unwanted words proposed by the system have been described above. Other ways of minimizing utterance needed to replace unwanted words with other words are described below.

According to one embodiment of the invention, when using the data input (e.g., including word prediction) systems of the present invention to enter a word, if the word selected by the system is not what the user wishes to enter, the user may (e.g. After pointing / selecting a word that you do not want, you can begin to speak the symbols corresponding to the word one by one. For example, after uttering the first symbol of the word, if the selection of two or more other words in the database (eg other than the unwanted word) is provided by the user and the first key presses begin with the spoken symbol If so, the system may provide the user with a word in the selection that has the highest priority with respect to other words in the selection. If the proposed word is still not what the user wants to enter, the user can utter a subsequent symbol of the desired word. The system may suggest another word (eg, within the selection) that begins with the two symbols spoken. If two or more words begin with the two symbols, the system can suggest the one with the highest priority. According to one method, if a previously proposed word also begins with the same two symbols, the system can determine the word even if it has the highest priority among the words corresponding to the two spoken symbols. Can be ignored. The fact that the user advances to the next symbol of the desired word can tell the system that the word suggested by the system is not the word the user wishes to enter. If the last word presented is still not what the user wants to enter, the user can speak the next letter of the desired word, and so on.

For example, FIG. 10 shows a keypad 13400 having at least four keys, wherein the four keys 13301-13404 of the keypad represent letters of the alphabet's alphabet. Values 1 to 4 are thus assigned to corresponding keys 13401 to 13404 respectively. Also shown is a portion 13405 of the system's word database corresponding to a chain of key presses 13301, 13401, 13402, 13402 (eg, the value “1122”). In this example, if the user presses keys 13401, 13401, 13402, 13402 (eg, key press value "1122") without utterance, the system pre-selects to select the highest priority word corresponding to the key presses. Can be defined. In this example, the word is the word "what". Immediately after providing the key presses, or later during the modification / selection procedure indicative of the proposed word (eg "what"), the user may speak the letter "n" (eg without providing key presses). The system is responsible for the first key presses 13301, 13401, 13402, 13402 having the highest priority with respect to other words that start with the letter "n" and correspond to the same key presses and begin with the same letter "n". You can suggest a word. In this example, the word is the word "next". If the proposed word is still not what the user wants to enter, the user can speak the next letter of the desired word, "e" (eg without providing key presses). According to one method, the fact that an additional symbol is spoken causes the system to ignore the currently proposed word (eg "next"), even if the additionally spoken symbol corresponds to a subsequent symbol of the word (eg "next"). To tell. The system has the same initial key press information, by excluding the proposed word (s) (eg "next"), the two (eg first) characters match the utterances of the corresponding user, and the same key press It is possible to propose a word with the highest priority in relation to other words which have information and whose two (eg first) letters match the utterances of the corresponding user. In this example, the word is the word "near". If the word "near" is still not the word the user wishes to enter, the user proceeds to the utterance of the next symbol of the desired word, the systems can behave based on the principles described, and so on. In this example, by uttering the subsequent letter "a", the system suggests the word "neat", for example, which the user wishes to enter.

As mentioned above, the data entry systems of the present invention, such as the predictive data entry systems of the present invention, may be combined with other data entry systems such as handwriting input and / or recognition systems, multi-tap systems, and the like.

For example, the word prediction method of the present invention is combined with a handwriting recognition system to allow the handwriting input and recognition system to be used as an additional recognition system or to replace the speech recognition systems of the present invention. For example, a user may enter a word by providing key presses corresponding to the word without utterance. If the word is not the only word corresponding to the chain key presses, the user may use the handwriting recognition system (eg, by using a handwriting recognition system for predefined time wraps, such as during the entry of the word or during the modification / selection procedure as described above). At least some of the symbols can be entered correctly. For example, as shown in FIG. 11A, a system using a separate keypad having at least four keys (13501 through 13504) (e.g., fitting one of the keypads described above) of the tablet PC 13500, By considering the portion 13508 of the word database of < RTI ID = 0.0 >, < / RTI > in order to enter the word " has " The system may suggest an unwanted word "was" that corresponds to the same chain key presses and has the highest priority. Then, as shown in FIG. 11B, the user wants the desired word “has”, for example, by writing at least several (eg, first) characters necessary to distinguish it from other words having the same key press chain. Words that do not (eg, by indicating / selecting 13507 the word "was" at that time / moment or later) can be changed. This is because the word "has" has the highest priority value with respect to other words corresponding to the same key presses provided by the user in the database and starts with the letter "h". Although the system can be slower and more confusing, it can be enabled for data entry and modification without the user's voice. According to one embodiment of the present invention, the handwriting input and recognition system can be realized with the word prediction data entry systems of the present invention, which can also use speech recognition systems, so that for example an environment (eg quiet) or situation ( For example, in accordance with the need for quick and easy data entry, the user may use a number of keys such as word prediction data entry systems (e.g., at least four minor keys are assigned at least four language keys to at least four minor keys). Branch is used with the keys of a keypad).

The above example illustrates the word correction procedure of the system. Clearly, as mentioned above, a user can enter hand-written letter (s) during word entry. For example, (eg if the words "was" and "has" correspond to the same chain key presses and the user knows in advance that the word "was" has the highest priority) of the word "has" For input, the user may first write the letter "h" and then press the keys corresponding to the remaining letters "as" of the word.

According to one embodiment of the invention:

Indicating a location relative to (eg, within) the text that is not desired, such as an undesired word, and providing a utterance such as uttering the desired word or uttering at least some of the symbols corresponding to the desired word without providing key presses (Eg, the system may take into account the initial key presses provided by the user for input of an unwanted word), and may cause the unwanted word to be replaced with the desired word;

Indicating a position relative to the text (eg, within the text), such as a word provided by the system, providing a chain key presses and uttering a utterance corresponding to the chain key presses or at least some of the symbols corresponding to the chain key presses Providing a utterance such as one may cause the text (eg a word) to be inserted at a location (eg within a word) relative to the provided word.

According to one embodiment of the present invention, a navigation method of pointing / selecting words in a document may be considered (eg to replace them with other words, for example). For example, according to one method (eg four) arrows of the data entry systems of the present invention can be used to navigate / seek on the words of the document to indicate / select the words of the document. According to another method, forward and / or backward means (e.g., keys) (e.g., word-by-one selection) for moving the selection means forward and / or backward on words may be used for the same purpose.

According to one embodiment of the invention, during data entry in an environment, such as during a memo in a meeting, the user may have a portion of text such as a word, a long word (s), or a non-dictionary word such as non-dictionary words. You may decide to enter a chain of symbols, or even (eg, complex) phrases, functions, etc. (eg, "input-trunk text", later, such as after a meeting). Different methods can be considered for this purpose, for example:

According to one method, the user can inform the system by providing a predefined signal, for example by pressing a predefined key and uttering an "input-trunk text" such as a word. The system may record the utterance and provide one or more symbols at that location in the literature so that when the user indicates / selects the location (eg one or more symbols), the system provides the recorded utterance to the user so that the user It is possible to remember what you have done and thus enter "input-trunk text" (eg words) in the literature, for example by:

Pressing the keys corresponding to at least some of the (e.g. first) symbols of the "input-distressed text" (e.g. word) and uttering said symbols or "input-confused text" (e.g. word); or

Inputting the word using handwriting input and recognition systems, or;

- Other.

For example, when a user takes a note, a non-dictionary word, such as the person's name (eg, "Alecandrite"), may press a predefined key, such as a key that corresponds to the first letter of the word, and utter the word. Can be entered. By not discerning the word, the system provides one or more (e.g., predefined) symbols at that location in the text and the corresponding speech to that location (e.g., with one or more (e.g., predefined symbols) corresponding). Recalling that, when the user indicates / selects one or more corresponding (e.g. predefined) symbols, the system provides the user's utterance (e.g., recorded voice) to the user to remember what he did and thus, As shown, text may be entered.

According to another method, as described above, the user provides chain press operations corresponding to at least some (e.g. preferably all) of the symbols of the "input-confused text" (e.g. word) and the text (e.g. word). (S)) can be spoken. If the system fails to recognize the word, it records the utterance and provides a set of symbols, such as a chain of predefined symbols, such as the values of the keys pressed by the user at that location in the document, so that the user knows the location (e.g., the symbol). In this case, the system can provide the user with a recorded speech to allow the user to remember what he did and thus to enter the text, for example by:

Uttering symbols one by one without retyping the key (eg, by considering a chain of symbols, such as the values of stored keys, as described above, the system remembers the first key presses), or;

Inputting a word using handwriting input and recognition systems, or;

- Other.

For example, when a user notes, a non-dictionary word, such as a person's name (eg, "Alexandrite"), may be entered by pressing the keys corresponding to the letters of the word and uttering the word. By not discerning the word, the system provides predefined symbols, such as the values of the predefined key presses provided by the user, at that location in the text and returns to the location (eg, corresponding (eg, predefined) symbols). Also remembers the corresponding utterance, so that when the user indicates / selects the corresponding (e.g. predefined) symbols, the system provides the user's utterance (e.g. a recorded voice) to the user to remember what he did Therefore, as described above, text may be input.

Furthermore, according to another method, the user immediately enters an "input-distressed text" (such as a word) by his handwriting as it is digitized or written and subsequently converts it into digitized symbols (such as printed symbols) or writes it literally. I can keep it inside.

For example, when a user takes a note, non-dictionary words, such as a person's name (eg, "Alexandrite"), may be entered, for example, using a handwriting input (eg, and recognition) system. According to one method, the system can provide the text as it is in that position in the literature so that the user can later convert it to digitized symbols (eg print symbols) if desired. According to another method, the system can immediately convert the text into digitized symbols (eg printed symbols).

Other input methods such as on-screen keyboards (e.g. providing clear symbols) or other input methods such as based on the principles of the data input systems of the present invention are also referred to as "input-trunk text" (eg words). Can be used to enter

It should be noted that the entire words of any portion of text may be entered by the handwriting input and recognition system. Switching to a typing procedure (such as a key press) can inform the system of the end of the handwriting input procedure of a portion of the text.

It should be noted that recording the user's utterance (eg, voice) for some of the portions of text during data entry requires a small amount of memory but can provide the user a lot of freedom and flexibility.

There are many ways to retrieve content using a standard PC keyboard. The data entry systems of the present invention allow at least to simulate the functionality of a PC keyboard to simulate content retrieval methods.

The data entry systems of the present invention can be used for general interactive procedures such as content retrieval procedures, interactive computing systems such as interaction with PCs, interactive TVs, and the like. For example, the system can be used to search for and play movies or songs, to answer questions asked by servers on TV channels, to remotely interact with other users of the TV (eg, via TV messaging), and the like. Can be used with remote control of media center PC.

According to one embodiment of the invention, to retrieve content in a content database (e.g., a song title in a song title database, a movie title in a movie title database, the name of a person or company in a telephone directory, etc.) A data entry system can be used. The content may consist of one or more chain symbols, such as one or more words. Each piece of content may be composed of different regions, each region of which may include one or more chain symbols. For example, FIG. 12 shows a portion 13600 of a database of song titles. For example, each record corresponding to a song may include several areas, such as the song's title 13601, singer 13602, album 13603, and the like, each of which may include one or more words.

Since the data entry systems of the present invention use a keyboard that includes a few keys with one or more Jamo characters assigned to at least one key, different input methods for content retrieval may consider those described below.

According to a first method, a user can provide key presses corresponding to symbols and, if necessary, provide utterances corresponding to the key presses (eg, uttering symbols corresponding to at least some of the key presses, or It is possible to enter all the symbols of at least one area) by uttering the desired word (s) that it represents. Speech may be provided concurrently with key press operations or after key presses have been entered (eg, speech may be dynamically required by the system after processing key presses, for example). May generally require more key presses to find the content.

According to the second method, the user can provide key presses corresponding to symbols and, if necessary, by providing utterances corresponding to key presses (e.g., voting symbols corresponding to key presses or voicing words associated with key presses). A part of symbols of at least one area may be input. Speech may be provided concurrently with key press operations or after key presses have been input (eg, voice may be dynamically requested by the system after processing key presses, for example). This method typically requires fewer key presses to retrieve content but may require more speech.

For example, if the user wants to listen to a song on a music playback device associated with a media center PC, the user may, for example, have keys corresponding to at least a portion of the title of the song (e.g. preferably at the beginning). By pressing, for example, at least some symbols of the title of a song can be entered using the data entry systems of the present invention. Preferably, the user at least part of another area, such as the singer's name of the song, at least via the data entry systems of the present invention, for example by pressing keys corresponding to at least a part of the singer (e.g. preferably the first) sign. You can also enter a symbol. A delimiting signal may be provided, such as pressing a predefined key between two chain key presses.

The system may search within the area of song titles in the database to find titles corresponding to a key press chain (eg, a value) provided by the user that corresponds to the title of the song the user wants. The system may find one or more titles that correspond to key presses provided by the user with respect to the desired title. If the system finds only one title that matches the user's corresponding key presses, the system can move on to the next task, such as selecting a song and sending it to the user's playback device to be played. According to one embodiment, the next task may be to ask the user to confirm the song selected by the user, and so forth.

If there are one or more titles corresponding to the key presses provided by the user with respect to the desired title, the system selects the titles and the corresponding key presses provided by the user in the mantissa field of the selected one (e.g., second Search for the name of the mantissa (s) that match the chain key press). If the system finds only one song that matches the user's corresponding key presses, the system can move on to the next task, such as selecting the song and sending it to the user's playback device to be played. According to one embodiment, the next task may be requesting to confirm the song selected by the user, and so forth.

If there are one or more titles corresponding to key presses provided by the user in relation to the desired title and mantissa, the system may, for example, at least some of the key presses provided by the user for input of the title and / or mantissa. More information, such as voicing symbols corresponding to and providing key presses (e.g., and utterance) corresponding to more symbols in any areas or other areas, such as album titles, parts of words in a song, etc. You can ask the user to provide.

Although the example described above refers to song search, it should be noted that the search method can be applied to any type of content. For example, the movie database may include different areas such as movie title, at least one actor of the movie, movie director, and the like. According to another example, the calling directory database may include different areas such as a person's last name, a person's first name, his address, a postal code, and the like. The above-described retrieval method is also useful for the validity and other applications of the data entry systems of the present invention, such as the predictive data entry systems of the present invention combined with the push / slide and voice data entry systems of the present invention. This is just an example way of showing. As mentioned, many types of content searches may be considered by those skilled in the art. For example, according to another search method, based on the principles of the predictive data entry systems of the present invention, if a user provides a chain key press corresponding to, for example, a movie title, the key presses may be more than one in the database. If the title corresponds to the title of the movies, then the system requires more information (such as the first part, the middle part, or the end part) to provide more symbols, such as the characters of the movie title (eg, as described above). May be required. According to other systems, a user can enter symbols, for example by using the user's speech or by writing symbols (eg by hand). If the information provided by the user corresponds to at least two movies, the system may request more information, such as at least (eg, the beginning) part of the name of one of the actors of the movie, where the user may, for example, watch the movie. The above information may be provided by providing the same procedures as used for providing the title, etc.

As mentioned, content retrieval as described may be used for general purpose retrieval and interaction, although movie retrieval is mentioned as an example, the retrieval or interaction procedures may be searched for songs, responses to queries, queries, It should be noted that it may be applied to any content such as text input.

An example is provided below with reference to FIG. 12, where the remote control 13605 has a telephone type keypad 13607. The remote controller may include a speaker 13608, a microphone 13604, an input pad 13609, a display unit 13606, and the like. 12 also shows a portion of the song database 13600 as described above.

According to one search method, in order to enter the title "who" of Frank Sinatra's song, the user may enter, for example, keys 136011, 13612, 13613 corresponding to the word "who". The user then provides a limit setting signal, such as pressing the "#" key 13615, and the key corresponding to the letter "F" (e.g., the first letter of the chain symbols "Frank Sinatra") corresponding to the name of the mantissa ( 13614). The system can then compare the pressing operations with the corresponding regions. In this example, there are two entries (eg songs) that correspond to key presses provided by the user:

-"Who", a song by Frank Sinatra, and "

-"Win", a song by David Bowie.

The system can recognize that the first letters of the two songs are similar (eg, the same letter "W"), so that the system can recognize which song the user wants to select at least two first letters of the title. Can dynamically determine to speak to the user. Instead of the utterance, the system may request the second or last symbol of the song title, or request other information such as pressing a key (e.g., and if necessary) corresponding to the singer's last name or other information as described above. It is understood that it can.

Another example relating to content search within the telephone directory (eg, to ask for a phone number or to make a call) is described with reference to FIG. 13, where a portion of an example database of the telephone directory is shown.

For example, if the user wants to request the telephone number of "Eve Miller" using the keypad 13708 of the telephone 13709 by calling the automatic telephone directory, the user may enter the first three characters of the last name "Miller", for example. Three key presses (13711, 13712, 13713) corresponding to the " mil ", followed by a limit set key press such as pressing the pound key 13714, the name of the person " Eva " Two key presses 13715 and 13716 may be provided corresponding to the first two characters. As described, based on the principles of the data entry systems of the present invention, the system can recognize that there are several entries in the database corresponding to the key press information as follows:

Milker Evelyn (212) 112 2222

Miller Eve (212) 111 2221

Nikky Dudu (212) 333 2222

By analyzing the selections, the system may ask the user to input utterance at least the first four characters (eg, or the last two characters) of the desired person's last name.

As described above, instead of pressing the keys and waiting for a user's command to provide the requested utterance, the user presses the keys and provides the utterance (e.g., at the same time, or immediately after providing the corresponding press action), for example. It should be noted that it is possible to input chain symbols based on the symbol-one-one push / slide-and-voice data input systems of the present invention.

According to one embodiment of the invention, selected entries in the database based on the information (eg key presses and / or speech) are provided to the user on the screen of the electronic device to provide some additional information to select the desired entry. Note that the user can make a decision.

If an area within an entry in the database contains more than one word (e.g. a space between words), the user may provide a space symbol by pressing a predefined key, for example a "*" key, a predetermined number of times. Can be. It should be noted that the content in these areas may correspond to any symbol such as a number, or the like. It should also be noted that the retrieval procedure may be executed dynamically after the entry of each piece of information, such as a push (e.g., speech) operation. Also, according to one method, the (eg, additional) information required by / provided by the system is only spoken (e.g. without that key press), such as to utter the last symbol of a word for which no key press is provided. Can be.

Content databases such as lists of songs, movies, etc. and their corresponding information, such as artist names, actor names, etc., are available locally, such as within the user's media center PC, or from service providers such as cable TV or satellite TV companies. It should also be noted that the server may be available in a remote location, such as a server of a company or individual accessible via the Internet.

Credit card numbers generally include a predefined number of numbers. Each of the numbers may be 0-9. When a user purchases an item via a phone call, he may pay for the item by entering his credit card number through the keys of a touch tone phone. In touch tone telephones, there is only one tone per key. By using the data entry systems of the present invention that enable the input of any symbol (eg on a PC keyboard) through the telephone (eg by pressing a key and emitting a symbol), the type of credit card number including at least alphanumeric symbols Makes the entry of numbers more secure and possible even through touch-tone telephones.

According to one embodiment of the invention, when using the predictive data entry systems of the present invention, words whose corresponding chain key presses may correspond to one or more words in the system's word database may be highlighted (eg, other Written in color, underlined, bold, etc.). For example, if a user wishes to replace an unwanted word suggested by the system with a word he wishes, in order to find the unwanted word, he would have noticeably highlighted words (eg those words). Only search) can be noted.

As mentioned above, in at least a digitizer, such as a display of a tablet PC, at least some special symbols, such as punctuation symbols, include interactions (e.g., by soft / virtual keys of the corresponding keypad, or keys and And / or fingers and / or finger caps that simulate interactions with the keys. These are described in detail in the present patent application and the references contained herein, filed by the inventors. For example, a predefined punctuation symbol may hold the slide action for that key in combination with the predefined spoken information and / or press the first defined key—keep pressing the corresponding key in combination with the predefined spoken information. Can be assigned to

As mentioned above, language rules may be used to enhance any data entry systems of the present invention. For example, taking into account the portion 13508 of the database of FIG. 11A and the keys 13501 through 13504 of the keypad, when entering a phrase such as “there is no way” by pressing the corresponding keys without utterance, You can enter the phrase "there is no was" (eg "was" has a higher priority than "has"). By using a system that includes language rules, the word "was" is the only word in the phrase among the words in the database that corresponds to the key presses provided for input of the word "was". The result provided can be automatically corrected to "there is no way".

As mentioned above, the words of the system's word database can be whole words or parts of whole words (eg, first part, middle part, end part). Also, as mentioned, after each key (e.g. press) interaction (e.g., and, if present, corresponding speech) information is provided during the entry of a word, the system displays its initials (e.g. or its entire symbols) by the user. Consider the corresponding words corresponding to the provided key (eg press) interaction (eg, and, if present, corresponding speech) information (eg up to that moment).

Key presses (eg, and, if provided, corresponding speech) during the entry of a word According to different principles and / or methods of presenting corresponding chain symbols, information during word entry may be considered.

According to one embodiment of the invention, each key press (eg, if present) provided by the user during chain key presses (eg, voice, if present) information provided by the user for entry of a word After the information, the system can present to the user the word with the highest priority of all the words in which all the symbols of each word correspond to the chain information provided by the user. If the system does not find such a word, the system presents the user with the highest priority word among the words in which the predefined portion of the words (such as preferably the initials) correspond to the chain information provided by the user. can do. By considering the keypad 13800, FIG. 14 proposes chain symbols corresponding to the beginning of chain information (e.g., only key presses in this example) provided by the user during input of the word "information" according to this method. A table 13805 is shown as an example showing several (eg, beginning) steps.

According to another embodiment of the invention, each key press (eg and if provided) provided by the user during a chain key press (eg and, if provided, speech) information provided by the user for entry of a word After the information, the system determines that at least a portion of the words (such as at least the initials, or the entire symbols) correspond to the chain information (eg key presses and, if present, corresponding speech) provided by the user. The word having the highest priority among the words can be presented to the user. By considering the keypad 13800, FIG. 14 shows, according to this method, the chain symbol corresponding to the beginning of the chain information (e.g., only key presses in this example) provided by the user during the entry of the word "information". A table 13806 showing the various (eg, beginning) steps suggesting these is shown by way of example. According to this method, at any moment if the user provides a word-ending signal (such as a space symbol) and the proposed chain symbols are (eg the first) part of the word (such as with longer chain symbols) ( For example, in other words, the chain symbols are incomplete words) the system may consider words in the database, where the information provided by the user (eg key presses and, if provided) information corresponds to all symbols in each of the words For example, one may suggest a word having the highest priority among words. In the example of 13806, if the user provides a word-end signal (such as a space symbol) after three key presses, the system replaces the chain symbols "int" (such as the initials of the word "into") with the word ". put "(eg, the information provided by the user has the highest frequency among the words corresponding to all the symbols of each of the words).

According to one embodiment of the invention, during the provision of the chain information, if the user enters one or more correct symbols:

According to one method, the system can keep other symbols as they are, and add the correct symbols to the end of other symbols.

According to another method, the system can combine the information provided for the input of one or more correct symbols with other symbols and compare them against words in a database to provide the corresponding words. If there are no such words, the system can keep other symbols as they are and add the correct symbols to the end of the other symbols.

It should be noted that the chain symbols may be input by any of the data input methods (e.g., or a combination of the methods) of the present invention, such as one symbol by one, word-part / word input, predictive input, and the like. As mentioned in detail above, each input of a symbol (e.g. one by one) or a word-part / word (e.g. part of a word / word) is each different, such as providing different interactions with the corresponding key (s). For example, according to one method, a symbol / character input may be provided by a single-pressing action on a key corresponding to a symbol and by the provision of the corresponding speech, and a word- Some input may be provided by the sliding action on the corresponding key (e.g., the key corresponding to the word-part) and the provision of the corresponding voice.In addition, for example, according to another method, the word-part input may generate two keys. Can be provided by pressing (e.g., pressing a shift key and a key corresponding to the word-part) and providing the corresponding utterance. These have already been described in detail. As such, by considering the keypad 13800, in order to enter the word "revolutionize", the user first enters the key presses (e.g., key presses) corresponding to the initial characters "revolu" of the word "revolution" (eg without speech). At least some may enter information including the corresponding utterance, which may be provided without vocalization, and then he may provide the sliding action on the key 13802 and the provision of utterance of the “tion” portion (eg “shen”). Speech), the system combines the first and second information and compares it with the words in the database and outputs the word "revolution". And provide third information, including the sliding action of the image and the utterance of the “ize.” The system can combine the first, second, and third information and compare it with words in a database. Combined The word corresponding to the information may not be found, in which case the system outputs the chain symbols (eg word) "revolution" as-is, and attaches the "ize" part to the end to provide chain symbols / word "revolutionize". can do. Integrating a portion by portion of data entry systems with the predictive data entry systems of the present invention accelerates the speed of the data entry systems of the present invention and requires less key presses and less speech. And minimize user spelling mistakes.

As described in detail above, according to another embodiment of the present invention, the input of each symbol or word-part / word input provides the same interaction as providing a single-press operation on the corresponding key (s). Note that the same input signal may be assigned to the same input signal.

As previously described in the references incorporated herein by Applicant, according to one embodiment of the present invention, a portion of a word corresponds to a key press corresponding to one or more characters, such as all the characters of the portion. It should be noted that they may be input by providing them and speaking the part. For example, to enter the "tion" portion, the user can press the keys 13802, 13804, 13803, 13804 and utter the portion (eg, by uttering "shen").

According to one embodiment of the invention, during the input of information provided by the user for inputting a word by using the predictive data input of the present invention, the system is unique to the information corresponding to at least one word in the database. The chain symbols suggested by the system may not be printed until only the chain symbols are present. This can eliminate screen shakes. According to one method, up to that point, other predefined symbols, for example the corresponding key numbers, can be printed on the screen.

As mentioned above, the data entry systems of the present invention, such as the predictive data entry systems of the present invention, include handwriting input and / or recognition systems, multi-tap systems, on-line, to accurately enter at least one or more symbols of a word. It can be combined with other data entry systems such as a screen (eg QWERTY) keyboard, and the like. Other data entry systems may be used by any of the methods (eg embodiments) of the data entry and / or replacement / modification system of the data entry systems of the present invention. For example, in the corresponding embodiments of the present invention, instead of inputting a word / word-part of one or more symbols, by a handwriting input / recognition system or by pressing a key corresponding to a letter and uttering the letter, the user May enter one or more characters of a word using a virtual (eg QWERTY) keyboard. These have already been described in detail.

As mentioned above, according to one of the modification methods of the present invention, when a user indicates a word that is not desired and provides a utterance to replace the word with a desired word, the utterance is one or more of the desired words. It may be the utterance of letters.

According to one embodiment of the invention, when a user provides a utterance to indicate an undesired word and replace that word with a desired word, the utterance is at least one letter of the word (eg, first letter or last). A letter or a character predefined as a middle letter), or to utter all letters of a desired word. This will significantly increase the accuracy of the system. According to one method, the system considers both premises and thus provides the best matching word. Since in most cases providing the first letter of a word is sufficient for the correction / replacement of the unwanted word, the user may preferably speak the first word of the desired word (eg the first letter) as a first attempt. If the system still provides other unwanted words that correspond to the user's utterance, the user can provide another modification / replacement attempt, for example, by pointing again to the unwanted words and uttering all the letters of the desired word. As described in detail above, it is understood that according to another method any number of symbols of words desired to be corrected / replaced during the modification / replacement procedure may be spoken.

During data (eg text) input, typing (eg key interaction) mistakes can occur. When using the predictive data entry systems of this survey, this prevents the system from considering real words or finding those words. The user can use a delete means such as a "back space" key to delete a key (eg press) interaction (eg, and, if provided, the corresponding voice). It should be noted that, according to one embodiment, each "back space" operation may delete the last key (eg press) interaction (eg, the corresponding voice if provided) and the corresponding one or more symbols. The system may consider words that correspond to the remaining key press (eg, speech) information. According to one method, the user can then provide the pressing action (s) (eg, and, if desired, corresponding speech) corresponding to the remaining symbols of the word. By considering the remaining initial information and the additional information provided after the back space procedure, the system can predict the word (s).

Special backspace means, such as a special backspace command assigned to a key to delete one or more symbols, have been described in the references contained herein, filed by the inventors. According to one embodiment of the invention, a special back space command of the present invention (e.g., providing a corresponding interaction, such as a pressing action on a key to which a special back space command is assigned) is entered by the user (e.g., in front of the cursor). ) You can delete the last completed or incomplete word. Each additional special back space operation may delete additional last completed or incomplete words (eg, before the cursor).

According to another method, providing a special back space action may delete the last chain symbols (eg, in front of the cursor) corresponding to the user's corresponding pronunciation (eg, including key and speech (if provided) information). . Each additional special back space operation may delete one additional last chain symbol (eg, before the cursor) that corresponds to another user's pronunciation (eg, key and utterance (if provided) information). The user's pronunciation (including key and voice, if provided) may correspond to providing information between two predefined time laps. The pronunciation may correspond to chain symbols including letters, chain symbols such as part of a word, complete words, or at least one portion of at least one word and another word, and the like. Chain symbols may also include special symbols.

As mentioned above, according to one method of the predictive data entry systems of the present invention utilizing the word / word-some data entry systems of the present invention, the interaction with the corresponding key (e.g. and providing the corresponding speech) is You can provide / output chain symbols. According to one embodiment of the invention, a single "back space" (e.g., a conventional or special "back space" according to the user's and / or designer's definition of the system) (e.g. when the cursor is at the end of the chain symbol) is The chain symbols may be deleted. According to one method, the special symbols may be deleted if the chain symbols and one or more special symbols such as a space symbol or a punctuation symbol after the cursor are located after the special symbols.

Words in the dictionary database can include any special symbols, such as punctuation symbols. For example, by considering FIG. 14, in order to enter the word “don't” (eg, present in the database), the user may use keys 13804, 13803, 13804, 13802 (eg, an apostrophe symbol 13809). 13802) and provide the corresponding vocalization if desired / needed.

According to one embodiment of the present invention, pressing the "shift" key before or during the entry of a word through the predictive data entry systems of the present invention may result in capitalization of the corresponding (eg first) letter of the word. . According to another method, the user can press (e.g. and hold) the shift key and press the key corresponding to the (e.g., the first) letter of the desired word input via the predictive data entry system of the present invention and utter the letter. have. The user can then provide key presses corresponding to the remaining symbols of the word (eg, to release the pressed key). Letters can be capitalized within words. According to another method, instead of the pressing action, the sliding action (e.g. and corresponding speech) for that key may be predefined to serve the same purpose. Clearly, when providing a key interaction, if the user does not provide that utterance, the system may consider the key interaction as an unclear key interaction and the corresponding letter in the predicted word (s) may be capitalized. .

According to one embodiment of the invention, one or more symbols in a word may be deleted or changed.

According to one method, the system may consider key interaction (eg, and utterance) information of the changed word and compare it with words in the database to provide that word (eg, as described in detail). According to another method, the system can keep new chain symbols (eg words) as they are.

Although keys have been mentioned as input means for providing input signals through many embodiments of the present invention, it is apparent that any other input means (eg finger interactions on a touch screen, finger caps, etc.) may input the input signals. It should be noted that it can be used for

According to one embodiment of the invention, during the entry of information corresponding to a word (eg at any moment), the user replaces the initial symbols (eg letters) of the corresponding chain symbols proposed by the system (eg the same corresponding key). To other symbol (s) with interactions. Without providing key interactions or using predefined means, such as a predefined key, a user can speak one or more symbols. According to one method, by considering the information provided for the changed symbols and the remaining chain symbols and based on a comparison with the system's database, the system can generate one (eg or more) new candidate chain symbols (eg words). ) And suggest it to the user.

According to one embodiment of the invention, in order to change one (e.g. or more) symbol of a chain of symbols such as a word (e.g. to another symbol (s) having the same corresponding key interactions), the user moves the cursor. It may be located in a predetermined position with respect to the symbol and provide a utterance corresponding to the desired symbol (s).

According to an embodiment of the present invention, during the modification / replacement procedure of the predictive data entry systems of the present invention, by using a predefined means such as a key, the user can speak the speech he provides for the correction / replacement of the corresponding proposed word. (Eg, the utterance of one or more letters) may inform the system that it is associated with a defined region of the word, such as the termination region of the word.

According to another embodiment of the present invention, during the modification / replacement procedure of the predictive data entry systems of the present invention, the user may use a predefined means such as selecting / indicating a portion (eg, one or more characters) of a word. The utterance he provides (eg utterance of one or more letters) for correcting / replacement of the corresponding proposed word may inform the system that it is associated with a portion of the word. The portion may be any portion of a word, and the (eg selection / instruction) means may be any means for selecting or indicating a portion of a word. According to one method, the user may indicate a position in a chain of symbols (such as a proposed chain) such as a completed or incomplete word (eg by placing the cursor at that position) and the system may be in front of and / or the cursor. Or may be predefined to select all or part of the symbols of the chain of symbols later (eg as designed). The user can then provide the desired utterance corresponding to the selected portion. According to one method, the system thus provides key (eg press) interaction information (eg and if provided by the user) for the changed symbols and other symbols of the chain of symbols in order to change the entire chain symbols. Corresponding vocalization) (eg, considering the candidates and suggesting one from them, as described above).

Although some paragraphs refer to the beginning of a word for the corresponding information provided by the user for the input of the correct symbol (s), it is apparent that any of the words for the entry of a word based on the principles of the combined data entry systems of the present invention. It should be noted once again that part of and the information of the corresponding user may be taken into account.

Also, although only some key presses are mentioned as information provided or considered by the user in some paragraphs, based on the principles of the data entry system of the present invention, the information may include the user's speech if provided. It should be noted once again.

The data entry systems of the present invention (such as / including prediction data entry systems of the present invention, such as the prediction data entry systems of the present invention) may use multiple languages simultaneously. For example, utterances of one or more languages may be assigned to symbols such as (eg Roman) letters, words, word-parts, etc. of the data entry systems of the present invention. For example, utterances of letters of two or more languages may be assigned to each of the letters of the Roman alphabet assigned to the keys of the keypad. For example, utterances of letters in two languages, such as English and French, may be assigned to each of the letters. For example, the letter "z" may thus be pronounced "ziy" or "zehd". The multi-lingual data entry system of the present invention may comprise a database comprising the (eg word) predictive data entry systems of the present invention and comprising words (eg and word-parts) of multiple languages or Or multiple databases of (eg words) (eg and words-part) of multiple languages such as, for example, English, French, and the like. Different embodiments may be considered in order not to consider other languages of the system (eg, to increase the accuracy of data entry) while entering data (such as text) in one of the languages.

According to one embodiment of the invention, means such as a key or a user's (eg voice) command may be used to switch from one language to another.

According to another embodiment of the invention, switching between languages can be made automatically by the system. According to one method, when a user enters a symbol (such as a letter, word, etc.), the voice is said to be a voice corresponding to one language by the system in order to interact with that key and provide that voice. Can be considered. The system may then switch to that language and not consider other languages until the user provides the utterance of the symbol in another language. For example, to enter the phrase "my mother in English means ma mere in French", for example, using the word prediction data entry systems of the present invention, the user may, first of all, use the first part "my mother in English means". May be entered by one of the data entry systems of the present invention as usual. He then presses the word "ma" (e.g. a French word) on the keys corresponding to the word and provides a utterance corresponding to at least one of the letters of the word (e.g. in French) or the word (e.g. in a data entry method). By speaking in French). The system may recognize that the word is a French word and act accordingly. For example, if the data entry method used to enter a word is the (eg word) predictive data entry method of the present invention, the system may use the French database of words of the present invention to enter the word. The system may assume that the user enters data in French until the user provides utterance in another language. In this example, the user may enter the subsequent word "mere" (eg French word) by pressing the corresponding keys (eg if the user uses the predictive data entry method and the system enters the word of the present invention). If you have access to their French database, you may not need to provide a voice).

The user can then enter the last part "in French" by one of the data entry systems of the present invention. While providing key presses corresponding to the word "in", the user may provide utterance information in English, such as to provide a utterance corresponding to at least one of the symbols of the word "in" (eg, to speak the word). Or at least one of the symbols such as "ay" in the word).

It is understood that other manual or automatic switching methods may be considered. For example, based on the previous word of the language entered / selected by the user, the system may transmit the subsequent word to the language unless the user provides information (such as those described above) for switching to another language. Can be regarded as the word of.

According to one embodiment of the predictive data entry systems of the present invention, a symbol, such as a symbol, is considered by the system as a word-part of a database (e.g., the symbol is its corresponding during chain key presses for entering the word). When a key press is provided (e.g., pre-defined as vocal or absent), it is considered to be part of the word to be predicted by the word prediction procedure, or predefined to be considered as an out-of-word symbol, or both in advance Can be defined.

The same symbol can be assigned to any predefined type of interaction with the first key or the second key, and can be considered as a word-part or a non-word symbol, respectively (eg or vice versa).

The same symbol may be assigned to the same key but to two different types of information, wherein by providing a first type of interaction with the key, the system considers the symbol to be a word-some symbol, and By providing an interaction of and providing second speech information, the system considers the symbol to be an out-of-word symbol (eg, or vice versa).

For example, the symbol "-" may be part of a word in a database, such as the word "semi-final", or may be used in a formula or a separator between two separate words (eg "single-double"). It may be a symbol other than the same word. Predefined interactions (eg, key interactions, and / or types of speech information), such as predefined user behavior, can inform the system of the nature of the symbol. For example, when entering the word "semi-final" (such as a word in a database), a user may enter the corresponding keys without providing a utterance corresponding to the symbol "-" (for example in this example, Not providing utterance when single-pressing the key corresponding to the symbol "-" may inform the system that the symbol corresponding to the key press is part of the word to be predicted by the system). On the other hand, for example, if a user utters a utterance corresponding to the symbol "-", the system may consider it an out-of-word symbol and regard it as the word-end signal of the previous word. For example, to enter two words "single-double" containing the separator symbol, the user can press the keys corresponding to the word "single" and then press the key corresponding to the separator symbol. Press and speak the symbol. The system understands that the spoken symbol (eg, punctuation) is not part of the previous word. The system enters the word "single" and the separator "-" and considers the next key presses (e.g. and corresponding utterance (s)) to be part of the next word (e.g. "double" in this example).

According to another example, when entering the word "semi-final" (e.g., a word in a database), the user may have a predefined utterance corresponding to the symbol when the user presses a key corresponding to the symbol "-". It is possible to press the corresponding keys (e.g., in this example, to provide a single-pressing action (e.g. with or without speech) on the key corresponding to the symbol "-"). The symbol corresponding to the key press action for may inform the system that it is part of the word to be input). As predetermined, other pressing actions on the key or other predefined key to which the symbol "-" is assigned may indicate that the symbol provided by the interaction may not be considered part of a word by the system. Can tell the system.

Based on the principles predefined by the manufacturer or user as described, each of the symbols of the data entry systems of the present invention are part of a word, for example based on the type of interaction provided by the user. Or, other than words, or both. As a further example, a word-part of the present invention, such as the suffix "tion", may be considered part of a word and may be a first type, such as a single or double pressing operation on that key and the provision of utterances corresponding to that word-part. Is assigned to the interaction. Also, for example, the same word-part (e.g., the suffix "tion") of the present invention may be considered as an out-of-word symbol and may provide a single or double glide action for that key and the provision of utterances corresponding to that word-part. It can be assigned to the same other type of interaction.

It should be noted that in some paragraphs of applications relating to the data entry systems of the present invention, such as the predictive data entry systems of the present invention, the term word-end signal is used to describe non-word symbols.

Based on the principles described above, symbols such as symbols provided by other systems (e.g. and used with the word prediction systems of the present invention) such as handwriting input and recognition system, complete keyboard, and the like are pre-defined. May be predefinely word-some symbols, or extra-word symbols, or both.

As mentioned above, in the predictive data entry system of the present invention, a symbol may be predefined and assigned to at least two types of user interactions, where in one of the interactions the symbol is part of a word. Or (eg, as compared to words in the system's word dictionary) or as an out-of-word symbol. For example, the letter "a" may be assigned to a key and may be entered as part of a word, for example by providing a corresponding press action on that key without providing utterance. On the other hand, for example, the same letter "a" may be regarded as an out-of-word letter and may be input, for example, by providing a corresponding pressing action on the key and uttering the letter.

As examples of the described methods, for example, the letter “a” can be assigned to a key and for example a word by providing a predefined pressing action such as a single pressing action for that key with or without providing utterance. (E.g., uttering the character may correctly enter the character, and not uttering the character, the keypress action is unclear and assigned to the predefined interaction with the key). The system may correspond to any symbol containing the letter “a”). On the other hand, for example, the same letter "a" can be considered as an out-of-word letter and for example, another predefined interaction with that key (eg slide action, longer time) with or without utterances. Can be entered by providing a pressing operation, pressing a key strongly, pressing-holding on a first key and pressing two keys, such as pressing the corresponding key, etc. (eg, according to one example, Characters can be input accurately and precisely by the interactions and utterances of the characters, According to another example, the characters can be input accurately and precisely by providing the key interaction without utterances. It may be the only character assigned to the key, although in this example the input of the character is used to describe the methods, preferably, the key interaction is performed without speech. The symbol to be assigned to the key to be input by providing it may be a symbol other than a punctuation symbol or a character such as a command / function).

It is understood that the described examples can be applied to any type of symbols of the data entry system of the present invention, such as characters and punctuation symbols.

As mentioned above, in the predictive data entry system of the present invention, the word-end signal can be used to inform the system of the end of chain key presses (e.g., and speech information) corresponding to a word. For example, as mentioned, in accordance with one embodiment of the present invention, the word-end signal may be an out-of-word symbol or a signal that does not provide any symbol (e.g., with a predefined pressing action for a predefined key). As also mentioned, according to another embodiment of the present invention, the word-ending signal may be a symbol, such as a letter, which may be defined as being the beginning of a word (e.g., next). This has already been explained in detail.

According to one embodiment of the present invention, one or more shift keys (eg, in the predictive data entry system of the present invention) may be considered to allow various types of information for data entry. For example, an uppercase letter may be provided by pressing (eg, holding) the first shift key and pressing another key corresponding to that letter. According to different methods, uttering the text may, for example, inform the system that:

The letters may be capitalized and may be predefined as non-word letters, or;

Said letter may be capitalized (eg next) and may be predefined to be regarded as the first letter of a word, or;

The letter may be capitalized and may be considered part of the current word, or;

- Other.

In addition, according to one method, for example, not uttering the letter may inform the system that the letter in the corresponding (eg recognized) word may be capitalized.

Also, for example, the second shift key may be considered for use with at least the word-some data entry system of the present invention. For example, a part of a word may be provided by pressing (eg and holding) the second shift key and pressing another key corresponding to the word-part and voting a predefined utterance corresponding to the word-part. Also, for example, pressing (eg and holding) the second shift key and pressing another key without utterance may correspond to other types of symbols, such as punctuation symbols or special symbols such as functions.

In devices with sensitive surfaces, such as touch screens, another or additional input method, such as single or double slide operations (e.g., tapping plus slide operation) (e.g. for a key) (e.g., and, if necessary, corresponding speech) These may be provided for the input of corresponding data, for example more and more symbols. For example, instead of pressing (e.g. and holding) the shift key and pressing another key corresponding to a word-part, the sliding action for that section / key corresponding to that part and a predefined utterance corresponding to that part Can be provided for input of the part.

As mentioned above, according to one embodiment of the present invention, the predictive data input systems of the present invention may include the symbol data input system of the present invention and the word-part data input system of the present invention. In some cases for less speech (eg, words with multiple syllables, such as preferably at least three to four syllables), a word-part may be added by interaction with the corresponding key without providing speech. Can be entered. Based on the portion of the word entered by the user and other portions, the system can recognize the word.

According to an embodiment of the present invention, based on the principles of the predictive data input systems of the present invention, a table of predefined symbols may be considered as each symbol being word-partial or non-word-part. Each of the symbols may be assigned to at least one predefined type of interaction with that key in the presence or absence of the utterance (eg, or both). The table of symbols may be created or changed by the manufacturer and / or the user.

As mentioned above, according to one embodiment of the present invention, tapping on four different locations on the surface may correspond to four keys used by the data entry systems of the present invention. According to one embodiment of the present invention, tapping at different locations on the surface may provide different corresponding sounds or vibrations, and the like. The system can recognize each of the sounds and corresponding predefined virtual keys and use them by the data entry systems of the present invention. This principle can be achieved, for example, by tapping on four different positions on a non-touch sensitive screen of a computer (e.g. combined with the corresponding speech based on the principles of the data input system of the present invention). It can be useful for data entry. According to another method, the inventors such as slide in four different directions on the surface to provide four different sounds (eg PCT / US / 01/49450 (“stylus computer”)). (As described above in other patent applications filed by)) may be used for the same purpose. At least some of the inventions of the PCT patent application may be combined / used with the combined data input systems of the present invention. Some of the combinations have been described in detail in this patent application.

As mentioned above, the data entry systems of the present invention may be combined with handwriting input and recognition systems. As also mentioned above, for example, during word entry based on the predictive data entry systems of the present invention, at least one correct, key interaction for each entry of a symbol, such as a letter (eg or part) of the word Instead of, or in addition to providing corresponding speech information, the letter (s) / portion (e.g., for separation, for example) may, for example, be placed on a corresponding input device (e.g. Can be written by user). The information corresponding to the other symbols of the word may be to press the corresponding keys without utterance (eg based on the principles of the predictive data input systems of the present invention). Such a data entry system may be of any type (eg, reduced), such as those described in the references contained herein, filed by the inventor, such as the dynamic keypad of the present invention (eg having at least four keys). It is understood that keyboards can be used.

(For example, as described in previous patent applications, the dynamic keypad system interacts with (eg, fictitious) a predefined keypad model in a non-predefined portion of the (touch-sensitive) surface). (E.g., by a finger, stylus, mouse, etc.) The procedures for recognizing interactions provided at a location on a surface and associating the location with the corresponding key in the keypad model are described below. Regarding the relationship between the positions of the keys of the predefined keypad model concerned, based on the recognition of the relationship between the contact position on the surface encountered by the user and the other positions (eg, previous and / or next) These have been described in detail in the present patent applications.

For example, by using the (e.g. dynamic) keypad of the present invention, a user can select a symbol of a word (e.g., on an input panel of a handwriting input / recognition system, or anywhere on the screen of the corresponding data entry surface). ) And use tapping / sliding interactions (eg, described above), which correspond to the input of other symbols of the word. The system can easily distinguish between write and tap operations. In this example, a dynamic keypad was used to illustrate the principle of an embodiment of the present invention. As mentioned, any other keypad can be used with embodiments of the present invention.

According to one method, when writing and / or tapping operations are allowed to use the same input surface (eg, any point on the touch screen surface), the writing / sliding operations may correspond to handwriting input and both The drip operations may correspond to a key press operation. For example, to enter the word "book", the user provides the first (eg accurate) information by writing the letter "b" (eg by the stylus or by his finger) on the touch screen of the tablet pc. And then press a second (eg, unknown) key (s) corresponding to the remaining symbols "ook" of the corresponding keypad / keyboard used with the system (eg hard / soft / virtual / dynamic, etc.) Information can be provided. As mentioned above, the user can write any one or more letters of the word and provide key presses corresponding to the other letters of the word. According to another example, the user presses the keys corresponding to the beginnings of the word (eg, "bo" in this example) on the surface of the electronic device, etc. to provide the first (eg indeterminate) information and the remaining characters of the word. Words (eg, "ok" in this example) may be used to enter the word "book" by providing second (eg accurate) information.

Any number of systems, methods, features, and the like of the data entry systems of the present invention can be combined together to provide a particular data entry system that can be used in the environment.

As mentioned above, each different type of input information may be a different type of interaction with the corresponding key (e.g., to provide information for entry of a word through the predictive word input systems of the present invention). Can be assigned to For example, entering an unclear character in lowercase can be assigned to a short single-press action on that key without speech, and entering the same letter in uppercase can be assigned to a long-press operation on a key without speech. have. Also, for example, entering the correct letter in lowercase may be assigned to a short single-press action for that key with the utterance of the letter, and entering the same exact letter in uppercase with the utterance of the letter May be assigned to a longer-press operation on. Also for example, entering a word-part may be assigned to a double-press or glide action (eg and to provide a corresponding utterance) for that key. Also, for example, entering one of the few punctuations or at least some of the functions or commands may press-hold the first (e.g. change / shift) king and provide a press action for that key with / without vocalization. Can be assigned to And so on.

It is understood that the above-mentioned interactions and assignments are used for demonstration purposes only. Any (eg, type of) interaction, and the assignment of (eg, any type) symbols for the interaction, may be considered based on the described principles.

Various methods for correcting unwanted words provided by the system corresponding to the predictive data entry systems of the present invention have been described above. Other modification methods may also be considered. For example, when an unwanted word is selected / indicated, according to one method, if the desired word starts with the same (first head) letter as the unwanted word, the user can enter the predefined word such as pressing a predefined key without utterance. Can provide a signal. The system suggests another word within the next higher priority word group of words in the word group that corresponds to the same key presses as the unwanted word and starts with the same letter as the unwanted word and preferably starts with the letter. can do. Also, for example, if the desired word does not begin with the same (first letter) letter as the unwanted word, the user can provide a utterance corresponding to (eg or writing to) the first letter of the desired word. The system corresponds to the same key presses as the unwanted word and starts with a letter that corresponds to the user's utterance (e.g., written symbol) and preferably has the highest priority among the words of the word group beginning with the letter. You can suggest words within.

As an example corresponding to the above-mentioned method, by considering the keypad 13900, FIG. 15 illustrates a system corresponding to chain key presses 13904, 13903, 13901 (e.g., their key press value is "431"). A table 13909 is shown containing words from the word database. As mentioned above, according to one method, providing chain key presses 13904, 13903, 13901 without utterance may correspond to the word "now" having the highest priority among such words. If the word is an unwanted word entered in the text, while the word is selected / indicated, the system starts with the letter “n” if the user presses a predefined (e.g., modified) key, for example, without providing speech. The word "nvq" corresponding to the word group 13906 may be proposed, where the word has the highest priority after the word "now" in the group. Also, for example, if a user provides a utterance (eg, with pressing a modifier key) such as uttering a "d", the system is most likely among words of the word group 13905 beginning with the letter "d". We can suggest the word "doe" with a higher priority. An additional pressing action (e.g. without speech) on the modifier key may provide another word of group 13905 (e.g., "dye" in this example), where the word "doe" within the group. Next highest priority. Another additional pressing action (e.g. without speech) on the modifier key may provide another word of group 13905 (e.g., "dow" in this example), where the word is "words" within the group. doe "and" dye "have the next highest priority. And so on.

According to one method of correction, if the user goes to correct an unwanted word suggested by the system, for example, pressing the modifier key and uttering the initial letter of the desired word with the same initial character as the unwanted word, the system May suggest a word from a word group that corresponds to the same key presses as the unwanted word and begins with the symbol, where the suggested word has the next highest priority after the unwanted word in the word group.

According to another method of correction, if the user undertakes correction of an unwanted word proposed by the system, for example, pressing the correction key and uttering the first letter of the desired word having the same first letter as the unwanted word, The system may suggest a word in a word group that corresponds to the same key presses as the unwanted word and begins with the symbol, where the suggested word has a higher priority among the word group (e.g. Having the highest priority among the word groups, the system may re-suggest words.

According to one modification method, in some cases the modification procedure may additionally provide a space symbol. For example, if the user wants to correct the currently proposed word immediately after input for information corresponding to the word (eg key presses and, if necessary, a voice) and after inputting a word-end signal such as a space symbol If so, he can provide a predefined signal, such as a space key or a correction key, and utter the corresponding utterance (eg, the initial letters of the desired word). The system can thus replace the word and provide a space symbol after the word.

Instead of the above-mentioned correction methods, it is understood that the user can use any other modification methods such as those described above such as uttering some or all of the letters of the word.

According to one embodiment of the invention, the system can automatically learn the user's text exchange habits and provide corresponding priorities to words used by the user based on the frequency of use for words by the user.

As mentioned above, in order to enter a character, the user may enter a key to which the character is assigned and provide a utterance corresponding to the character (eg, uttering the character). According to one embodiment, for example, if the character is a consonant, the utterance may be a pronunciation of a phoneme of a consonant followed by at least a phoneme of a vowel. According to one method, the vowel phoneme may be at least any of the vowel phonemes of the language. For example, the letter "t" can be entered by pressing the corresponding key and uttering at least one of the following combinations:

"Ta" (e.g. corresponds to the beginning of the word "tap")

"Te" (e.g. the beginning of the word "test")

"Te" (e.g. corresponding to the beginning of the word "tee")

-"To ^" (e.g. the beginning of the word "told")

-"Tu" (e.g. equivalent to the beginning of the word "tune")

-"Ti ^" (e.g. equivalent to the beginning of the word "ride")

This method can be used with the combined (eg, including prediction) data entry systems of the present invention. For example, to enter the word "test", the user presses the key corresponding to the letter "t" and utters "teh" (corresponding to the vowel of the corresponding (eg first) syllable of the word), without speech The key corresponding to the remaining symbols of the word can be pressed. Also, for example, to enter the word "told", the user presses the key corresponding to the letter "t" and utters "toh", and presses the keys corresponding to the remaining symbols of the word without speech. For some users this is more user friendly than speaking letters (eg, "ti"). As mentioned above, it is understood that the user can provide more than consonants and subsequent vowels. For example, the user may pronounce the first consonant, the subsequent vowel, and the subsequent consonant of the word. These have already been described in detail in the references incorporated herein by the inventors.

As noted above, it should be noted that instead of the user's speech, other user interactions may be combined with the user's key interaction. Less input based on the user's idea is being tested in laboratories. For example, although not yet realized, in order to enter a symbol such as a character, the user may press the character and think about the character. The symbols can be distributed to the keys of the corresponding keypad to increase the accuracy of the data entry.

As mentioned above, other systems for inputting correct symbols, such as letters, in addition to or in place of the user's utterance may be used with the predictive data entry systems of the present invention (e.g. substantially all characters of the language). Using a few keys, such as the four keys assigned to them). As also mentioned, in addition to interaction with the input means of the system to provide input signals (corresponding to a group of symbols) corresponding to unclear symbols, the data input systems of the present invention may be implemented using a virtual or physical keyboard ( Provided for corresponding correct symbols, such as for example using a QWERTY keyboard in addition to the four unclear keys of the system), interacting with handwriting input and recognition methods (eg to enter correct symbols), and the like. Other input signals may be used.

FIG. 16A shows, by way of example, a multi-directional keypad with four keys assigned to letters of a language (eg, as shown for each key). For example, pressing the center of the keys may correspond to ambiguously entering one of the letters assigned to the key. A chain of such key presses corresponding to the desired word may be considered by the predictive data entry system to provide at least one corresponding word. On the other hand, a pressing operation on a symbol such as a character on a key (for example, a character section) may correctly input the symbol. For example, by considering the keypad 14000 of FIG. 16A, the (eg, central) chain press operation of the keys 1402, 14001, 14003, 14002 may correspond to the words “test” and “best”. . The word "best" may generally have a higher frequency of use. The system can thus suggest the word as a result of the pressing operations. In this example, if the user wishes to enter the word "test", he can first press the letter "t" (14005) (area) of the key 1402 and press the centers of the keys 1401, 14003, 14002. have. The system understands that the word corresponding to the chain key presses for keys 1402, 14001, 14003, 14002 begins with the letter "t". Thus in this example, the word "test" can be suggested by the system.

FIG. 16B shows a keypad 1410 with touch sensitive keys to mimic the functionality of the multi-directional keypad 14000 by, for example, touching the corresponding regions of the keys of the keypad 14010. In this example, according to another input method, the pressing actions for any zone on the key of keypad 14010 may correspond to an indefinite symbol on the key, for example towards a symbol on the key (such as printed). Or the sliding action on the symbol may correspond to a (eg correct) symbol on the key.

Based on the principles of symbol assignments for a few keys of the (eg combined) data entry systems of the present invention, any special symbols, such as numbers, punctuation symbols, instructions, functions, etc., may be combined with the provision of speech. Or may be assigned to predefined interactions with the keys (eg or additional keys) without providing for vocalization.

In the predictive data entry system of the present invention, single characters / symbols may have priorities with respect to each other. For example, the letters "a" and "i" (eg, taking English into account) can clearly have a higher usage-frequency / priority than other English letters. For example, taking into account the keypad 14000 of FIG. 16A, the letters “m”, “a”, “s”, and “i” are different from their respective corresponding keys 1401, 14002, 14003, 14004. It may have the highest priority over characters. The user would otherwise benefit from priority assignments for the letters to enter some words based solely on the provision of a key press, although they would have needed some speech. For example, by considering the keypad 14000, if the user wants to enter the word "as" by pressing the corresponding keys 1402, 14003 (e.g., the center of those keys) without speaking, the system presses the same chain key. Can be suggested, but the word "to" has a higher frequency of use / priority. To enter the word "as" without utterance, the user presses the key 1402 corresponding to the letter "a" and then can provide a word-end signal that does not provide a space symbol (any symbol such as). (E.g., by pressing a corresponding key), and then a key 14003 corresponding to the letter "s" may be pressed. In this example, the letter "a" has the highest priority among the letters assigned to the key 1402 (eg, in a single press operation of the key). The system suggests letters. By providing a word-end signal, the system recognizes that the next key press to hold the letter and enter a word corresponds to the entry of the next / other word. In this example, the letter "s" has the highest priority among the letters assigned to key 14003 (e.g., a single press operation of that key). By pressing key 14003 without speech, the system suggests the symbol "s" and prints it immediately after the letter "a" to provide the word "as".

The use of the handwriting input and recognition system with the (e.g. predictive) data entry system of the present invention has already been described (e.g., the predictive system of the present invention is described in such a way as to interact with the keypad of the system and write down the correct symbols on the surface. Help us predict more accurately). According to one embodiment of the invention, a user may write a symbol, such as a character, on a (eg, sensitive / virtual) key / area of the keypad of the system corresponding to the character. The handwriting recognition system may match a character having a small number of characters with one of symbols assigned to the key / area. Thus, processing time can be reduced and accuracy of recognition can be increased.

The writing input may be performed by writing on a surface by means such as a stylus or a user's finger. According to one embodiment of the invention, when using a split version of the keypad of the system (e.g. located on the opposite side of a device with a touch sensitive surface) the user presses the keys with the corresponding thumb and the surface if necessary. Write down on your finger (e.g. thumb, index finger, etc.) on the top to provide accurate symbols that can be used as the (e.g. predictive) data entry system of the present invention (e.g. at least one of the symbols of a word by handwriting input / recognition system) Providing a first type of information consisting of correctly entering a key and providing at least a second input of information consisting of providing a key press corresponding to (eg, indeterminate) at least some of the remaining symbols of the word. The input of a word through is described in detail above).

As mentioned above, the data entry systems of the present invention can be used as any application or devices. One popular device is the iPOD music / media player. The apparatus comprises content search means, consisting of scrolling a menu by a circular sliding action on a circular touch sensitive surface. If the list of contents is long, the procedure takes time. The data entry systems of the present invention may replace the search (eg and selection) procedure or assist to operate faster. For example, the user may enter one or more (eg start) symbols corresponding to the desired content (eg song, singer or both, etc.). The device's pop-up menu can thus provide (eg a reduced number of) content to the user, and the user can search within the reduced list using the device's (eg circular) content search means. can do.

The use of movements, such as the user's body movements (such as the user's lips to provide vocalization corresponding to a user's key press) with the data entry systems of the present invention, has been described above. The movements can be used with the predictive data input systems of the present invention to reduce the number of speech utterances provided by the user.

According to one embodiment of the invention, the action sensitive means replaces the next word in the list of words corresponding to the user's information (e.g., key presses, and / or vocal chains) during the input of the word or in the system; It may be used with (e.g., predictive) data entry systems of the present invention in providing the correct procedure of the present invention for the purpose of informing. For example, by considering the 14000 keypad of FIG. 16, the chain of key press operations (eg on the middle) of the 14002, 14001, 14003, 14002 keys may correspond to “test” and “best”. The word "best" can generally have more frequent uses. Thus, the system can suggest the result of the word press operations. In this example, for example, if the user wants to enter the word "test" while pressing the corresponding keys, the user can provide lip movement actions (eg, without providing a voice) (eg, the letter "t"). By saying or saying "next" or the like). The camera of the device can detect the movements and the system can use the next word in the list (e.g. second use priority) after a word provided mainly by the system (e.g. "best") if no lip movements are provided by the user. Understand that you want. This system is different from the lip reading and recognition system. In this embodiment the system does not need to recognize what the user is saying but only detects that there is movement. Understand that other interactions and corresponding sensing means can be used for the same purpose. For example, heat sensitive means may be used instead of the camera to detect what the user is saying. According to another embodiment, the movement detecting means provided by the user may be a movement sensing means provided in a corresponding device such as a mobile telephone. Instead of lip movements, the user may provide a predefined movement of the device upon providing a chain of corresponding key presses.

As mentioned above, in order to increase the accuracy of the system, several symbols with indistinctly similar utterances related to certain symbols (eg on one key) are assigned different utterances so that they can be distinguished from other symbols. For example, to better distinguish the utterance of the letter "m" (e.g. pronounced "em") from the utterance of the letter "n" (e.g. pronounced "en"), instead of the standard utterance of the letter "m" Or in addition, another utterance such as "ma" is assigned to the letter "m" to distinguish it from the utterance of the letter "n".

As described in detail above, the data entry systems of the present invention may use a keypad with a few keys, such as a keypad with at least six keys, and substantially all of the characters of one language are assigned to six of those keys. Assigned. The assignment of such characters to the keys can be made to mimic a chain of QWERTY keyboards. For example, the QWERTY of the PC keyboard 14100 using a group of six keys of the reduced keyboard 14110 (e.g., 3 rows 14117-14119 x 2 columns 14111-14112) as shown in FIG. The assignment can be simulated (eg, by simulating imaginary rows 14107-14109 x columns 14101-14112 constituting the alphabetic symbols of keyboard 14100). For example, the upper left key 14113 may include “QWERT” characters corresponding to characters located on keys of the left side 14101 of the upper row 14107 of the PC keyboard 14100. The other keys of the group of six keys follow the same principle and include the corresponding characters located on the corresponding row- and convenience keys of the PC keyboard 14100.

Experienced users of Qwerty keyboards often know the exact location of each character on the keyboard. The movement reflex allows the user to quickly type on the QWERTY keyboard. Simulating a QWERTY assignment on six keys, as described herein, allows a user to touch-type (quick typing) on a keyboard with a reduced number of keys. For example, a user may use both hand thumbs (left thumb for the left column and right thumb for the right column) for data entry. Typing on the PC keyboard to enable rapid data entry reminds of this. As mentioned above, as an example, each of the key columns can be located on the opposite side of the electronic device and used as data input systems of the present invention.

It is understood that the left and right symbol definitions of the keyboard described in the above examples are shown by way of example only. The definition can be reconsidered according to the user's habits. For example, the letter "G" may be considered to belong to the right side rather than the left side.

As mentioned above, in order to increase the accuracy of the system, there is an unclearly similar utterance associated with each other with a customized utterance (eg, "m" and "n" characters on the telephone keypad or on the keypad 14110 of FIG. 17). Assign at least some characters (eg on other keys or on the same key) so that they can be distinguished from each other. For example, the letter "m" (such as "em") can be assigned another name (such as "ma") such as "ma" so that it can be distinguished from the letter "n" (such as "en"). do.

With continued reference to the example of FIG. 17, the same principle is applied to unambiguously similar speech such as "i, y", "s, f", "d, g", "j, k", "z, c", and the like. Make sure you have different characters from each other.

It is understood that the keypad, key configuration, symbols such as letter assignment on the keys of the keypad, customized designations of the symbols, etc. are described and shown as examples. Other keypads, key configurations, symbols such as character assignments on the keys of the keypad, customized designations of the symbols, and the like may be considered by those skilled in the art. For example, the keypad may be in alphabetical order and a symbol, such as a letter assignment, may allow, for example, to assign two or more customized utterances (eg, in addition to the standard utterance) to a letter.

Dialing by name is described in the references contained herein submitted by the inventor. Dialing by name can result in ambiguity, for example, where two other people can have the same name. According to one embodiment of the present invention, call forwarding / switching centers, telephone operators, service providers, etc. may use a database in providing a call or obtain information secretly (e.g., upon request) for a user / call. You can give the person information about the person you are calling. For example, when a user makes a call based on the user's name, the center provides information such as the address, photo, predefined phrase, password request, etc. of the calling party so that the user You can be sure you are calling. The calling party may be a person, a company, an organization, or the like. Information about the calling party may be provided by itself and / or by a third party or through a database used by a call forwarding center. For example, in addition to an address, a photo, a favorite song, or other information of the user may be provided to the call forwarding center so that at least some of the information is delivered to the caller, so that the caller makes a call based on the information. Matches can be confirmed and the call can be confirmed or canceled based on that. The information provided to the user may be voice or displayed on the display of the user's electronic device (eg, a telephone).

As mentioned above, when using the predictive data entry systems of the present invention using several words and utterances to enter a word instead of or in addition to utterance use, at least one of the letters of the word is handwritten (and recognized). ) A predetermined press action on one key (eg in some cases one or more keys) of a keyboard (eg, a standard QWERTY keyboard) or other data input means such as a system generally corresponds to a symbol or function (eg mechanical, Virtual input). These have already been described above. As described above, it is understood that other input means or methods may have other forms. For example, the keyboard may be an on-screen keyboard having a standard PC keyboard form, a circular form, and a circle divided into two semicircle portions or two quarter circle portions on the opposite side of the device.

As another example, FIG. 18 shows an electronic device 14200 such as a tablet PC with a touch screen display. The predictive data input system of the present invention using some keys (eg mechanical keys divided into two groups 14205 and 14206 in this example) is included in the apparatus. For example, as described above, almost all characters of a language are assigned to four keys 14401, 14202, 14203, 14204 (e.g., each of 14211, 14212, 14213, 14214 of the keypad model 14210). Each of the four keys unambiguously represents any character of the group of characters assigned to that (e.g., corresponding) key (e.g., assignment to some keys, such as four keys of characters of a language, and the keys of the invention Its use as a predictive data entry system has already been described in detail). In this example, an on-screen keyboard (e.g. divided into two parts 14207, 14208) that provides correct characters (e.g. as described above) in addition to some mechanical keys is applicable to the (e.g. predictive) data input system of the present invention. Do. In one example, the on-screen keyboard can be used to correctly enter one or more characters of a word while entering a word through the predictive data entry systems of the present invention. For example, in order to enter the word "day", the user can first press the key 14209 of the on-screen keyboard to correctly enter the letter "d". The user can then press the opaque keys 14202 and 14203 corresponding to "a" and "y". Based on the information provided, the system provides the word "day" with the highest priority among the words corresponding to the input information (e.g., pressing a chain of indeterminate keys 14204, 14202, 14203 in this example). May correspond to the word "its", which has a higher priority than the word "."

It should be noted that the examples shown above are used for illustrative purposes only. As mentioned above, it is understood that other input means (eg having different forms) may be used to input the correct symbols (eg characters) or functions.

As described in the references contained herein submitted by the inventor, an external (eg mobile) keypad (eg mechanical, virtual, etc.) with some keys and some other components such as a microphone, camera, mouse, etc. It can be manufactured for use with the data entry systems of the.

According to one embodiment of the invention, the external keypad may be a removable keypad designed to be attached to or accommodated in the device (eg, when not in use). For example, the keypad may have a flat shape to be mounted to an external surface of the device, or may be designed to have a PCMCIA card shape to slide in the PCMCIA slot of the device. The keypad may be connected to the device by wireless or wired.

According to one embodiment, the external keypad may comprise a rechargeable battery such that when the keypad is mounted to the device (eg, slides in a PCMCIA slot, fits in a USB port, etc.) a battery can be charged through the device. have.

It is understood that the keypad may have any other form, may be attached to the device by any other means, and may be charged using any other means. For example, the keypad may have a credit card size.

As mentioned above, words in which at least part of them exist in the system's word dictionary can be entered via the combined data entry systems of the present invention. According to one embodiment of the invention, in order to enter a word, a user may input a predefined discrete input signal (such as a predefined interaction with one or more predefined key (s)) in two consecutive words. End-of-the-entry-of-the-previous of the word provided between the input of the corresponding information (e.g., key presses and, if provided, vocalization) -portion) and / or beginning-of-the-entry-of-the-following-portion can be informed to the system. In this case, the input corresponding to the subsequent part of the word (e.g. key presses and possibly vocalization if necessary / if desired) information (e.g., eventually including a search procedure if the word prediction system of the present invention is used) In the course of, the system does not consider the input information (eg key presses and utterances) corresponding to the portion (s) preceding the subsequent portion of the word.

A word may combine one or more parts, where each part is an entry in a system's word dictionary (eg, a word, an item such as a word-part) and / or is not an entry in a system's word dictionary (eg, not included in the dictionary). ) One or more parts. It should be noted that the word can be arbitrarily divided into several parts by the user.

As an example for one of the possibilities of compound words described above, when entering a word (eg the word "pcboard", which can be divided into two parts "pc" and "board"), the input of a chain of single symbols (eg In the example, after the data input system, one by one of the symbols of the present invention for inputting the letters "p" and "c", the user inputs a separate input signal so that the user enters a new part of the word (e.g., predictive data input of the present invention). In this example input by the system, when entering the part / word "board", it is possible to prevent the system from considering the preceding part (s) (e.g., in this example, a chain of single symbols) within the search procedure. It is understood that. A compound word (eg "handheld") may consist of at least two consecutive parts (eg words / parts) (eg "hand" and "held") that may be included in the system's word dictionary. In this case, for example, both parts may be input by the predictive data input system of the present invention, where a separate input signal may be provided after input of the first of the two parts.

As described above, when using the predictive data entry system of the present invention for entering words, according to a first search-and-select procedure, user input information (e.g., key presses and For each moment of input information (e.g., each successive key press and, if provided, the corresponding voice) during the speech, if provided, the system is such that the entire chain symbol of each word corresponds to the input information provided by then. A word having the highest priority among the words of the word dictionary in may be provided (eg, shown to the user). For example, by considering the keypad 14300 of FIG. 19A, if the user wishes to enter the word “forget” (eg, without providing vocalization in this example), he may use keys 14302, 14303, 14302, 14303, 14301,. 14302) can be pressed. Table 14309 shows, by way of example, a word database of the system. Table 14305 shows, for example, successive key presses (eg, 1 ^st key press, 1 ^st and 2 ^nd key press, 1 ^st and 2 ^nd and 3 ^rd keys) provided without utterance during input of the word “forget”. Six words 14306 of the system's word database corresponding to the six stages of input information consisting of the provision of pressing, etc.). This word search, selection, and presentation method has already been described in detail.

As mentioned above, according to one method, at the moment of each input information, if all of the symbols correspond to the user's input information (e.g. key presses and vocalization if provided), the word prediction of the present invention If the system does not find, the system can find longer words in the database whose initials correspond to the input information and generally those initials have the highest priority (e.g. at least) of the input information. Note that you can present the initial symbols. If the system does not find a longer word whose initials correspond to the input information, the system may define, for example, a predefined symbol (e.g., the last key press, if provided) , Corresponding to utterance, may be added to the end of the last chain symbols (e.g. or word) provided by the user (e.g., corresponding to a previous step of input information provided for entry of a word).

Further, as mentioned above, according to the second search-and-selection procedure, in order to reduce the fluctuation of the display unit during the provision of the user's input information (eg, key presses and, if provided, voice), the input information For each step of (e.g., each successive key press and, if provided, utterance), the system can determine that at least the first letters of the words are input information (e.g., input information corresponding to the word provided so far) (e.g., the step of the input information). Among the words of the system's database corresponding to at least the highest priority (eg corresponding) initials of the word (eg if so designed, the system may also show the remaining symbols of the word). To separate the remaining symbols from the system, the system may, for example, place them behind a cursor or other By printing can be provided to distinguish between them it may be printed) (which may e.g. show to the user). For example, considering the keypad 14300 of FIG. 19A, if the user wishes to enter the word "forget" (eg, in this example, without the provision of vocalization), he may have keys 14302, 14303, 14302, 14303,. 14301, 14302 may be pressed. The table 14307 shows, by way of example, that each of the words corresponds to a corresponding corresponding step of the input information, at least the initials of the words being provided during the entry of the word "forget" (eg in this example a chain key provided without speech) Presses) consecutive key presses (eg 1 ^st key press, 1 ^st and 2 ^nd key presses) during the entry of the word "forget" with the highest priority (eg the highest frequency of use) among the words of the system's word database, Six chain symbols 14308 corresponding to six levels of input information consisting of providing 1 ^st and 2 ^nd and 3 ^rd key presses, etc.) (eg, in this example, without speech). According to this procedure, for example, when the user presses the first key 14302, the system is the highest among the words in the system's word database 14309 starting with the letter corresponding to the first key press provided by the user. The letter "a" of the word "and" having a frequency of use can be displayed. The user can then press the second key 14303. The system includes at least the beginnings of the word " to " having the highest frequency of use among the words in the system's word database 14309 starting with the letters corresponding to the first and second key presses provided by the user. For example, in this example, the chain letters "to" corresponding to all symbols) may be presented. The user can then press the second key 14302. The system has at least the beginning of the word " for " having the highest frequency of use among the words in the system's word database 14309 starting with the letters corresponding to the first and second and third key presses provided by the user. It is possible to suggest the chained letters "for" corresponding to the symbols (eg in this example all the symbols). The user can then press the fourth key 14302. The system appends the letter "g" to the chain letter "for" that is already displayed (e.g. the display is not shaken) so that most of the words in the system's word database start with the letters corresponding to the key presses provided by the user. It is possible to provide the chained letters "forg" corresponding to the initial signs of the word "forget" with a high frequency of use. The user can then press the fifth key 14401. The system appends the letter "e" to the already displayed chained letter "forg" (for example, the display is not shaken) and is the highest among the words in the system's word database starting with the letters corresponding to the key presses provided by the user. It is possible to provide the chain letters "forge" corresponding to the initial signs of the word "forget" with the frequency of use. The user can then press the sixth key 14302. The system appends the letter "t" to the already displayed chained letter "forge" (for example, the display is not shaken) so that the word with the highest frequency of use among the words in the system's word database corresponding to key presses provided by the user. You can provide "forget".

As shown, the display unit compares the second search-and-selection procedure by comparing the proposed chain symbols 14306 and 14308 provided during the entry of the word "forget" by each of the search-and-selection procedures. Note that you may have less fluctuations if you use them.

According to one embodiment of the invention, the first and second procedures can be combined to provide an efficient third procedure that can provide the user with search results of both procedures. At each point in the input information for entering a word, the word with the highest priority among those words whose all symbols of the word correspond to the input information provided by the user up to that point and the first symbols of the word are presented to the user up to that point in time. It is easier for the user to show the highest priority word among those words corresponding to the input information provided by the user (e.g., as described above, if the first procedure finds the corresponding word). If not, this may help to provide for printing / suggesting the corresponding initial symbols of the word proposed by the second procedure.

As an example, FIG. 19B shows a database of words 14319 of the present invention. By considering the keypad 14309 of FIG. 19A, based on a first procedure (eg, as shown in step 4), to enter the word “informative” (eg, in this example, by providing key presses without speech) After pressing the keys corresponding to the first four symbols of the word, the system inserts the four symbols into line 14315 of text on screen 14310 to select the first word (eg, word “into”) 14311. It is possible to propose, and preferably at the same time based on the second procedure, the system may, for example, place a second word (eg, word “information”) 14312 below the first word (eg “into”). By printing, a second word can also be proposed as an alternative. In addition to the cursor 14314 located at the end of the first word 14311, the cursor 14313 may be automatically placed at the corresponding position of the second word 14312. To continue, the user may have other choices, such as using one of the following methods.

1) Press the key corresponding to the next symbol (eg "r") of the desired word (eg "informative"). or

2) provide a predefined input signal, such as a predefined pressing action on a predefined key, to replace the word “into” in line 14153 of the text with the word “information”. or

3) the first N + 1 symbols of the second word (eg, "information") proposed by the second procedure with a first word (eg "into") with N symbols proposed by the first procedure Provide a predefined input signal, such as a predefined pressing action on a predefined key to replace with (eg, “infor”). This makes it possible to enter at least the correct initial letters of the desired word, without having to press the keys corresponding to the letters. Obviously, each additional predefined input signal (eg, a predefined key press operation) may add to the N + 1 symbols the same symbol as the next symbol / character of the second word. or

4) defined to replace the first word (eg "into") with the longest chain first symbols (eg "informati") of the second word (eg "information") proposed by the second procedure. Provides a predefined input signal, such as a predefined pressing action on a defined key. Wherein the chain symbols are the first symbols of all words in the database of the present invention corresponding to the same chain key presses corresponding to the chain symbols.

In this example, for example using one of the methods 3, or 4 mentioned above, the user can replace the word "into" with the chain symbols "informati" (see, eg, step 5). The user may then press a key 14304 corresponding to the letter “v”, for example (see step 6). Based on the word prediction system of the present invention, the system may add the letter "v" to the chain symbols "informati" to provide "informativ" as the first choice. And of course a combined input signal providing the first symbols of the word proposed as the first selection corresponding to the word "informative" (e.g., in this example, the two words "informative" and "informatively") as the second selection by the second procedure. , Where the word "informative" has the highest priority. The user can then press the key 14301 corresponding to the letter "e". Based on the word prediction system of the present invention, the system may provide the word "informative" as the first suggested word by adding the letter "e" to the initial chain symbols "informativ". At this stage, the system may, by a second procedure, combine the word “informatively” (eg, the combined input signals providing the word “informative” as the first choice may correspond to the first symbols of the single word “informatively”). Can also be provided. The user may provide a word-ending signal, such as providing a spacing symbol to terminate the input of the desired word, "informative".

In order to have better accuracy and still fewer display fluctuations while entering a word, the user can also provide corresponding speech information based on the principles (eg, press-and-speak) of the data entry systems of the present invention. . By way of example and for comparison purposes, FIG. 19 illustrates the steps of entering the word “informative” by providing input information (eg, key presses) without speech (eg, as described above). 19D shows steps for entering the same word, “informative” by providing input information (eg, key presses). Here, in step 3, in addition to the key press corresponding to the letter "f", the user speaks the letter. In view of the input information, the system may provide the word “udf” by the first procedure and also in this step (eg, in this example, in comparison with the input of the same word in the example of FIG. 19C), the word information. May be proposed in the initial stage by the second procedure), the word "information" by the second procedure.

As mentioned above, in the case of using the predictive data input systems of the present invention, a user can correct a word while inputting a word-end signal such as a space symbol. For this purpose, for example in the case of pressing the space key, the user can speak at least one of the symbols (eg, letters) of the word. The at least one spoken symbol may be any one of the symbols of the word. For example, in order to enter the words "this way" using the keypad 14300 of FIG. 19A, the user first enters the keys 14302 (eg, corresponding to each of the letters of the word "this") without speech. 14301, 14304, 14303 may be pressed. The system may not detect any speech while entering a word (eg, providing key presses corresponding to letters). Thus, the system can provide the highest priority word (eg, "this") of the words in the database that correspond only to key presses (eg, provided without speech). At the end of the word entry, the user can press the "space" key without a utterance to provide a space symbol after the word. The user can then proceed with the input of the word "way", for example, by pressing the keys corresponding to the letters "wa" without the utterance, and then pressing the key corresponding to the letter "y". I can press and speak. The key presses and speech information provided by the user may correspond to at least two words "may" and "way" in the system's word database. The system may output the word “may” having the highest priority among words corresponding to the user's input information (eg, key presses and utterances provided by the user for input of the word). Different modification methods have been described in detail above. According to one method of the present invention, at the end of the input of a word, if the word provided by the system is not the word desired by the user, a predefined signal such as pressing the " space " During providing, the user may speak (eg, at least) one of the letters of the word as additional input information corresponding to the word. By considering the already provided (eg, preferably all) input information and additional information, the system can provide a new search process to provide new results. In this example, while pressing the "space" key (eg, preferably), the user may speak the letter "w". In view of the phonation of the information (e.g., the resulting word "may") already provided and the letter "w" (e.g., corresponding to key 14401), the system (e.g., the letter "m" of the word "may") Letter "w" may be provided (eg, in the present example, since the letter "y" has already been correctly entered by pressing the corresponding key and uttering the letter "y"), the letter "w" "y" cannot be replaced). In this example, with modifications, the system may provide a space symbol after the word. In another example, to enter the word "no", if the user presses the keys 14304, 14303 without utterance, the system may provide the word "is" that corresponds to the input information and has the highest corresponding priority. have. The user can then press the "space" key and utter one of the letters of the desired word "no" (eg, "n"). The system may provide the word "no" corresponding to the binding information provided by the user (eg, initial key presses, and additional speech). In another example, to enter the word "ok", the user may provide key presses 14303 and 14303 without utterance. The system may provide the word "so" with the highest priority. The user can then press the "space" key and utter the letter "k", for example. In this example, the character corresponds to two key presses (eg, both key presses). The system says that words in the dictionary corresponding to key presses and at least one of the letters of each of the words includes a letter “k” (eg, at a location corresponding to the key press corresponding to the letter “k”). In consideration of this, it is possible to provide the word having the highest priority. In this example, the word "ok" may be provided.

It should be understood that the modification procedures of the present invention can be performed during the entry of a word (eg, in the middle). For example, if during the entry of a word (eg, intermediate), the user provides speech information (eg, with / without a correction signal, as defined previously), and the speech information provides speech information. The predefined key press information provided during the previous word entry can be reached. According to one method, it should be noted that speech information may be predefined to correspond to key presses that may be provided after providing the speech information.

Also, in many cases, non-voice key presses may be sufficient for input of the desired word, and according to the principles of the word prediction data entry systems of the present invention, the screen of the device may suggest different words during key presses. Care must be taken. In order to reduce or sometimes eliminate screen fluctuations (e.g., especially for long words), during input of key presses corresponding to a word, the user is the most (e.g., preferably first) symbol / character of the word A small portion of the speech information can be provided to accurately enter the symbols. This can significantly reduce the fluctuation of the screen. For example, using the keypad 14300 of FIG. 19A, while entering the word “for” only by key presses, the system may suggest chain symbols “a”, “to”, “for”. Can be. For example, if the user also utters the letter "f" during the first push operation, the system suggests the chain symbols "f", "fo", "for" (e.g., screen shake in this example). none).

As mentioned above, during word entry through a word prediction data entry system (e.g., of the present invention), at the stage of the input information (e.g., a key press operation and, if necessary, vocal information), the system displays the first symbols of the user. It is possible to suggest a word / chain-symbols different from the word / chain-symbols provided in the preceding step of the input information corresponding to the word provided by (eg the display is shaken). According to an embodiment of the present invention, if the user wants to maintain previous chain symbols and adds additional symbol (s) to the chain symbols to provide the desired word, the user has information corresponding to the previous chain symbols. After inputting, a predefined input signal may be provided. For example, taking into account the keypad 14300 of FIG. 19A, in order to enter the word "then", if the user provides single press operations on the corresponding keys without utterance, the system is less than the word "then". It is possible to provide the word "been" with a high priority. The first three key presses provided by the user may provide the word "the". In order to inform the system to maintain the chain symbols (eg, "the"), the user must provide a user with a key on the predefined key before providing input information corresponding to the next symbol (s) (eg "n"). Provide predefined interactions (such as, for example, double pressing or sliding operations). The user may then provide input information (eg, press key 14304) corresponding to the text. The prediction system of the present invention comprises a chain symbol "the" (e.g., not input information corresponding to chain symbols) and input information corresponding to the letter "n" (e.g., in this example, the prediction system is a chain character). Find the word with the highest priority among the words starting with "the" and ending with one of the letters "iucdnvj", giving the word "then"). The user may then provide a word-end signal, such as a space symbol, to terminate the entry of the word.

As mentioned in detail above, input information corresponding to one symbol of a word may also include information about previous (eg, and / or next) symbols of the word. For example, a single press operation on key 14401 may inform the word prediction system that the system should consider one of the letters "ehmqwz" in the previous procedure for entry of the current word. On the other hand, other interactions on the key 14401, such as in one example (eg, a double press operation, a longer press operation (eg, a press and hold operation on the key for a predefined time), and the key Pressing different keys simultaneously, etc. may, for example, notify the word prediction system that the current input information belongs to the next word (e.g., it can of course also notify the system that the input of the previous word has ended). has exist). As a continuing reference to the previous embodiment and in view of the principles just described, an input signal informing the system that the previous chain symbols (e.g. "the") is retained is a word (e.g. "then"). It may be provided while providing input information corresponding to the next symbol (e.g., the letter "n" in this example). For the same purpose as above, according to another method, a predefined utterance input (eg, utterance of "keep", "end", etc.) may be provided during / with the key interaction corresponding to the following symbol. For example, in addition to a single press action on a key corresponding to the letter "n", the user also utters "keep" to give the chain symbol "the" the additional information just entered (eg, corresponding to the letter "n"). The system can be considered as the first symbols of the desired word combined with a single press on a key).

Including an input signal corresponding to previous or next symbols in input information provided for input of a word can confuse a user. Therefore, the input signals may preferably be provided separately (eg, as described in detail).

23 illustrates keypad 14700 of the present invention as an example. Here, substantially all the symbols of the PC keywalk (eg, letters, punctuations, instructions, functions) are assigned to the keys of the keypad 14700. The keypad is based on at least some of the principles of the predictive data input systems of the present invention. For example, English characters are assigned to four keys 14701, 14702, 14703, 14704 of the keypad 14700, thus:

Letters with ambiguity-like utterances are assigned to different keys apart from each other. And / or;

In the case of entering a word using the word prediction methods of the present invention based on key presses (e.g., and, if necessary / provided, if desired), the words most often used are in different chain key presses. Correspondingly, the target word in most cases (eg, while entering text) does not require additional input information (eg, utterance for the input of correct symbols, handwriting, etc.) and the characters of the word. The characters are assigned to different keys apart from one another so that the keys corresponding to the symbols can be entered by pressing / interacting with the keys. Also, preferably, the number of words corresponding to the same chain key presses is assigned to be minimized.

When entering text, the characters are assigned to different keys apart from one another so that the user can interact with the four keys as evenly as possible. And / or;

If the system also uses handwriting for the input of at least some of the letters of a word to improve the accuracy of handwriting (eg, by adding / speaking / speaking information), the system assigns a letter to it. Handwriting on the corresponding sensitive key enables the entry of correct characters / symbols. In this case, characters with obscure writing graphs can be assigned to different keys.

Character assignments for keys of the keypad based on the example keypad 14700 and at least one of the described principles are shown by way of example, and other character assignments for keys of the keypad based on at least one of the principles may be considered. . For example, keypad 13200 of FIG. 8 (eg, partially shown. The keypad may have more keys as described in many embodiments, where Roman / English alphabet letters are assigned). Keys shown) are also constructed based on the principles described above.

Based on the principles of the predictive data input systems of the present invention, with reference to one of the keys 14702 as an example, the system identifies symbols "abfrtx_" 147021 (eg, letters are lowercase) of the present invention. Word prediction systems can be thought of as part of the elements that can make up a word. The input information corresponding to any of the symbols may be the provision of a predefined first type of interaction with the key (eg, a single press on the key). Providing a predefined first type interaction (e.g., with the key) without providing a corresponding utterance may unambiguously correspond to one of the symbols (e.g., characters), and (e.g., the symbol) Providing the predefined interaction (eg with the key) in combination with the provision of a predefined utterance corresponding to one of these can accurately provide the preference.

In this example, the predefined second type interaction with the key 14702 (eg, a double press operation, a press-and-hold operation for a (at least) predefined time, etc.) is any one of opaque characters. For capital letters 147022, the provision of the predefined interaction, combined with the provision of utterances corresponding to one of the letters, may correspond to exactly the letters in capital letters.

Continuing with reference to keypad 14700, based on the principles of the predictive data entry systems of the present invention, for example, with reference to key 14702, a predefined second type interaction with key 14702 (eg, The provision of a single slide operation, the press of a modification key 14707 and the preferably simultaneous provision of a single press operation, etc. may be accomplished through the word prediction systems of the present invention (eg, as mentioned above). -of-a-word) may correspond to a symbol that may not be considered by the system as part of the members that may constitute a symbol (e.g., the symbol "" 147023). Providing a third type of interaction (eg, as defined with / without vocalization) does not provide for the symbolic data input system of the present invention without going through the word prediction procedures of the present invention (eg, such as search and selection methods). By the symbol by the corresponding symbol may be provided.

In this example, a predefined fourth type of interaction with the key 14702 (eg, dual slide operation, press the change key 14707 and preferably provide simultaneous double press action, press the change key 14707 and preferably Simultaneous provision of a press-and-hold operation for at least a predefined time period, etc.) may not be considered by the system as part of members capable of constructing a word through the word prediction systems of the present invention (eg, other ) Symbol (for example, symbol "_" 147024 in the present example).

The principles corresponding to the key 14702 as described can be applied to the keys of any keypad used with the combined data entry systems of the present invention, where the keypad has a reduced number of keys, wherein the At least some of the keys each represent several letters of the language. In this example, the principles corresponding to key 14072 as described can also be applied to keys 14701, 14703, 14704 (eg, for input of corresponding symbols).

Regarding the key 14706, in this example, the predefined first type interaction (eg, single push operation) with the key 14706 may correspond to providing a spacing symbol. The predefined second type of interaction with the key (eg, dual press operation, etc.) may correspond to an “enter” command (eg, a PC keyboard).

Providing a predefined third type of interaction with the key 17606 (eg, simultaneously providing a single slide action, pressing the change key 14707 and preferably providing a single press action) (eg, as described in detail above) Likewise) the prediction methods of the present invention may correspond to maintaining the input of a word, the proposed first chain symbol of the previous case of the input information. A predefined fourth type of interaction with the key 14706 (e.g., dual slide operation, pressing the change key 14707 and simultaneously providing a double pressing operation, pressing the change key 14707 and preferably at least a predefined Simultaneous provision of a press-and-hold operation over time) may correspond to a "tap" function.

Principles corresponding to key 14706 as described may also be applied to key 14705. For example, the provision of a predefined first type interaction with key 14705 without utterance may correspond to the provision of a "BkSp" instruction. Also for example, the provision of a predefined third type interaction with key 14705 without utterance may correspond to "SupBk" (eg, a super / special back space command, pattern Delete one or more symbols as described in the references submitted by the inventor and incorporated herein by reference).

Each of the keys 14707 and 14708 may at least function as a change (eg, “shift”) key.

With regard to the key 14707, in this example, the provision of the predefined first type interaction with the key without speech may correspond to the provision of the symbol ",". To enter any of the symbols "() @ # '$ &", the user can provide a predefined first type interaction with the key 14707 and utter the symbol.

Providing a predefined second type of interaction with key 14707 without utterance may correspond to a "shift" function. To input any one of the combined chain symbols "s", "d", "m", "ll" (eg, chain symbols comprising an apostrophe symbol as described in detail above) May provide a predefined second type interaction with the key 14707 and provide speech information corresponding to the concatenated chain symbols (e.g., according to another method of symbol assignment, each of the chain symbols). May be assigned to the key representing the first symbol of the chain symbols).

With continued reference to keypad 14700 and based on the principles of the predictive data entry systems of the present invention, for example, referring to key 14707, a predefined third type interaction with key 14707 without utterance Providing may correspond to the "CapsLocK" function. In this example, the provision of the predefined fourth type of interaction with the key 14707 without utterance may correspond to a symbol / symbol (eg, symbol “(”) in this example).

The principles corresponding to the key 14706 as described may be applied to the key 14708 described for the key 14707 (eg, key interactions, and utterance corresponding to the case provided). For example, providing a predefined first type of interaction with key 14708 without utterance may correspond to providing a symbol ".". To enter any of the (eg, arithmetic) symbols "<> 0 9 +-/ * =% #", the user can provide a predefined first type interaction with key 14707 and the symbol You can speak.

The provision of a predefined second type interaction with the key 14708 without utterance is directed to a "EOW" function (eg, a word-end signal generally used to provide compound words as described above). This may be the case. In order to input any one of the chain numbers (eg, "00" "000" "000,000", each of which has a corresponding utterance among "Hundred", "Thousand", "Million"), the user may interact with the key 14708. It may provide a predefined second type interaction, and may provide speech information corresponding to the chain numbers.

Continuing with reference to keypad 14700, based on the principles of the predictive data entry systems of the present invention, for example, with reference to key 14708, a predefined third type interaction with key 14708 without utterance (Eg, where the modifier key used with the key 14708 can be the key 14707), the provision of the " modify " key (eg, as described above, the predictive data input systems of the present invention). A modification key indicator indicating the start of the modification / selection procedure for correcting the words proposed by In order to provide one of the functions of F1 to F12, the user can provide a predefined third type interaction, and the predefined speech information corresponding to the function (eg, the number of the function, for example) Vocalization).

In this example, the provision of a predefined fourth type of interaction with the key 14708 (e.g., where the change key used with the key 14708 can be the key 14707) without utterance is a symbol / symbol ( For example, in the present example, the symbol ")" may correspond. In order to enter the combined digit-and-letter chain symbols (e.g., 0 ^th , 1 ^st , 2 ^nd , 3 ^rd , 4 ^th , ....., 9 ^th ), the user is assigned a fourth defined type Speech information that can provide interaction and correspond to the concatenated chain symbols (eg, "zero", "first", "second", "third", "fall", ....., "nineth" Speech can be provided.

It should be noted that the assignment of symbols and the different types of interactions for the keys are used only as an example. Any one of the symbols may be assigned to a (eg key) interaction different from the (eg key) assigned interaction in this example. Other (eg key) interactions may also be considered for the keypad.

As noted, some predefined forms of interactions with each key of the keypad in the absence of utterance or in the presence of the corresponding utterance (eg, different) corresponding symbols (eg, letters, chain symbols, punctuation symbols) , Functions, etc.). By taking this principle and other variables into account, such as using a different number of keys, many types of keypads can be provided for use in the present invention (e.g., used as a data input prediction system. For example, at least one PC keyboard). Functions (e.g., Ctrl, Alt, Del) can be assigned to the key 14707 and provide a corresponding interaction (e.g., in this example, a fourth predefined form of interaction with the key 14707). And by providing corresponding predefined speech information assigned to the keyboard function), any function may be entered.

As mentioned above, the data entry system of the present invention may be used as a circular touch sensitive surface such as that used in an iPod (iPOD) player to navigate and / or select content. In addition to the sliding action on a circular surface (eg to select content from a list), the circular surface is used to reproduce (eg, reproduce) the keypad used with the data entry system of the present invention for entering data such as text. May respond to tapping motions (at predefined locations on the surface). 20 illustrates a circular touch sensitive surface as described by way of example. The circular surface may have several predefined areas, each of which may correspond to a key of a predefined (eg imaginary) keypad of the system. In this example, areas 14401-14406 may reproduce at least some of the principles of the keypad of the present invention as described in the references filed by the inventor and incorporated herein. For example, four keys 14401,14402, 14403, and 14404 may be allocated and allocated substantially all characters of the language.

In accordance with one embodiment of the present invention, the keypad may be used to input text, and a sliding action on a circular surface (eg, circular) may be used to select an object. Can be used as a combined data / text input system. For example, to enter the word "bull", in accordance with one embodiment of the present invention, the user may not be able to speak (eg, the prediction system of the present invention may be based only on key presses, or the user may Etc. may not want to speak) may press the corresponding keys (14402, 14404, 14403 and 14403). The system may show the user a list of words 14212 that corresponds to the input information (eg, the frequency of the usage numbers is indicative). At this time, the user provides a (eg circular) slide action (14410/14411) on the circular surface to navigate the list to select the desired word. For example, after selecting a desired word, the user may provide an operation of tapping the central area 14407 of the circular keypad to select / enter the word. According to another example, in addition to the key press, for example, while pressing the first key 14402, the user may speak the corresponding letter "b". In this case, the system may show the list 14413 (eg, directly indicate the frequency of the usage numbers). At this time, the user may provide a sliding action (14410/14411) to the circular surface to navigate the list to select the desired word.

The described data entry clause can be applied for content retrieval and selection. For example, by using a tapping action (eg, by providing a corresponding utterance), a user can enter text content information corresponding to a partial or full symbol of the content (eg, the title of the content), (eg, After providing the list of contents (e.g., by printing them on the screen) corresponding to the user's input information), the user can use the slide action on the surface to select the desired content.

It is clear that the circular keypad provided in this embodiment is provided only as an example. It is apparent that other types of keypads may be considered by those skilled in the art. For example, as shown in FIG. 20A, a circular keypad 14410 that includes eight predefined areas resembling the eight keys used in the data entry system of the present invention may be considered.

As shown in FIG. 21, according to one embodiment of the present invention, an external keypad 14500 that can be used as the data entry system of the present invention is manufactured separately to form at least two portions 1501 and 14502. Each part may be independently / separated and attached to the corresponding electronic device 14509 (eg, each part is preferably independent, preferably wirelessly connected) to the data input system of the present invention. Is used. The separate portions are designed to be attached to each other to form an intensive keypad 14500 and to be used as an external keypad in an electronic device (and, for example, to be used with the data entry system of the present invention). In this example (eg, by way of example), all characters of the language may be assigned to four keys (eg, dark colored keys in the figure). The integrated keypad is used in the data input system of the present invention including at least one microphone. For example, each portion of the separated keypad includes an integrated microphone so that when the keypad is located separately on the sides of the electronic device, the microphones 14503 and 14504 can also better capture the user's voice. Distributed on the side of the electronic device.

Telephone keypads have become a familiar dialing means for telephones. Therefore, the keypad can be included in most voice communication devices. In accordance with one embodiment of the present invention, a voice and / or data communication device capable of utilizing the system's data input (e.g., including data input prediction system) system may be a telephone (e.g., for data / text input). Keypads (eg for dialing) and (eg, separate) keypads of the present invention. As is known, the telephone keypad occupies a significant portion of that side of the telephone, reducing the surface to be assigned to the display on that side. To solve this problem, the keypad can be manufactured to extend from the device body (eg, preferably in the long axis direction of the device, as in FIG. 22B). 22 illustrates, by way of example, a voice and / or data communication device 14600 comprising a (eg, combined) data entry system of the present invention using a keypad separated into two parts 14601 and 14602. The device may also be a keypad in the form of a telephone (eg, in a retracted position, not shown here). When the telephone keypad is in the retracted position or when the user enters data such as text by using a separate keypad through the data entry system of the present invention, the display unit 14608 (eg, automatically, automatically) is displayed on the screen. The text 14609 to be output may be in a landscape position to better view the text 14609. The apparatus may also include a point 14613 and a selection 14414 means resembling a computer mouse function (eg, in conjunction with the data entry system of the present invention). FIG. 22B shows the same device 14600 with telephone keypad portion 14601 extending outward from the device body (eg, the device of FIG. 22A). When the keypad is in an extended position, the display portion 14608 of the device is switched to a direction (e.g., automatically, a portrait posture automatically), which is indicated by the telephone keypad 14611 and the telephone number 14641. Make sure to match the phone keypad to show the input information such as).

It should be noted again that the lip reading system (in addition to or in addition to the speech recognition system can be used as the data entry system of the present invention to recognize the vocalization of the user). Because it requires the utterance of mostly separated characters, it can be precisely matched with the data input prediction system of the present invention, which makes the procedure of inputting data even more individually.

As mentioned and shown in the references submitted by the inventors and incorporated herein, the keypads used as data input systems (e.g., including data input prediction systems) of the present invention (e.g., different forms May comprise different key configurations, and obviously, the keys of the keypad may have different shapes. Also, as mentioned, for example, the keys may be separated into two groups and located on opposite sides on the surface of the electronic device. For example, the keys 14701 to 14708 (and, for example, their correlations) of the keypad 14700 of FIG. 23 are shown in FIG. 23A, in this example 14711 to 14718, respectively, of FIG. 23. Keys corresponding to keys 14701 to 14708 may be reproduced / manufactured in an ellipse (or, for example, a circle). The keys may be separated into two key groups 147111 and 147112, which are attached / integrated on opposite sides on the surface of the electronic device 14719. The configuration of the keys of each group in relation to each other can resemble the configuration of the keys of the directional keypad (eg, of an electronic game console). In this example, in each group, two keys are located in the longitudinal direction, the third key of the group is located between the right side of the two keys, and the fourth key of the group is located between the right side of the two keys. Preferably, the mobile data entry system allows the user to enter data preferably while on the move. To grasp the mobile device, the user needs to use several fingers. Therefore, ideally the user's thumb is used for data entry. The configuration of the keys described herein allows the letter keys to be placed under the user's thumb, so that the user's finger does not have to move over the key while writing a word. By using the user's thumb, this configuration allows for quick entry of text characters.

The use of input means such as an on-screen keypad / keyboard for data entry through the data entry system of the present invention is described in detail in the references submitted by the inventor and incorporated herein. Also, the user's finger, finger cap, which resembles the interaction of the keys and keys of the keypad (e.g., with eight keys, several keys, such as keypad 14700 in Figure 23) used with the data entry system of the present invention. The use of other means, such as gloves, gloves, stylus, etc., is submitted in detail by the inventors and is described in detail in other references incorporated herein (e.g., obviously, the input means is a data input of the present invention). May be used as a prediction system). For example, as shown in FIG. 24A and described above, a dynamic keypad including at least four (eg, eight to eight) keys may be provided on a surface by a user's finger interaction (eg, where , At least four letters of the language are distributed and assigned based on the principles described in the references submitted by the inventor and incorporated herein). The principles of defining the keys of the dynamic keypad and the interaction between them are described in detail in the references submitted by the inventor and incorporated (eg, schematically in the present patent application). As mentioned above, the dynamic keypad can use the data entry system (including the data entry prediction system) of the present invention. For example, by finger interaction on a sensitive surface, such as a (eg, touch) screen of a tablet PC, using at least one finger, such as forefinger 14901 (or, for example, a stylus), as described above, The user can reproduce at least the keys of the keypad 14700 of FIG. 23 and the interaction between the keys (eg, single press operation, double press operation, slide operation, etc.). Pressing and holding a key (eg a shift key) and other keys (eg, simultaneously) may also be provided. For example, a user may provide an action of pressing and holding the screen using thumb 14803 (eg, located on the left side of index finger 14401), while at the same time, for example, Forefinger 14901 may provide a corresponding interaction (eg, single push action, double push action) with respect to the key. Each combined interaction may be assigned a corresponding symbol group of the present invention. Further, for example, a user may press and hold the screen using a middle finger (eg, placed on the right side of the index finger) while simultaneously using the index finger 14001 to press the desired key of the dynamic keypad, for example, Press action, double press action, etc.) to provide the corresponding interaction. Each of the combined interactions may be assigned symbols of the corresponding group of the present invention. It should be noted that the system may react differently to each pressing and holding interaction (eg, to the left or to the right of the actions of pressing the dynamic keypad). It should be noted that the fingers of the user reproducing the predefined keypad shown and described herein are used as an example. Obviously, based on the principles of the present invention, the user can use different fingers to provide keypads or other keypads.

Each finger can reproduce the keys of one or more (eg, preferably two) keypads of the present invention and utilize a (eg integrated) touch sensitive system / surface that utilizes them with the data entry system of the present invention. Means and methods consisting of providing interactions such as a push (eg, tapping) action (and / or sliding actions) by a portion of the finger capped finger (eg, tip, flat portion, etc.) having (E.g., as described in detail in this patent application, for example, tapping once or twice with the tip of a finger / finger cap may be performed once or once with a defined first key of a predefined keypad model. The two-press action can be reproduced or corresponded to each other, and, for example, tapping once or twice with a flat portion of the finger / finger cap (e.g., above) can be a predefined key. One or two presses of the second predefined key of the de-model may correspond to, or correspond to, respectively .. The means and methods of interaction may be touch sensitive, such as a touch screen of an electronic device for user input interaction. Reduce the use of panels). Obviously, the input means and method can be used as a data input system (including a data input prediction system) of the present invention. Also, as described in the previous patent applications, by considering the example of FIG. 24B, the present invention, such as the keypad 14700 of FIG. 23, by using four fingers 14111-14814, respectively, with the finger cap of the present invention The interaction between the keypad and the keypad keys can be reproduced and used as the data input prediction system of the present invention.

For example, using a flat portion of the finger cap 14911 to provide interaction, such as a predefined push action, is to interact with (eg, provide for interaction with) the key 14501 of the keypad 14700. It may correspond to). In addition, for example, providing an interaction, such as a predefined push action, using the tip of fingercap 14811 may interact with (eg, interact with) key 14703 of keypad 14700. To provide). Further, for example, providing interaction, such as a defined push action, with a flat portion of the finger cap 14812 may interact with (eg, provide for interaction with) the keys 14702 of the keypad 14700. It may correspond to). In addition, for example, providing interaction, such as a push action defined as the end of finger cap 14812, may interact with (eg, provide for interaction with) keys 14704 of keypad 14700. It may correspond to).

Also, for example, an interaction, such as a pressing action defined as a flat portion of the finger cap 14913, may interact with (eg, provide interaction) with the keys 14705 of the keypad 14700. This may be the case. Further, for example, providing interaction, such as a push action defined with the end of fingercap 14913, is to interact with (eg, provide for interaction with) the keys 14707 of the keypad 14700. It may correspond to). In addition, providing interaction, such as, for example, a pressing action defined as a flat portion of the fingercap 14414, may interact with (eg, provide for interaction with) the keys 14706 of the keypad 14700. It may correspond to). Finally, providing interaction, such as, for example, a pressing action defined as a flat portion of the fingercap 14414, is to interact with (eg, provide for interaction with) the keys 14708 of the keypad 14700. May correspond to

Pressing and holding a key (eg shift / change key) and pressing another key (eg at the same time) can also be reproduced. For example, a user may use the keypad 14700 of FIG. 23 to provide a press-hold operation with a flat portion of his finger (eg, apparently a fingercap) relative to a surface, and to provide a symbol "/" (eg, the keypad 14700 of FIG. 23). In order to reproduce the interaction required for the input of the symbol), without a word, for example, using the end of the finger cap (14811) can provide the desired pressing action.

For each of the different parts in contact with the sensitive surface for reproducing the keys of the keypad of the present invention, including several keys and used as the data entry system of the present invention, where the system is the size of the contact area of the surface. Recognize the portion based on an interaction based on interacting with different (eg, end, flat) portions of a user's finger with respect to a touch that the surface can sense to provide a different input signal. The method of is described in detail in the references submitted by the inventors and incorporated herein. The method of interaction can be used with the data input prediction systems of the present invention. Using an interactive system can reduce the need for a fingercap in the embodiments described above. For example, each user finger can reproduce two keys of the corresponding keypad model of the system. Based on the location of the contact impact of the user's finger on the sensitive surface relative to the location of the contact impact of another user's finger on the surface (eg, with respect to different vertical areas) (eg, during a series of data inputs, The typing / tapping operation of the surface with any part of the different user's fingers is typically provided within the same different vertical area on the surface), when typing text (eg, by using the principles of the above embodiments), By considering a finger (eg, end, flat) portion of the finger that can recognize a finger and where an interaction is provided, the system can associate the interaction to a corresponding key of a predefined keypad of the present invention.

As mentioned above, a handwriting recognition system may be used for the input of one or more correct (eg, unambiguous) word symbols, wherein at least one of the other symbols of the word is used to improve the recognition rate of the data input prediction system. One symbol is entered incorrectly. In the above examples (eg, as shown and described in FIGS. 20-24B), in order to reproduce key interactions when using hardkeys or softkeys, or when using fingers, fingercaps, gloves, or the like, In addition to tapping / sliding actions corresponding to key interactions, the user can write symbols, such as letters, on the corresponding sensitive surface, such as a touch screen. According to one method, the system can be designed to distinguish handwriting and tapping / sliding operations (eg, interacting / representing with a keypad). According to another method, the tapping motion may correspond to a key interaction, and the sliding motions may correspond to a handwriting motion. According to another method, the tapping / sliding action on the (eg soft) keys of the keypad of the system may correspond to key interactions and may be applied to the sensitive surface / touch screen (eg on any of the areas) of the keypad. A write operation (eg, tap, slide) may correspond to a handwriting input (eg, in the example of the media player of FIGS. 20-20A, region 14407 may be used for handwriting).

It is to be understood that the examples provided are for demonstration purposes only. Different fingers, different fingers with finger caps, different numbers of fingers or finger caps, stylus, gloves, etc. can be used as the combination (eg, data input prediction system) data input system of the present invention.

Although in different embodiments, different keypads, such as a keypad or a keypad in the form of a telephone comprising at least four keys to which four keys are assigned at least substantially all the characters of the language, may be used in different embodiments of the present invention. It should be noted that it has been used to demonstrate the principle, and obviously, at least some of the keys on the keypad are unclear, any kind of keys such as hard or soft keys, symbols assigned to the keys of the keypad and to any predefined interaction. Any other type of keypad including any number of keys, such as any configuration of these, may be used in embodiments of the combined data input system of the present invention.

Although in many paragraphs describing the combined data input systems of the present invention, "keys" have been used to describe the embodiments, they are used by predefined types of interaction to provide different predefined input signals. It will be appreciated that other input means may be considered to substitute for reproducing / replacing the keys and key interactions of the present invention. For example, interactions of use, such as finger or fingercap interactions, and corresponding interactions that reproduce the keys and key interactions described herein and in the references submitted by the inventor and incorporated herein It can be used in the above embodiments.

It will be noted that in some paragraphs the term “pressing” is an example of an interaction form with a key, which may be in other interaction forms such as contact / slide / pointing in relation to the key.

Although in many paragraphs a keypad (e.g., or some keys) has been used to demonstrate an embodiment of the combined data entry system of the present invention, the keypad may have some unclear keys (e.g., an unclear key previously described). Will be understood to include).

In order not to repeat the principles of the data entry systems of the present invention frequently, in many paragraphs of the present application, one or more symbols, such as symbols / words / word parts / functions, etc., may be assigned to a key (e.g., a non-key input object). It has been mentioned. Unless stated otherwise, it is to be understood that the symbols are intended to be assigned to the simplest predefined interaction, which is generally an act of pressing a key (as described in many embodiments of the present invention).

Although in different embodiments of the invention, a voice recognition system (eg, including a microphone) is mentioned as being used to recognize and recognize a user's speech, a lip reading system (eg, including a camera) May be used to recognize and recognize a user's speech in place of or in addition to a speech recognition system (or vice versa).

In addition, as mentioned above, some or all methods of the data entry systems of the present invention employ a linguistic rule text input recognition system such as the number of syllables of the matching words, the number of words in the recognized sentence, the position of the words in the paragraph, etc., as possible. It is available. These matters are known to those skilled in the art.

It will be noted that in some paragraphs the term “part-part” is used to simplify the term “at least part of a word (eg, by at least part of a word)”.

In many paragraphs, for the sake of simplicity, the data input system of the present invention is referred to as "data input systems of the present invention", "press / slide data input systems of the present invention", "press / slide and voice data input systems of the present invention. "," And the like, it will be understood that these phrases refer to the principles of the data input systems of the present invention in view of the push / slide operations combined with the user's utterance information. . Here, speech information is generally in the presence of a corresponding speech or in the absence of a user's speech (e.g., where at least one correct symbol is input through the above-mentioned system during the input of a word, the systems are generally of the present invention. Used with joint data entry systems, where at least one of the key presses provided for entry of a word is unclear). These have already been described in detail. Furthermore, in many paragraphs, the data entry systems of the present invention combining the push / slide and voice data entry systems of the present invention and word data entry prediction systems such as embodiments of the present invention are referred to as "combined" data of the present invention. It could be mentioned in the same phrase as the input systems.

Although in many embodiments of the data entry systems of the present invention, a sensitive surface such as a touch sensitive pad or touch screen has been used as an example, any other technique for detecting and analyzing interaction with a user's surface is soft. It can be used to define and / or use areas / keys of a (eg dynamic) keypad. For example, as mentioned, the technique is based on an optical detection technique, or on a (normal) surface (e.g., at least four keys / regions of the keypad) to detect a user's finger touching the keys / areas of the keypad. It may be an IR technology that provides a virtual keypad (including some keys / areas to which substantially all characters of the language are assigned).

In addition, according to one method, while a user enters a word by using the prediction system of the combined data entry systems of the present invention, the system up to that point in time (eg, before providing a word-signal-end). The closest word corresponding to the input information provided by the user may be output. The displayed word may change every time the user presses an additional key corresponding to the word.

For example, apparently, some or all of the principles of the embodiments of the present invention and other embodiments of the present inventors and / or their features may be used separately or in combination with each other. For example, at least one-word / at least-part of the text may be entered by combining different methods or embodiments of the present invention. For example, tapping / sliding operations on a sensitive surface submitted by the inventor and described in the references incorporated herein (eg, using interaction with several keys, such as using at least four keys on a keypad) Tapping and / or sliding operations provided by a user's finger, pen, glove, etc., reproducing the device, wherein at least the characters of the language are described in detail in the present application and references incorporated by this inventor and incorporated herein Assigned to several predefined keys, such as four of them, the glove of the data entry systems of the present invention reproduces the keys described in the references submitted and incorporated herein by this invention) It can be used as a data input prediction system.

In many instances, although the input information provided for the input of words is said to be based on key presses, in addition to key presses (eg, for more accuracy), if desired / allowed, the user may have at least some keys It will be appreciated that speech may be provided corresponding to the presses.

Although in many paragraphs describing the data entry methods of the present invention, it is mentioned that the input information provided by the user can obviously correspond to a symbol of a word based on the structure of the system, but in some cases the input information is It will be noted that they may correspond to the chain symbols of the word (see, for example, the word-of-part input system of the present invention). Obviously, as mentioned in detail previously, the symbolic data input systems of the present invention in combination with the word-of-part data input system of the present invention may be used as the (eg, word) data input prediction system of the present invention. For example, to enter the word "action", the user first presses the keys corresponding to the letters "a" and "c", and then defines a word-of-part of the word "tion". Based on the input information provided by the user, providing a sliding action on the key and uttering a portion (e.g., " section "), the data input prediction system of the present invention, for example, has two start symbols We can find a word that corresponds to a key press and the remaining symbols of the word are the chain symbols "tion". The system considers the word "action" and based on the frequency of use, the system may suggest the word to the user. In the above example, the user may correctly enter the symbol "a" or "c" by uttering a letter (e.g., while pressing the corresponding key) so that the system / user can provide the desired word more quickly / accurately. The use of the combined symbols and the use of the combined symbol and word-of-part data entry system of the present invention as the productive data entry system of the present invention has already been described in detail. At least one other data entry system, such as an on-screen full keyboard or a handwriting entry / recognition system, may be included / cooperated to enter the correct symbols that may correspond to. Additional sign and correct symbol (s) corresponding to the word provided through other data entry methods of the present invention, such as Information may be considered by the predictive data entry system of the present invention to provide the desired word.

For example, the combined data input systems of the present invention may be used in the illustrated expandable computer (eg, see FIGS. 70A-70H and the corresponding description in incorporated PCT / US04 / 12082) (eg, the isolated keypad of the present invention). May be used / executed).

Based on the structure of the system, part of the word-of-language of the language may be included in the word database dictionary of the language or may comprise a separate database, the combination of the present invention (eg, including a data input prediction system) data It should be noted once again that it can be used as input systems.

It will be noted that the utterance information corresponding to a letter is generally a utterance of a letter.

It will also be noted that key press (es) corresponding to at least one symbol of a word means providing consecutive key presses corresponding to the symbols in succession.

It will be noted that, like the foregoing arson, providing a predefined interaction with a key also includes pressing and holding another key and simultaneously pressing the key.

Although in many paragraphs describing the methods / features of the present invention, key press information is referred to for the entry of words, it is apparent that the method / features are in addition to the key press information that other input information such as It may be contemplated that this may be provided by the user.

As such, while important, inventive features of the invention are shown, described, and indicated as applied to alternative embodiments of the invention, various omissions, substitutions, and changes in form and detail of the invention disclosed may be made. It will be understood that those skilled in the art may be implemented without departing from this spirit of the invention. Therefore, the invention will 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 wholly conceptual.

Claims (50)

(a) receiving, via the first input means, a first input signal inaccurately corresponding to the plurality of characters; (b) predicting a first word from the word dictionary database 12601 (the predicted word corresponds to the received first input signal). (c) if the predicted word is not a required word, correcting information is received through a second input means different from the first input means in order to clearly identify at least the first character of the required word. Making; And (d) predicting a second word from the word dictionary database, wherein the second word corresponds to the first input signal and the at least first identified clearly identified word input data. Way. The method of claim 1, And receiving correction information to clearly identify at least the beginning of the required word and the subsequent symbol of the required word until the required word is obtained. Way. The method of claim 1, And a step of receiving an input signal of all symbols of a word. (a) first input means for providing an input signal inaccurately corresponding to a plurality of symbols; (b) second input means for providing correction information, which is different from the first input means; (c) word dictionary database 12601; (d) a word prediction system for predicting a first word from a word dictionary database based on the input signal; In a data input system comprising: When the second input means receives the correction information, the word prediction system provides a second word based on the input signal and the correction information, And said second input means is used to provide corrected information to accurately identify at least the first letter of a required word. 5. The method of claim 4, And said first input means is a keyboard comprising four keys to which all characters of a language are assigned indefinitely. 5. The method of claim 4, And said symbol comprises a character. delete The method of claim 1, The input of the input signal and the input of the correction information is a word prediction data input method, characterized in that provided by pressing a key (key). delete delete delete The method of claim 8, wherein And the input signal is provided through a key press on a first keyboard, and the correction information is provided through a key press on a second keyboard (14207, 14208). delete delete delete 13. The method of claim 12, 4. The word prediction data input method of claim 4, wherein all characters of a language are assigned indefinitely to four keys of the first keyboard. delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete 5. The method of claim 4, And said first input means provides an unclear symbol input signal for every symbol of a word. The method according to any one of claims 1 to 3, 8, 12 and 16, And receiving the correction information comprises receiving a plurality of correction information signals respectively corresponding to a single symbol. The method of claim 46, And the plurality of correction information signals are received while the cursor of the first input means is located at a specific position. 49. The method of claim 47, And the plurality of correction information signals are received while the cursor of the first input means is just after the last sign of a word. 13. The method of claim 12, And the second keyboard is a phone type keyboard. The method of claim 1, The symbol is a word prediction data input method, characterized in that the character.

Images