WO2014101410A1 - 输入处理方法和装置 - Google Patents

输入处理方法和装置 Download PDF

Info

Publication number
WO2014101410A1
WO2014101410A1 PCT/CN2013/080480 CN2013080480W WO2014101410A1 WO 2014101410 A1 WO2014101410 A1 WO 2014101410A1 CN 2013080480 W CN2013080480 W CN 2013080480W WO 2014101410 A1 WO2014101410 A1 WO 2014101410A1
Authority
WO
WIPO (PCT)
Prior art keywords
input
vocabulary
information
user
mode
Prior art date
Application number
PCT/CN2013/080480
Other languages
English (en)
French (fr)
Inventor
饶章文
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to KR1020137033362A priority Critical patent/KR101586890B1/ko
Priority to KR1020167000682A priority patent/KR20160011230A/ko
Priority to AU2013270485A priority patent/AU2013270485C1/en
Priority to EP13795149.7A priority patent/EP2765473A4/en
Priority to RU2014129212/08A priority patent/RU2589873C2/ru
Priority to JP2014553610A priority patent/JP5860171B2/ja
Priority to BR112013034056A priority patent/BR112013034056A2/pt
Priority to US14/097,467 priority patent/US20140184514A1/en
Publication of WO2014101410A1 publication Critical patent/WO2014101410A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/017Gesture based interaction, e.g. based on a set of recognized hand gestures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0487Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
    • G06F3/0488Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16Sound input; Sound output
    • G06F3/167Audio in a user interface, e.g. using voice commands for navigating, audio feedback
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/22Procedures used during a speech recognition process, e.g. man-machine dialogue
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2203/00Indexing scheme relating to G06F3/00 - G06F3/048
    • G06F2203/038Indexing scheme relating to G06F3/038
    • G06F2203/0381Multimodal input, i.e. interface arrangements enabling the user to issue commands by simultaneous use of input devices of different nature, e.g. voice plus gesture on digitizer
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/08Speech classification or search
    • G10L15/18Speech classification or search using natural language modelling
    • G10L15/183Speech classification or search using natural language modelling using context dependencies, e.g. language models
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M1/00Substation equipment, e.g. for use by subscribers
    • H04M1/72Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724User interfaces specially adapted for cordless or mobile telephones
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M2250/00Details of telephonic subscriber devices
    • H04M2250/70Details of telephonic subscriber devices methods for entering alphabetical characters, e.g. multi-tap or dictionary disambiguation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M2250/00Details of telephonic subscriber devices
    • H04M2250/74Details of telephonic subscriber devices with voice recognition means

Definitions

  • the present invention relates to communication technologies, and in particular, to an input processing method and apparatus. Background technique
  • the embodiment of the invention provides an input processing method and device for realizing fast and accurate input of complex vocabulary and effectively improving input efficiency.
  • a first aspect of the present invention provides an input processing method, including:
  • the terminal acquires first input information that is input by the user and matches the input mode according to an input mode selected by the user; the input mode is a gesture input mode or a voice input mode;
  • the terminal performs encoding processing on the first input information according to a preset encoding rule, to obtain an encoding corresponding to the first input information;
  • the terminal queries a pre-established vocabulary to obtain a candidate vocabulary corresponding to the encoding.
  • the acquiring, by the user input, the first input information that matches the input mode includes:
  • the encoding processing is performed on the first input information according to a preset encoding rule, and the encoding corresponding to the first input information is obtained, including:
  • the obtained first input information is a first touch trajectory of the user on the virtual keyboard of the terminal, combining characters corresponding to the keys sequentially passed by the first touch trajectory on the virtual keyboard, Obtaining an encoding corresponding to the first input information;
  • the hash information is used to encode the first sound information to obtain The code corresponding to the first input information.
  • the hidden Markov model HMM algorithm or the dynamic time warping DTW algorithm is used to encode the first sound information to obtain The code corresponding to the first input information.
  • the method further includes:
  • Receiving, by the terminal, a vocabulary processing request, requesting to add a vocabulary to the a vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added; the vocabulary processing mode is a gesture vocabulary processing mode or a speech vocabulary processing mode;
  • the terminal establishes a mapping relationship between the code corresponding to the second input information and the vocabulary to be added, and stores the mapping relationship in the vocabulary library;
  • the receiving, by the user, the second input information that is corresponding to the vocabulary processing mode includes: when the vocabulary When the processing mode is the gesture vocabulary processing mode, acquiring second touch track information input by the user on the virtual keyboard of the terminal; or
  • the second sound information is obtained by the sound sensor of the terminal.
  • the acquired second input information is a second touch trajectory of the user on the virtual keyboard of the terminal
  • the character corresponding to the button sequentially passed by the second touch trajectory on the virtual keyboard Combining to obtain an encoding corresponding to the second input information
  • the acquired second input information is the second sound information
  • encoding the second sound information by using a hash algorithm, a hidden Markov model HMM algorithm, or a dynamic time warping DTW algorithm, A code corresponding to the second input information is obtained.
  • the method further includes: when acquiring When the number of candidate vocabulary corresponding to the code is multiple, according to the word frequency from large to small a sequence of the plurality of candidate words, and displaying the arranged plurality of candidate words on a display screen of the terminal for selection by a user;
  • a vocabulary selected by the user from the arranged plurality of candidate vocabularies is displayed on a specific area on the display screen of the terminal.
  • a second aspect of the present invention provides an input processing apparatus, including:
  • An information obtaining module configured to acquire, according to an input mode selected by the user, first input information that is input by the user and that matches the input mode; the input mode is a gesture input mode or a voice input mode;
  • An encoding processing module configured to perform encoding processing on the first input information according to a preset encoding rule, to obtain an encoding corresponding to the first input information
  • the query obtaining module is configured to query a pre-established vocabulary to obtain a candidate vocabulary corresponding to the encoding.
  • the information acquiring module is configured to acquire, when the input mode is a gesture input mode, acquiring first touch track information of a user on a virtual keyboard of the input processing device.
  • the information acquiring module is specifically configured to: when the input mode is a voice input mode, obtain the first sound information by collecting the sound sensor of the input processing device;
  • the encoding processing module is specifically configured to: when the first input information acquired by the information acquiring module is a first touch track of a user on a virtual keyboard of the input processing device, the first touch track Combining characters corresponding to the keys sequentially passed on the virtual keyboard to obtain an encoding corresponding to the first input information; or
  • the encoding processing module is specifically configured to be used when the first input is obtained by the information acquiring module
  • the information is the first sound information
  • the first sound information is encoded by using a hash algorithm, a hidden Markov model HMM algorithm, or a dynamic time warping DTW algorithm, and the first input information is obtained. Corresponding code.
  • the input processing device further includes: a receiving module and a vocabulary storage module, wherein the receiving module And for receiving a vocabulary processing request, the vocabulary processing request is used to request adding a vocabulary to the vocabulary; the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added; the vocabulary processing mode is a gesture vocabulary processing mode or a voice Vocabulary processing mode;
  • the information obtaining module is further configured to receive, according to the vocabulary processing request, second input information that is input by the user and that matches a vocabulary processing mode;
  • the encoding processing module is further configured to perform encoding processing on the second input information according to the preset encoding rule, to generate an encoding corresponding to the second input information;
  • the vocabulary storage module is configured to establish a mapping relationship between the code corresponding to the second input information and the vocabulary to be added, and store the mapping relationship in the vocabulary library.
  • the information acquiring module is specifically configured to: when the vocabulary processing mode is a gesture vocabulary processing mode, obtain a user The second touch track information on the virtual keyboard of the input processing device; or
  • the information acquiring module is specifically configured to: when the vocabulary processing mode is a voice vocabulary processing mode, obtain the second sound information by using the sound sensor of the input processing device according to the voice vocabulary processing mode.
  • the encoding processing module is specifically configured to: when the information acquiring module acquires the second input When the information is the second touch trajectory of the user on the virtual keyboard of the input processing device, the characters corresponding to the keys sequentially passed by the second touch trajectory on the virtual keyboard are combined to obtain the The code corresponding to the two input information; or,
  • the encoding processing module is specifically configured to: when the second input information acquired by the information acquiring module is the second sound information, use a hash algorithm, a hidden Markov model HMM algorithm, or a dynamic time warping DTW An algorithm, performing encoding processing on the second sound information to obtain an encoding corresponding to the second input information
  • the input processing device further includes:
  • a sorting processing module configured to: when the number of candidate vocabularies corresponding to the encoded code is multiple, arrange the plurality of candidate vocabularies according to a sequence of word frequencies, and arrange the plurality of candidate vocabularies
  • the candidate vocabulary is displayed on the display screen of the input processing device for the user to select;
  • a display module configured to display a vocabulary selected by the user from the arranged plurality of candidate vocabularies on a specific area on a display screen of the input processing device.
  • the first input information that is input by the user and matched with the input mode is obtained according to the input mode selected by the user, and the first input information is encoded according to a preset encoding rule, and the The code corresponding to the first input information, finally querying the pre-established vocabulary, obtaining the candidate vocabulary corresponding to the code, acquiring the first input information according to different input modes, and acquiring the corresponding code according to the first input information, Then, the vocabulary that the user needs to input is obtained by coding, thereby realizing the fast and accurate input of the complex vocabulary, effectively improving the input efficiency and improving the user experience.
  • FIG. 1 is a flow chart of an embodiment of an input processing method of the present invention
  • FIG. 2 is a flow chart of still another embodiment of an input processing method of the present invention.
  • FIG. 3 is a flow chart of still another embodiment of an input processing method of the present invention.
  • FIG. 4 is a schematic diagram of an input interface of a terminal device
  • FIG. 5 is another schematic diagram of an input interface of the terminal device
  • FIG. 6 is a flow chart of still another embodiment of an input processing method of the present invention.
  • FIG. 7 is a schematic diagram of an input interface of the terminal device
  • FIG. 8 is a schematic structural diagram of an embodiment of an input processing apparatus according to the present invention.
  • FIG. 9 is a schematic structural diagram of still another embodiment of an input processing device according to the present invention.
  • Figure 10 is a block diagram showing another embodiment of the input processing device of the present invention.
  • FIG. 1 is a flowchart of an embodiment of an input processing method according to the present invention. As shown in FIG.
  • the execution body of the input processing method provided by the embodiment is a terminal, and the input processing method specifically includes the following steps: Step 101: The terminal acquires first input information that is input by the user and matches the input mode according to the input mode selected by the user, and the input mode is a gesture input mode or a voice input mode.
  • the terminal may be a mobile phone or a tablet computer.
  • the user selects an input mode by triggering a corresponding button on the terminal input interface.
  • the input mode is a voice input mode or a gesture input mode.
  • the different input modes correspond to different input information, and therefore, the first input information is different according to different input modes.
  • the voice input mode the first input information is a voice information
  • the gesture input mode the first input information is a touch track information
  • the touch track information may be, for example, text graphic information.
  • Step 102 The terminal performs encoding processing on the first input information according to a preset encoding rule, to obtain an encoding corresponding to the first input information.
  • the first input information is encoded according to a preset encoding rule, and the code corresponding to the first input information is obtained.
  • corresponding coding rules are set correspondingly, and corresponding codes are generated.
  • the gesture input mode is preset with an encoding rule
  • the voice input mode is also preset with an encoding rule.
  • Step 103 The terminal queries a pre-established vocabulary to obtain a candidate vocabulary corresponding to the code.
  • the pre-established vocabulary may be a vocabulary on a server connected to the terminal, or a local vocabulary established by the terminal itself, or a local vocabulary that the terminal downloads from the vocabulary on the server and saves to the local storage.
  • Library the vocabulary can include popular vocabulary, proprietary vocabulary, user-defined vocabulary, etc., and the code is established with the above A mapping relationship of vocabulary to enable rapid indexing of corresponding candidate vocabularies in a local vocabulary based on encoding.
  • the form of the candidate vocabulary may specifically be one or a combination of the following: a single word, a phrase, a sentence, a long string of numbers and letters, etc.; for example: QQ number 1234657; mailbox yagneos235 @hotmail.com.cn.
  • the one candidate vocabulary is the vocabulary that the user needs to input.
  • the candidate vocabulary may be sorted according to the order of the word frequency from large to small.
  • the plurality of candidate words after the sorting process are displayed on the display screen of the terminal for the user to select, and the words selected by the user from the arranged plurality of candidate words are displayed on a specific area on the display screen of the terminal.
  • the first input information that is input by the user and matched with the input mode is obtained according to the input mode selected by the user, and the first input information is encoded according to a preset encoding rule, and the Corresponding to the first input information, finally querying the pre-established vocabulary, obtaining the candidate vocabulary corresponding to the encoding, acquiring the first input information according to different input modes, and obtaining the corresponding encoding according to the first input information, and then By encoding the vocabulary that the user needs to input, the fast and accurate input of the complex vocabulary is realized, and the input efficiency is effectively improved.
  • the specific implementation manner of obtaining the first input information that is input by the user and matching the input mode in step 101 may be as follows: Several kinds:
  • the first type when the input mode is the gesture input mode, acquiring a first touch track input by the user on the virtual keyboard of the terminal;
  • the first sound information is obtained through the sound sensor of the terminal.
  • a vocabulary library is required to be pre-established on the server, and the vocabulary includes hot words, proprietary vocabulary, user customized vocabulary, and the like.
  • the customized vocabulary is pre-customized by the user, and is backed up to the vocabulary in the vocabulary by the terminal device, and the customized vocabulary needs to be associated with the user's personal account.
  • the vocabulary stores the user identifier and the user identifier.
  • the server indexes the collected words, such as pinching, pinyin letters, strokes, and frequency of occurrence.
  • the user can download the vocabulary from the local vocabulary on the terminal device such as a mobile phone or a tablet computer through the terminal device.
  • the user customized vocabulary in the local vocabulary is related to the user. The corresponding logo.
  • the local vocabulary can also have powerful customization features. Users can add some common vocabulary to a separate file in the local vocabulary, upload the file to the server, and the server writes the vocabulary in the file to the user's customized vocabulary for backup processing.
  • the server may also retrieve the user-customized vocabulary corresponding to the plurality of user identifiers, and index the words in the order of the word frequency to generate the hot words to be added to the vocabulary.
  • the terminal device can preferentially download the vocabulary associated with the user account, that is, the user-defined vocabulary into the local vocabulary.
  • the server will prompt the user to have an update, the user selects the update, and downloads the professional vocabulary and popular vocabulary into the local vocabulary.
  • the local vocabulary has a powerful custom function, and the user can add his or her commonly used vocabulary to the vocabulary through the self-defined function, so as to facilitate subsequent use by the user.
  • the following sections detail how users customize their vocabulary.
  • FIG. 2 is an input processing method of the present invention.
  • a flow chart of another embodiment the embodiment is based on the embodiment shown in FIG. 1 , and before the step 101, the method may further include the following steps:
  • Step 201 The terminal receives a vocabulary processing request, where the vocabulary processing request is used to request to add a vocabulary to the vocabulary; the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added; the vocabulary processing mode is a gesture vocabulary processing mode or a speech vocabulary processing mode.
  • the vocabulary processing mode is a gesture vocabulary processing mode or a speech vocabulary processing mode
  • the vocabulary processing request is used to request to add a vocabulary to the vocabulary.
  • the user can select a vocabulary processing mode by triggering a corresponding operation button on the input interface, and input a vocabulary to be added in the vocabulary processing mode, and the vocabulary to be added is a vocabulary that does not exist in the local vocabulary, and is a user-defined vocabulary, to be
  • the added vocabulary can be used to add the vocabulary commonly used by the user to the vocabulary by using a vocabulary processing request for the user's commonly used QQ number, mailbox, password, and phone number.
  • Step 202 The terminal receives, according to the vocabulary processing request, the second input information that is input by the user and matches the vocabulary processing mode.
  • the second input information is encoded according to a preset encoding rule, and the second input is generated.
  • the second input information input by the user that matches the vocabulary processing mode is received, and the second input information may be different for different vocabulary processing modes.
  • the second input information is a touch trajectory of the user on the virtual keyboard of the terminal.
  • the voice input mode the second input information is sound information collected by the sound sensor of the terminal.
  • the received second input information is encoded according to a preset encoding rule, and a code corresponding to the second input information is generated, the code uniquely identifying the second input information.
  • different encoding methods are used.
  • the pre-set encoding rule of the gesture vocabulary processing mode is a character corresponding to the keyboard that the user sequentially passes through the first touch trajectory on the virtual keyboard of the terminal.
  • the vocabulary processing mode preset encoding rules are a hash algorithm, a Hidden Markov models ( ⁇ ) algorithm, or a Dynamic Time Warping (DTW) algorithm.
  • Step 203 The terminal establishes a mapping relationship between the code corresponding to the second input information and the vocabulary to be added, and stores the mapping relationship in the sink.
  • a mapping relationship between the code corresponding to the second input information and the vocabulary to be added is established according to the code and vocabulary corresponding to the second input information, and the mapping relationship is stored in the vocabulary library. In this way, the user input information and the vocabulary are associated.
  • the vocabulary added by the user is saved in the vocabulary, and a mapping relationship between the user input information and the vocabulary is established.
  • the input information of the user can be identified and the code corresponding to the input information is obtained.
  • the index is quickly indexed to the candidate vocabulary required by the user.
  • the execution body of the method is a terminal device, such as a mobile phone
  • the input mode is a gesture input processing mode
  • the first input information is used.
  • the touch track information input by the user is taken as an example to describe the technical solution of the embodiment in detail.
  • the method includes:
  • Step 301 Receive a vocabulary processing request, where the vocabulary processing request includes a gesture vocabulary processing mode and a vocabulary to be added, the vocabulary processing request is used to add a vocabulary to the vocabulary.
  • the vocabulary processing request is input by the user through the input interface of the terminal device.
  • 4 is a schematic diagram of an input interface of a terminal device
  • FIG. 5 is another schematic diagram of an input interface of the terminal device.
  • the user opens a page for adding a customized vocabulary on the input interface, and corresponding buttons are displayed on the page.
  • the user selects the gesture vocabulary processing mode, and the page shown in FIG. 5 appears.
  • the user inputs the vocabulary to be added 12345678, and sets the remark information for the vocabulary, and the remark is the QQ number. .
  • Step 302 Receive, according to the vocabulary processing request, a second touch trajectory that is input by the user and that matches the gesture vocabulary processing mode, and perform encoding processing on the second touch trajectory according to a preset encoding rule to generate the second touch The code corresponding to the track.
  • the user in the gesture vocabulary processing mode, the user generates a second touch trajectory by sliding on the virtual keyboard of the terminal, and when acquiring the second touch trajectory of the user on the virtual keyboard of the terminal, according to the preset encoding.
  • Rule encoding the second touch track to generate a code corresponding to the second touch track.
  • the encoding rule is preset.
  • the characters corresponding to the keys sequentially passed by the second touch track on the virtual keyboard are combined to obtain a code corresponding to the second touch track.
  • the character is a number or a letter, so the number corresponding to the button that the second touch track passes in sequence may be encoded, or the letter corresponding to the button through which the second touch track passes sequentially may be encoded.
  • the second touch track information is an L-shaped figure drawn by the user, and the figure is drawn by a user's stroke.
  • the touch track of the L-shaped figure sequentially passes through the number key 1478 on the keyboard, and the starting point of the track is 1 , the end point is 8, the end of the touch track marks the end of the second touch track.
  • the characters corresponding to the keys that the second touch track sequentially passes on the virtual keyboard are combined as an encoding rule.
  • the virtual keyboard on the input interface is a conventional numeric keyboard, and the numeric keyboard has 0.
  • the user can draw the graph arbitrarily on the numeric keypad, and encode the number corresponding to the button that the user touches the touch track in sequence. Therefore, for the embodiment, the code corresponding to the second touch track is 1478.
  • the input interface is an analog keyboard of a PC, it can be encoded according to the letter corresponding to the button that the touch track passes.
  • Step 303 Establish a mapping relationship between the code corresponding to the second touch track and the to-be-added vocabulary, and store the mapping relationship in a pre-established vocabulary.
  • the code corresponding to the second touch track is generated as 1478, and the code 1478 is mapped to the to-be-added vocabulary input in step 201, and the to-be-added vocabulary is QQ number 12345678, that is, the code 1478 is unique. Identifies the vocabulary 12345678 and stores the mapping into the vocabulary.
  • a touch track of the user corresponds to a unique code.
  • one code may also correspond to multiple words, for example, the code 1478 may also correspond to other words, for example: when the user has multiple QQ numbers.
  • the code 1478 can correspond to multiple QQ numbers, it is only necessary to increase the mapping relationship between the code 1478 and the vocabulary.
  • the vocabulary can also store the mapping relationship between the encoding and the vocabulary in the following form, as shown in Table 1:
  • Step 304 Acquire a first touch trajectory that matches the gesture input mode according to the gesture input mode selected by the user.
  • Step 305 Perform encoding processing on the first touch track according to a preset encoding rule, and acquire a code corresponding to the first touch track.
  • the set encoding rule is the same as the encoding rule used to encode the second touch track in step 302, specifically, corresponding to the button that sequentially passes the first touch track on the virtual keyboard. The characters are combined to obtain an encoding corresponding to the first input information.
  • Step 306 Query a pre-established vocabulary to obtain a candidate vocabulary corresponding to the code.
  • the candidate vocabulary corresponding to the encoding can be obtained by performing a search in the vocabulary according to the obtained encoding.
  • the input interface on the terminal device can be switched to the gesture input mode.
  • the user draws a graphic through the input interface to trigger the terminal device according to the trigger.
  • the gesture input mode acquires the touch track information, and identifies the touch track information to obtain a corresponding code. For example, if the area where the user touch track passes is 1478, the code is 1478. Then, the local vocabulary is queried, and the candidate vocabulary corresponding to the code 1478 is obtained, that is, the QQ number 12345678.
  • the candidate vocabulary (ie, QQ number 12345678) is a vocabulary that the user needs to input; if the number of the candidate vocabulary is multiple, more
  • the candidate vocabulary sorts the plurality of candidate vocabularies according to the order of the word frequency, and displays the sorted plurality of candidate vocabularies on the display screen of the terminal, so that the user selects one of the candidate vocabularies as the user needs.
  • the input vocabulary finally, the terminal displays the vocabulary selected by the user from the arranged plurality of candidate vocabularies in a specific area on the display screen of the terminal, wherein the specific area refers to an application currently used by the user or waiting for text The input area specified by the input application.
  • the input processing method provided in this embodiment establishes a correspondence between a cumbersome vocabulary and a simple graphic by storing a touch trajectory of a user-drawn simple graphic, a code corresponding to the touch trajectory, and a mapping relationship between the encoding and the vocabulary in a local vocabulary.
  • This vocabulary is some of the cumbersome words commonly used by users, such as users. Personal QQ account, email, password, etc.
  • the user When in use, the user only needs to switch the input mode to the gesture input mode, draw a graphic on the input interface, and the touch track information of the device recognition graphic obtains the code corresponding to the touch track information, according to the mapping relationship between the code and the vocabulary in the vocabulary
  • the corresponding candidate vocabulary is found, and the candidate vocabulary is used as the vocabulary that the user needs to input, so that the user can input long and complex vocabulary with only one input, thereby improving input efficiency and accuracy.
  • the input processing method provided in this embodiment is applicable to the input of complex and cumbersome vocabulary. Compared with the traditional handwriting input, the input processing method provided by the embodiment does not perform the drawing of the graphic and the text, but the drawing of the graphic.
  • the touch track establishes a correspondence between the code and the vocabulary. For example, when an L-shaped graphic is input, it is not necessary to identify whether the graphic is a letter L, but a corresponding encoding is obtained according to the touch trajectory of the graphic, and the corresponding vocabulary is searched according to the encoding, which not only improves the input efficiency but also improves the input efficiency. The accuracy of the input.
  • FIG. 6 is a flowchart of still another embodiment of the input processing method of the present invention.
  • the execution body of the method is a terminal device, such as a mobile phone, and the input mode is a voice input processing mode, and the first input information is used.
  • the sound information input by the user is taken as an example, and the technical solution of the embodiment is described in detail.
  • the method includes:
  • Step 601 Receive a vocabulary processing request, where the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added; the vocabulary processing request is used to add a vocabulary to the vocabulary.
  • the vocabulary processing request is input by the user through the input interface of the terminal device.
  • FIG. 7 is still another schematic diagram of the input interface of the terminal device.
  • the user opens a page for adding a customized vocabulary on the input interface, and corresponding buttons are set on the page to respectively represent different vocabulary processing modes, as shown in FIG. 4 .
  • a speech vocabulary processing mode When the user selects a speech vocabulary processing mode, a page as shown in FIG. 7 appears, and in the speech vocabulary processing mode, the user inputs the Add the vocabulary, and set the remark information for the vocabulary to be added.
  • the remark is 163 mailboxes.
  • Step 602 Receive second voice information that is input by the user and that matches the voice vocabulary processing mode according to the vocabulary processing request, and perform encoding processing on the second voice information according to a preset encoding rule to generate the second sound.
  • the code corresponding to the information is
  • the second sound information input by the user that matches the voice vocabulary processing mode is received, and the second sound information is obtained by the terminal through the sound sensor.
  • the second sound information is encoded according to a preset encoding rule, and a code corresponding to the second sound information is generated.
  • the hash algorithm, the hidden Markov model ⁇ algorithm, or the dynamic time warping DTW algorithm may be used to encode the second sound information to obtain an encoding corresponding to the second input information.
  • a button for starting the utterance can be set on the input interface of the terminal device (also as an end sound button), and the utterance button is activated by pressing and holding.
  • a microphone or other voice input device on the terminal device the user utters a voice, for example, the user sends out the voice information of the mailbox, the terminal device starts to collect the user's voice information, the user releases the start sound button, completes a voice information collection, and collects the voice information.
  • the sound information is the second sound information.
  • the second sound information is encoded to generate a code corresponding to the sound information.
  • the coding rule is preset, and the coding rule may use a speech recognition algorithm commonly used in the prior art, such as a hash algorithm, a hidden Markov model, a dynamic time warping DTW algorithm, etc.
  • the obtained sound information is converted into a character string or a text, and the embodiment of the present invention does not particularly limit the encoding rule.
  • the code corresponding to the second sound information is obtained by encoding the second sound information input by the user. In this embodiment, for example, the code corresponding to the second sound information is Sjdegsogea4512.
  • Step 603 Establish a mapping relationship between the code corresponding to the second sound information and the vocabulary to be added, and store the mapping relationship in the vocabulary.
  • a code Sjdegsogea4512 corresponding to the second sound information is generated, and the code Sjdegsogea4512 and the to-be-added word Jackv.Chen@163.com input in step 601 are mapped, and the code Sjdegsogea4512 uniquely identifies the word Jackv. .Chen@163.com , and store the mapping in the vocabulary.
  • a voice information of the user corresponds to a unique code. It should be noted that one code can correspond to multiple words, for example, the user has multiple mailbox numbers. For the convenience of memory, the user encodes the voice information mailbox correspondingly. To establish a correspondence between mailbox numbers, simply add the mapping relationship between the code and the corresponding vocabulary in the local vocabulary.
  • the local vocabulary can also store the mapping relationship between encoding and vocabulary in the following form, as shown in Table 2:
  • the audio information input by the user is recorded in the audio 1, the audio 2, and the audio 3.
  • Step 604 Acquire first sound information that matches the voice input mode according to a voice input mode selected by the user.
  • Step 605 Perform encoding processing on the first sound information according to a preset encoding rule, and acquire an encoding corresponding to the first sound information.
  • the coding rule used in this step is the same as the coding rule used in step 602. Specifically, the hash algorithm, the hidden Markov model HMM algorithm, or the dynamic time warping DTW algorithm may be used, and the first sound information is used. Encoding processing is performed to obtain an encoding corresponding to the first input information.
  • Step 606 Query a pre-established vocabulary to obtain a candidate vocabulary corresponding to the code.
  • the candidate vocabulary corresponding to the encoding can be obtained by performing a search in the vocabulary according to the obtained encoding.
  • the input interface of the terminal device is first switched to the voice input mode.
  • the voice input mode the user sends a voice message by pressing and holding the start sound button on the input interface.
  • the triggering terminal device acquires the sound information according to the voice input mode, and identifies the sound information, and obtains a code corresponding to the sound information. For example, if the code corresponding to the user input voice information is Sjdegsogea4512, the local vocabulary is queried. Get the 4 vocabulary corresponding to the code Sjdegsogea4512, namely Jackv.Chen (3 ⁇ 4 163. com.
  • Jackv.Chen@163.com is a vocabulary that the user needs to input; if the number of the candidate vocabulary is plural, multiple candidate vocabularies can be sorted according to the frequency of the word frequency, and the plurality of candidate vocabularies are sorted, and The plurality of candidate vocabulary words are displayed to the user, so that the user selects one of the candidate vocabulary words as the vocabulary that the user needs to input, and the user selects the vocabulary input that needs to be input from the plurality of sorted candidate vocabulary corresponding to the encoding.
  • the input processing method provided in this embodiment establishes cumbersome vocabulary and sound by storing the short voice information of the user, the code corresponding to the sound information, and the mapping relationship between the code and the vocabulary in the local vocabulary.
  • the vocabulary is some cumbersome words commonly used by users, such as user mailbox, password, mobile phone number, and so on.
  • the user only needs to switch the input mode to the voice input mode, perform corresponding operations, input the user's voice information, the device recognizes the sound information and obtains the code corresponding to the voice information, according to the mapping relationship between the code and the vocabulary, locally Find the corresponding candidate vocabulary in the vocabulary. Simply input short voice information to complete complex and long vocabulary input and improve input efficiency.
  • the input processing method provided in this embodiment is applicable to an operation scene that is shaken and difficult to be handwritten, such as a user needs to input during driving.
  • the input mode is switched to the voice input mode, the user only needs to input a short voice information, and the complicated word input can be completed, which is convenient and the input efficiency is high.
  • the input processing method provided in this embodiment does not make the sound and the text only correspond to the input sound and the vocabulary, thereby improving the accuracy of the input.
  • the present invention can also use the character input mode to obtain the vocabulary that the user needs to input.
  • the user can switch the input interface to the character input mode, and the terminal device acquires the user input character information in the mode, and retrieves the candidate vocabulary corresponding to the character information from the local vocabulary, and retrieves the retrieved vocabulary.
  • the plurality of candidate vocabulary corresponding to the character information are sorted in descending order of lexical frequency, and the sorted plurality of candidate vocabularies are displayed to the user through the interface of the terminal device. The user selects the vocabulary to be input from the returned candidate vocabulary.
  • the terminal device may perform a secondary index in the local vocabulary according to the context of the vocabulary, and predict After the user can input the vocabulary in the next step, after the secondary indexing, the user can directly select the vocabulary to be input from the candidate vocabulary obtained after the secondary index, without Enter the characters again.
  • the user inputs "ao” through pinyin and retrieves a plurality of candidate vocabulary (words) corresponding to it in the local vocabulary by "ao”: Austria, ⁇ , ⁇ , ⁇ , etc., these candidate words can be based on The word frequency is sorted and outputted in descending order.
  • the user selects the candidate vocabulary "Olympic” as the vocabulary that the user needs to input, and can also perform secondary indexing in the local vocabulary according to "Ao", and obtain from the local vocabulary.
  • "Affinity” is associated with multiple candidate vocabulary, such as Yun, Ba, Lin, Mei, etc., for users to choose.
  • FIG. 8 is a schematic structural diagram of an embodiment of an input processing device according to the present invention.
  • the input processing device 800 of the present embodiment includes an information acquiring module 801, an encoding processing module 802, and a query obtaining module 803.
  • the information obtaining module 801 is configured to obtain, according to an input mode selected by the user, first input information input by the user that matches the input mode; the input mode is a gesture input mode or a voice input mode.
  • the encoding processing module 802 is configured to perform encoding processing on the first input information according to a preset encoding rule to obtain an encoding corresponding to the first input information.
  • the query obtaining module 803 is configured to query a pre-established vocabulary to obtain a candidate vocabulary corresponding to the encoding.
  • the input processing device 800 of this embodiment may perform the technical solution of the method embodiment shown in FIG. 1 , and the implementation principles thereof are similar, and details are not described herein again.
  • the first input information that matches the input mode is obtained according to the input mode selected by the user, and the first input information is encoded according to a preset encoding rule to obtain a correspondence with the first input information.
  • Encoding finally querying a pre-established vocabulary, obtaining candidate vocabulary corresponding to the encoding, acquiring first input information according to different input modes, and according to the first input information
  • the corresponding code is obtained, and the vocabulary that the user needs to input is obtained by coding, thereby realizing the fast and accurate input of the complex vocabulary and effectively improving the input efficiency.
  • the input processing device further includes: a touch screen and a sound sensor, and the virtual screen is displayed on the touch screen for The user performs a touch operation, and the sound sensor is configured to collect sounds.
  • the information acquiring module 801 is specifically configured to acquire a first touch of the user on the virtual keyboard of the input processing device when the input mode is the gesture input mode.
  • the trajectory; the encoding processing module 802 is specifically configured to: when the first input information acquired by the information acquiring module 801 is a first touch trajectory of the user on the virtual keyboard of the input processing device, according to a preset encoding rule, the first A touch track is combined with characters corresponding to the keys sequentially passed on the virtual keyboard to obtain a code corresponding to the first input information.
  • the information acquiring module 801 is further configured to: when the input mode is the voice input mode, obtain the first sound information by collecting the sound sensors of the input processing device.
  • the encoding processing module 802 is specifically configured to: when the first input information acquired by the information acquiring module 802 is the first sound information, according to a preset encoding rule, such as a hash algorithm, a hidden Markov model, or a dynamic HMM algorithm, or dynamic The time warping DTW algorithm performs encoding processing on the first sound information to obtain a code corresponding to the first input information.
  • the query obtaining module 803 queries the pre-established vocabulary to obtain the candidate vocabulary corresponding to the encoding, and is further configured to: when the number of candidate vocabularies corresponding to the acquired encoding is multiple, according to the order of the word frequencies from large to small And arranging the plurality of candidate vocabularies, and displaying the arranged plurality of candidate vocabularies to the user.
  • the vocabulary corresponding to the code input by the user is received, and the vocabulary corresponding to the code is a vocabulary selected by the user from the plurality of candidate vocabulary words after the arrangement.
  • FIG. 9 is a schematic structural diagram of still another embodiment of an input processing apparatus according to the present invention, as shown in FIG.
  • the input processing device 900 provided in this embodiment includes: a receiving module 901, an information obtaining module 902, an encoding processing module 903, a vocabulary storage module 904, a query obtaining module 905, a sorting processing module 906, and a display module 907.
  • the receiving module 901 is configured to receive a vocabulary processing request, where the vocabulary processing request is used to request to add a vocabulary to the vocabulary, where the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added, and the vocabulary processing mode is a gesture vocabulary processing mode or a voice
  • the vocabulary processing mode 902 is configured to receive second input information that is input by the user and that matches the vocabulary processing mode according to the vocabulary processing request.
  • the encoding processing module 903 is configured to use the second input information according to a preset encoding rule.
  • the vocabulary storage module 904 is configured to establish a mapping relationship between the encoding corresponding to the second input information and the vocabulary to be added, and store the mapping relationship into the vocabulary library in.
  • the information obtaining module 902 is further configured to: acquire, according to an input mode selected by the user, first input information input by the user that matches the input mode; the input mode is a gesture input mode or a voice input mode.
  • the encoding processing module 903 is further configured to perform encoding processing on the first input information according to a preset encoding rule to obtain an encoding corresponding to the first input information.
  • the query obtaining module 905 is configured to query a pre-established vocabulary to obtain A candidate vocabulary corresponding to the code.
  • the vocabulary processing request received by the receiving module 901 is input by the user through the terminal input interface, and the vocabulary processing request is used to request to add a vocabulary to the vocabulary, the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added.
  • the vocabulary processing mode is a speech vocabulary processing mode or a gesture vocabulary processing mode.
  • the vocabulary to be added is a vocabulary that does not exist in the vocabulary.
  • the vocabulary can be a single word, phrase, sentence, long string of numbers and letters; for example: QQ number 8945145182; mail box yagneos235@hotmail.com.cn.
  • the information obtaining module 902 is configured to receive, according to the vocabulary processing request, second input information input by the user that matches the vocabulary processing mode.
  • the information obtaining module 902 is specifically configured to use the vocabulary processing mode Obtaining a second touch trajectory of the user on the virtual keyboard of the input processing device when the vocabulary processing mode is gestured, and obtaining a second sound by collecting the sound sensor of the input processing device when the vocabulary processing mode is the vocabulary processing mode information.
  • the encoding processing module 903 is configured to encode the second input information according to a preset encoding rule, and generate a code corresponding to the second input information.
  • different encoding methods are used for different second input information, and the encoding rules are predefined.
  • the encoding processing module 903 is specifically configured to: when the second input information acquired by the information acquiring module 902 is a second touch track on the virtual keyboard of the input processing device, the second touch track is sequentially pressed on the virtual keyboard.
  • the corresponding characters are combined to obtain an encoding corresponding to the second input information; when the second input information acquired by the information acquiring module 902 is the second sound information, using a hash algorithm, a hidden Markov model HMM algorithm, or The dynamic time warping DTW algorithm performs encoding processing on the second sound information to obtain an encoding corresponding to the second input information.
  • the code corresponding to the second input information uniquely identifies the second input information.
  • the vocabulary storage module 904 is configured to establish a mapping relationship between the code corresponding to the second input information and the vocabulary to be added, and store the mapping relationship in the vocabulary library.
  • the vocabulary storage module 904 establishes a mapping relationship between the encoding corresponding to the second input information generated by the encoding processing module 903 and the vocabulary to be added received by the receiving module 901, and stores the mapping relationship in a preset vocabulary. in. In this way, a mapping relationship between user input information and vocabulary is established, and the mapping relationship is saved in the local vocabulary.
  • the user's input information can be identified and the code corresponding to the input information can be obtained.
  • the information obtaining module 902 is further configured to: acquire, according to an input mode selected by the user, first input information that is input by the user and that matches the input mode, where the input mode is a gesture input mode or a language. Sound input mode.
  • the information acquiring module 902 is specifically configured to: when the input mode is the gesture input mode, acquire a first touch track of the user on the virtual keyboard of the input processing device; when the input mode is the voice input mode, input the sound sensor of the processing device The first set of sound information is obtained.
  • the encoding processing module 903 After the information acquisition module 902 obtains the first input information input by the user, the encoding processing module 903 performs encoding processing on the first input information according to a preset encoding rule to obtain a code corresponding to the first input information.
  • the encoding processing module 903 is specifically configured to combine the characters corresponding to the keys sequentially passed by the first touch track on the virtual keyboard when the acquired first input information is the first touch track of the user on the virtual keyboard of the input processing device.
  • the character may be a number or a letter, that is, the number corresponding to the button sequentially passed by the first touch track on the virtual keyboard may be combined to obtain a code, or the first touch may be The trajectory is combined by the letters corresponding to the keys that pass through the virtual keyboard in sequence.
  • the encoding processing module 903 is further configured to: when the obtained first input information is the first sound information, use a hash algorithm, a hidden Markov model HMM algorithm, or a dynamic time warping DTW algorithm to encode the first sound information. , obtaining an encoding corresponding to the first input information.
  • the query obtaining module 905 queries the pre-established vocabulary to obtain candidate vocabulary corresponding to the encoding. Specifically, the query obtaining module 905 searches for a pre-established vocabulary according to the code corresponding to the first input information obtained by the encoding processing module 903, where the vocabulary stores the mapping relationship between the encoding and the corresponding vocabulary, and encodes and A mapping relationship of words, and obtaining candidate words corresponding to the code.
  • the candidate vocabulary is a vocabulary that the user needs to input; if the number of the candidate vocabulary is multiple, the sorting processing module 906 arranging a plurality of candidate vocabularies according to a sequence of word frequencies from large to small, and displaying the arranged plurality of candidate vocabularies on a display screen of the input processing device for selection by the user; Block 907 is configured to display a vocabulary selected by the user from the arranged plurality of candidate vocabularies in a specific area on a display screen of the input processing device.
  • the input processing device provided in this embodiment may be used to perform the technical solution of the embodiment shown in the method of the present invention.
  • the implementation principle and technical effects are similar, and details are not described herein again.
  • FIG. 10 is a schematic structural view of another embodiment of an input processing apparatus according to the present invention.
  • the input processing apparatus 700 of this embodiment includes: at least one processor 701, a memory 702, and a bus.
  • the processor 701 and the memory 702 are connected by a bus and complete communication with each other.
  • the bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Wait.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 10, but it does not mean that there is only one bus or one type of bus. among them:
  • Memory 702 is for storing executable program code, the program code including computer operating instructions.
  • the memory 702 may include a high speed random access memory (RAM) and may also include a non-volatile memory such as at least one disk memory.
  • processor 701 runs a program corresponding to the executable program code by reading executable program code stored in memory 702 for:
  • the first input information input by the user and matching the input mode is obtained according to the input mode selected by the user, and the input mode is a gesture input mode or a voice input mode.
  • the first input information is encoded according to a preset encoding rule, and an encoding corresponding to the first input information is obtained.
  • the pre-established vocabulary is queried to obtain candidate vocabulary corresponding to the code.
  • the input processing device further includes: a touch screen and a sound sensor, and the virtual screen is displayed on the touch screen for the user to perform a touch operation, and the sound sensor is configured to collect sounds; Encoding the first input information according to a preset encoding rule, and obtaining an encoding corresponding to the first input information, specifically: when the obtained first input information is a user on a virtual keyboard of the terminal When the first touch trajectory is combined, the characters corresponding to the keys sequentially passed by the first touch trajectory on the virtual keyboard are combined to obtain an encoding corresponding to the first input information; and the obtained first input information is the first In the case of the sound information, the first sound information is encoded by a hash algorithm, a hidden Markov model HMM algorithm, or a dynamic time warping DTW algorithm to obtain a code corresponding to the first input information.
  • the processor 701 runs the program corresponding to the executable program code by reading the executable program code stored in the memory 702, and is further configured to:
  • the vocabulary processing request includes a vocabulary processing mode and a vocabulary to be added; the vocabulary processing mode is a gesture vocabulary processing mode or a speech vocabulary processing mode.
  • the aforementioned program can be stored in a computer readable In the storage medium.
  • the steps including the foregoing method embodiments are performed; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

Abstract

本发明实施例提供一种输入处理方法和装置,输入处理方法包括:终端根据用户选择的输入模式;获取所述用户输入的与所述输入模式匹配的第一输入信息;所述输入模式为手势输入模式或语音输入模式;根据预先设定的编码规则,对所述第一输入信息进行编码处理,得到与所述第一输入信息对应的编码;查询预先建立的词汇库,获取与所述编码对应的候选词汇。本发明实施例提供的输入处理方法,通过根据用户选择的输入模式获取第一输入信息,并根据该第一输入信息得到其对应的编码,再通过编码获取用户需要输入的词汇,从而实现了复杂词汇的快速准确的输入,有效地提高了输入效率。

Description

输入处理方法和装置
技术领域
本发明涉及通信技术, 尤其涉及一种输入处理方法和装置。 背景技术
当前智能手机、平板电脑等智能设备日渐普及,这些设备大都釆用触摸屏, 需要通过虚拟键盘进行输入操作。 由于受到屏幕面积的限制, 用户无法用十个 手指头同时进行输入操作, 因此, 通过虚拟键盘进行输入的效率比通过物理键 盘输入的效率相差较远。 特别是在特殊场景下 (如开车或坐在晃动的公交车 上), 针对大小写字母、 数字、 符号混合输入的情况, 如用户要输入自己的邮 箱 (如 Jackyl2345@huawei.com ) , 需要来回在数字、 字母、 标点符号等输入 方式之间不断切换,才能完成完整的输入。不仅输入效率低,且容易按错按钮, 输入准确率也难以保证。 为了解决上述问题,现有技术中智能手机、平板电脑等设备主要通过釆用 语音输入和 /或手写输入等方式来提高用户的输入效率。 但是, 语音输入需要 网络支撑才能完成,且只能识别特定语言的标准发音,对不同口音识别率不高; 对于手写输入来说, 手写的字迹需要符合一定规范, 对于笔画较多的字, 还需 要二次挑选, 因此, 釆用上述方式仍然存在输入效率较低的问题。 发明内容
本发明实施例提供了一种输入处理方法和装置,用以实现复杂词汇的快速 准确的输入, 有效地提高输入效率。 本发明第一方面提供一种输入处理方法, 包括:
终端根据用户选择的输入模式 ,获取所述用户输入的与所述输入模式匹配 的第一输入信息; 所述输入模式为手势输入模式或语音输入模式;
所述终端根据预先设定的编码规则, 对所述第一输入信息进行编码处理, 得到与所述第一输入信息对应的编码;
所述终端查询预先建立的词汇库, 获取与所述编码对应的候选词汇。 在第一方面的第一种可能实现方式中,所述获取用户输入的与所述输入模 式匹配的第一输入信息, 包括:
当所述输入模式为手势输入模式时,获取用户在所述终端的虚拟键盘上输 入的第一触摸轨迹信息; 或者, 当所述输入模式为语音输入模式时, 通过所述 终端的声音传感器釆集得到第一声音信息。
则所述根据预先设定的编码规则,对所述第一输入信息进行编码处理,得 到与所述第一输入信息对应的编码, 包括:
当获取的所述第一输入信息为用户在所述终端的虚拟键盘上的第一触摸 轨迹时,将所述第一触摸轨迹在所述虚拟键盘上依次经过的按键所对应的字符 进行组合, 得到与所述第一输入信息对应的编码;
当获取的所述第一输入信息为第一声音信息时, 釆用哈希算法、 隐马尔可 夫模型 HMM算法、 或动态时间规整 DTW算法, 对所述第一声音信息进行编 码处理, 得到与所述第一输入信息对应的编码。
结合第一方面或者第一方面的第一种可能实现方式,在第一方面的第二种 可能实现方式中, 该方法还包括:
所述终端接收词汇处理请求,所述词汇处理请求用于请求添加词汇到所述 词汇库; 所述词汇处理请求包含词汇处理模式和待添加词汇; 所述词汇处理模 式为手势词汇处理模式或语音词汇处理模式;
所述终端根据所述词汇处理请求,接收所述用户输入的与词汇处理模式匹 配的第二输入信息; 并根据所述预先设定的编码规则,对所述第二输入信息进 行编码, 生成与所述第二输入信息对应的编码;
所述终端建立所述第二输入信息对应的编码和所述待添加词汇之间的映 射关系, 并将所述映射关系存储到所述词汇库中;
结合第一方面的第二种可能实现方式,在第一方面的第三种可能实现方式 中, 所述接收用户输入的与所述词汇处理模式对应的第二输入信息, 包括: 当所述词汇处理模式为手势词汇处理模式时,获取用户在所述终端的虚拟 键盘上输入的第二触摸轨迹信息; 或者,
当所述词汇处理模式为语音词汇处理模式时,通过所述终端的声音传感器 釆集得到第二声音信息。
相应地,当获取的所述第二输入信息为用户在所述终端的虚拟键盘上的第 二触摸轨迹时,将所述第二触摸轨迹在所述虚拟键盘上依次经过的按键所对应 的字符进行组合, 得到与所述第二输入信息对应的编码;
当获取的所述第二输入信息为所述第二声音信息时, 釆用哈希算法、 隐马 尔可夫模型 HMM算法、 或动态时间规整 DTW算法, 对所述第二声音信息进 行编码处理, 得到与所述第二输入信息对应的编码。
结合第一方面, 第一方面的第一至第三种可能的实现方式中, 所述查询预 先建立的词汇库, 获取与所述编码对应的候选词汇之后, 所述方法还包括: 当获取到的所述编码对应的候选词汇的数量为多个时,按照词频由大到小 的顺序,对所述多个候选词汇进行排列, 并将排列后的多个候选词汇显示到所 述终端的显示屏上以供用户选择;
将用户从所述排列后的多个候选词汇中选择的词汇,显示在所述终端的显 示屏上的特定区域。
本发明第二方面提供一种输入处理装置, 包括:
信息获取模块, 用于根据用户选择的输入模式, 获取所述用户输入的与所 述输入模式匹配的第一输入信息;所述输入模式为手势输入模式或语音输入模 式;
编码处理模块, 用于根据预先设定的编码规则,对所述第一输入信息进行 编码处理, 得到与所述第一输入信息对应的编码;
查询获取模块, 用于查询预先建立的词汇库, 获取与所述编码对应的候选 词汇。
在第二方面的第一种可能实现方式中,所述信息获取模块具体用于当所述 输入模式为手势输入模式时,获取用户在所述输入处理装置的虚拟键盘上的第 一触摸轨迹信息; 或者,
所述信息获取模块具体用于当所述输入模式为语音输入模式时,通过所述 输入处理装置的声音传感器釆集得到第一声音信息;
则所述编码处理模块具体用于当所述信息获取模块获取到的所述第一输 入信息为用户在所述输入处理装置的虚拟键盘上的第一触摸轨迹时,将所述第 一触摸轨迹在所述虚拟键盘上依次经过的按键所对应的字符进行组合,得到与 所述第一输入信息对应的编码; 或者,
所述编码处理模块具体用于当所述信息获取模块获取到的所述第一输入 信息为所述第一声音信息时, 釆用哈希算法、 隐马尔可夫模型 HMM算法、 或 动态时间规整 DTW算法, 对所述第一声音信息进行编码处理, 得到与所述第 一输入信息对应的编码。
结合第二方面或者第二方面的第一种可能实现方式,在第二方面的第二种 可能实现方式中,所述输入处理装置还包括:接收模块和词汇存储模块,其中, 所述接收模块, 用于接收词汇处理请求, 所述词汇处理请求用于请求添加 词汇到所述词汇库; 所述词汇处理请求包含词汇处理模式和待添加词汇; 所述 词汇处理模式为手势词汇处理模式或语音词汇处理模式;
所述信息获取模块还用于根据所述词汇处理请求,接收所述用户输入的与 词汇处理模式匹配的第二输入信息;
所述编码处理模块还用于根据所述预先设定的编码规则,对所述第二输入 信息进行编码处理, 生成与所述第二输入信息对应的编码;
所述词汇存储模块,用于建立所述第二输入信息对应的编码和所述待添加 词汇之间的映射关系, 并将所述映射关系存储到所述词汇库中。
结合第二方面的第二种可能实现方式,在第二方面的第三种可能实现方式 中,所述信息获取模块具体用于,当所述词汇处理模式为手势词汇处理模式时, 获取用户在所述输入处理装置的虚拟键盘上的第二触摸轨迹信息; 或者,
所述信息获取模块具体用于当所述词汇处理模式为语音词汇处理模式时, 根据所述语音词汇处理模式,通过所述输入处理装置的声音传感器釆集得到第 二声音信息。
结合第二方面的第三种可能的实现方式,在第二方面的第四种可能的实现 方式中, 所述编码处理模块具体用于, 当所述信息获取模块获取的所述第二输 入信息为用户在所述输入处理装置的虚拟键盘上的第二触摸轨迹时,将所述第 二触摸轨迹在所述虚拟键盘上依次经过的按键所对应的字符进行组合,得到与 所述第二输入信息对应的编码; 或者,
所述编码处理模块具体用于当所述信息获取模块获取的所述第二输入信 息为所述第二声音信息时, 釆用哈希算法、 隐马尔可夫模型 HMM算法、 或动 态时间规整 DTW算法, 对所述第二声音信息进行编码处理, 得到与所述第二 输入信息对应的编码
结合第二方面, 第二方面的第一至第四种可能的实现方式中,在第二方面 的第五中可能的实现方式中, 所述的输入处理装置还包括:
排序处理模块, 用于当获取到的所述编码对应的候选词汇的数量为多个 时, 按照词频由大到小的顺序, 对所述多个候选词汇进行排列, 并将排列后的 多个候选词汇显示到所述输入处理装置的显示屏上以供用户选择;
显示模块, 显示模块, 用于将用户从所述排列后的多个候选词汇中选择的 词汇, 显示在所述输入处理装置的显示屏上的特定区域。
在本发明实施例中,通过根据用户选择的输入模式, 获取用户输入的与输 入模式匹配的第一输入信息, 并根据预先设定的编码规则,对该第一输入信息 进行编码处理,得到与该第一输入信息对应的编码, 最后查询预先建立的词汇 库, 获取与该编码对应的候选词汇, 根据不同的输入模式获取第一输入信息, 并根据该第一输入信息获取其对应的编码,再通过编码获取用户需要输入的词 汇, 从而实现了复杂词汇的快速准确的输入, 有效地提高了输入效率, 提升了 用户体验。 附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施 例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地, 下面描述 中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲, 在不付 出创造性劳动的前提下, 还可以根据这些附图获得其他的附图。
图 1为本发明输入处理方法的一个实施例的流程图;
图 2为本发明输入处理方法的又一个实施例的流程图;
图 3为本发明输入处理方法的还一个实施例的流程图;
图 4为终端设备的输入界面的一个示意图;
图 5为终端设备的输入界面的另一个示意图;
图 6为本发明输入处理方法的再一个实施例的流程图;
图 7为终端设备的输入界面的还一个示意图;
图 8为本发明的输入处理装置的一个实施例的结构示意图;
图 9为本发明的输入处理装置的又一个实施例的结构示意图;
图 10为本发明的输入处理装置的另一个实施例的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚, 下面将结合本发明 实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然, 所描述的实施例是本发明一部分实施例, 而不是全部的实施例。基于本发明中 的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其 他实施例, 都属于本发明保护的范围。 图 1为本发明输入处理方法的一个实施例的流程图,如图 1所示, 本实施 例提供的输入处理方法的执行主体为终端 ,则该输入处理方法具体包括以下步 骤: 步骤 101、 终端根据用户选择的输入模式; 获取用户输入的与输入模式匹 配的第一输入信息, 该输入模式为手势输入模式或语音输入模式。
在本实施例中该终端可以为手机或者平板电脑等, 具体地, 用户通过触发 终端输入界面上的相应按钮来选择输入模式,优选地, 该输入模式为语音输入 模式或者手势输入模式。 并获取用户输入的与输入模式匹配的第一输入信息, 不同的输入模式对应不同的输入信息, 因此, 第一输入信息根据不同的输入模 式有所不同。 例如在语音输入模式下, 第一输入信息为一声音信息, 在手势输 入模式下, 第一输入信息为一触摸轨迹信息, 该触摸轨迹信息例如可以是文字 图形信息。
步骤 102、 终端根据预先设定的编码规则, 对该第一输入信息进行编码处 理, 得到与该第一输入信息对应的编码。
在本实施例中,在获得用户输入的与输入模式匹配的第一输入信息后,根 据预先设定的编码规则,对该第一输入信息进行编码处理,得到该第一输入信 息对应的编码, 对于不同的输入模式, 对应设定不同的编码规则, 生成相应的 编码, 具体来说, 手势输入模式预先设定有编码规则, 语音输入模式也预先设 定有编码规则。
步骤 103、 终端查询预先建立的词汇库, 获取与该编码对应的候选词汇。 在本实施例中,在获得该第一输入信息对应的编码后, 查找预先建立的词 汇库, 获取与该编码对应的候选词汇。 具体的, 该预先建立的词汇库可以为该 终端所连接的服务器上的词汇库, 或者该终端自身建立的本地词汇库, 或者该 终端从服务器上的词汇库下载并保存到自身存储器的本地词汇库;该词汇库中 可以包括热门词汇、 专有词汇、 用户自定义的词汇等, 并且建立了编码与上述 词汇的映射关系,以使得能够根据编码快速的在本地词汇库中索引到对应的候 选词汇。 另外, 候选词汇的形式具体可以为如下一种或者几种的组合: 单个词 语、 短语、 句子、 长串的数字和字母等字符; 例如: QQ号码 1234657; 邮箱 yagneos235 @hotmail. com.cn。
优选地, 当获取到的该编码对应的候选词汇为一个时, 则该一个候选词汇 即为用户需要输入的词汇。 当获取到的该编码对应的候选词汇的数量为多个 时, 可以对多个候选词汇按照词频由大到小的顺序, 进行排序处理。 再将排序 处理后的多个候选词汇显示到终端的显示屏上以供用户选择,并将用户从排列 后的多个候选词汇中选择的词汇, 显示在终端的显示屏上的特定区域。
在本实施例中,通过根据用户选择的输入模式, 获取用户输入的与输入模 式匹配的第一输入信息, 并根据预先设定的编码规则,对该第一输入信息进行 编码处理, 得到与该第一输入信息对应的编码, 最后查询预先建立的词汇库, 获取与该编码对应的候选词汇,根据不同的输入模式获取第一输入信息, 并根 据该第一输入信息得到其对应的编码, 再通过编码获取用户需要输入的词汇 , 从而实现了复杂词汇的快速准确的输入, 有效地提高了输入效率。
进一步的,在本发明的另一个实施例中,在上述图 1所示实施例的基础上, 步骤 101 中的获取用户输入的与该输入模式匹配的第一输入信息的具体实现 方式可以有如下几种:
第一种: 当该输入模式为手势输入模式时, 获取用户在所述终端的虚拟键 盘上输入的第一触摸轨迹;
第二种: 当该输入模式为语音输入模式时,通过终端的声音传感器釆集得 到第一声音信息。 在本实施例中, 需要在服务器上预先建立一个词汇库, 该词汇库中包括有 热门词汇、 专有词汇、 用户定制词汇等。 其中, 该用户定制词汇为用户预先自 定义, 并通过终端设备备份到词汇库中的词汇, 需要将该用户定制词汇和用户 个人账号关联, 例如: 词汇库中存储有用户标识和所述用户标识对应的用户定 制词汇。 服务器对收集到的词汇进行索引排序, 如按照汉字拼音、 拼音字母、 笔画及出现频率等进行索引。 为了使词汇库提供更加丰富的词汇, 可通过专门 的维护人员及时的搜集网络上的常用语、 出现频率高的词汇, 并及时的添加到 词汇库中。 另外, 为了方便用户使用, 用户可以通过终端设备将词汇库下载都 手机、 平板电脑等终端设备上的本地词汇库中, 需要说明的是, 该本地词汇库 中的用户定制词汇是与所述用户的标识相对应的。同时为了满足不同用户的需 求, 该本地词汇库还可以具有强大的自定义功能。 用户可以将一些自己常用的 词汇添加到本地词汇库中单独的文件中, 并将该文件上传到服务器,服务器将 该文件中的词汇写入到用户定制词汇中, 以进行备份处理。
还需要说明的是,服务器还可以对多个用户标识对应的用户定制词汇进行 检索, 按照词频由大到小的顺序建立索引, 以生成热门词汇添加到词汇库中。
在用户使用的过程中, 终端设备可以优先下载用户账号关联的词汇, 即用 户自定义的词汇到本地词汇库中。对于专业词汇和热门词汇等,服务器会提示 用户有更新, 用户选择更新, 将专业词汇和热门词汇下载到本地词汇库中。
在本实施例中, 该本地词汇库具有强大的自定义功能, 用户可以通过自定 义功能将自己常用的词汇添加到词汇中, 以方便用户后续的使用。 以下将具体 说明用户如何自定义词汇。
在本发明的又一个实施例中,如图 2所示, 图 2为本发明输入处理方法的 又一个实施例的流程图, 本实施例在上述图 1所示实施例的基础上, 步骤 101 之前, 该方法还可以包括如下步骤:
步骤 201、 终端接收词汇处理请求, 该词汇处理请求用于请求添加词汇到 词汇库; 该词汇处理请求包含词汇处理模式和待添加词汇; 该词汇处理模式为 手势词汇处理模式或语音词汇处理模式。
在本实施例中, 该词汇处理模式为手势词汇处理模式或语音词汇处理模 式, 该词汇处理请求用于请求添加词汇到词汇库。 用户可以通过触发输入界面 上相应的操作按钮来选择词汇处理模式,并在该词汇处理模式下输入待添加词 汇, 待添加词汇为本地词汇库中不存在的词汇, 是用户自定义的词汇, 待添加 词汇可以为用户常用的 QQ号、 邮箱、 密码及电话号码等, 通过词汇处理请求 将用户常用的词汇添加到词汇库中。
步骤 202、 终端根据词汇处理请求, 接收用户输入的与该词汇处理模式匹 配的第二输入信息; 并根据预先设定的编码规则,对该第二输入信息进行编码 处理, 生成与该第二输入信息对应的编码。
本实施例中,根据词汇处理请求,接收用户输入的与词汇处理模式匹配的 第二输入信息, 对于不同的词汇处理模式, 第二输入信息会有所不同。 在手势 词汇处理模式下, 该第二输入信息为用户在终端的虚拟键盘上的触摸轨迹,在 语音输入模式下, 该第二输入信息为终端的声音传感器釆集得到的声音信息。 另外, 根据预先设定编码规则, 对接收的第二输入信息进行编码处理, 生成与 该第二输入信息对应的编码, 该编码唯一标识该第二输入信息。对于不同的输 入模式, 釆用不同的编码方式。 其中, 手势词汇处理模式预先设定的编码规则 为用户在终端的虚拟键盘上的第一触摸轨迹依次经过的键盘所对应的字符的 组合; 语音词汇处理模式预先设定的编码规则为哈希(Hash )算法、 隐马尔可 夫模型( Hidden Markov models , 简称 ΗΜΜ )算法、或动态时间规整( Dynamic Time Warping, 简称 DTW ) 算法。
步骤 203、 终端建立该第二输入信息对应的编码和该待添加词汇之间的映 射关系, 并将该映射关系存储到汇库中。
在本实施例中,根据第二输入信息对应的编码和词汇, 建立该第二输入信 息对应的编码与待添加词汇的之间的映射关系,并将该映射关系存储到词汇库 中。 这样, 便将用户输入信息和词汇对应起来。
在本实施例中,将用户添加的词汇保存到词汇库中, 并且建立用户输入信 息和词汇的映射关系, 当用户输入再次输入信息时, 可以识别用户的输入信息 并获取该输入信息对应的编码,根据本地词汇库中存储的该编码和词汇的映射 关系, 快速的索引到用户需要的候选词汇。
图 3为本发明输入处理方法的还一个实施例的流程图,在本实施例中, 以 所述方法的执行主体为终端设备, 例如手机、 输入模式为手势输入处理模式, 且第一输入信息为用户输入的触摸轨迹信息为例,详细介绍本实施例的技术方 案。 如图 3所示, 该方法包括:
步骤 301、 接收词汇处理请求, 该词汇处理请求包含手势词汇处理模式和 待添力 p词汇, 该词汇处理请求用于添加词汇到词汇库。
在本实施例中,词汇处理请求是用户通过终端设备的输入界面输入的。 图 4为终端设备的输入界面的一个示意图, 图 5为终端设备的输入界面的另一个 示意图, 首先, 用户在输入界面上打开增加自定义词汇的页面, 在该页面上有 相应的按钮, 分别代表不同的词汇处理模式, 如图 4所示, 有手势词汇处理模 式和语音词汇处理模式, 用户选择手势词汇处理模式, 则出现如图 5所示的页 面, 在手势词汇处理模式下, 用户输入待添加词汇 12345678 , 并对该词汇设 置备注信息, 备注为 QQ号码。
步骤 302、 根据词汇处理请求, 接收用户输入的与该手势词汇处理模式匹 配的第二触摸轨迹; 并根据预先设定的编码规则,对该第二触摸轨迹进行编码 处理, 生成与该第二触摸轨迹对应的编码。
本实施例中,在手势词汇处理模式下, 用户通过在终端的虚拟键盘上滑动 产生第二触摸轨迹, 当获取到用户在终端的虚拟键盘上的第二触摸轨迹时,根 据预先设定的编码规则,对该第二触摸轨迹进行编码处理, 生成与该第二触摸 轨迹对应的编码。 该编码规则是预先设定的, 本实施例中, 将第二触摸轨迹在 虚拟键盘上依次经过的按键所对应的字符进行组合作为编码,得到与第二触摸 轨迹对应的编码。 该字符为数字或字母, 因此可将第二触摸轨迹依次经过的按 键所对应的数字作为编码,或将第二触摸轨迹依次经过的按键所对应的字母作 为编码。
如图 5所示,该第二触摸轨迹信息为用户绘制的一 L型图形,该图形由用 户一笔绘制成, 该 L型图形的触摸轨迹依次经过键盘上的数字键 1478, 轨迹 的起点为 1 , 终点为 8, 触摸轨迹的终点标志着该第二触摸轨迹的结束。 本实 施例中,将第二触摸轨迹在虚拟键盘上依次经过的按键所对应的字符进行组合 作为编码规则, 参照 图 5 , 输入界面上的虚拟键盘是常规的数字键盘, 该数 字键盘上有 0~9十个数字,每个数字对应一个按钮,用户可以在该数字键盘上 任意绘制图形,以用户绘制图形时的触摸轨迹依次经过的按键所对应的数字进 行编码。 因此, 对于本实施例来说, 该第二触摸轨迹对应的编码为 1478。 当 输入界面为 PC机的模拟键盘时, 可以根据触摸轨迹经过的按钮对应的字母进 行编码。
步骤 303、 建立该第二触摸轨迹对应的编码和该待添加词汇之间的映射关 系, 并将该映射关系存储到预先建立的词汇库中。
具体为, 在步骤 302中生成了第二触摸轨迹对应的编码为 1478, 将该编 码 1478和步骤 201 中输入的待添加词汇建立映射关系, 该待添加词汇为 QQ 号 12345678, 即该编码 1478唯一标识该词汇 12345678, 并将该映射关系存储 到词汇库中。 本实施例中, 用户的一条触摸轨迹对应唯一的编码, 但是, 需要 说明的是, 一个编码还可以对应多个词汇, 如编码 1478还可以对应其他的词 汇, 例如: 当用户有多个 QQ号时, 编码 1478可以和多个 QQ号码对应, 只 需增加编码 1478和词汇的映射关系即可。
另外,举例来说, 该词汇库中还可以存储如下形式的编码和词汇的映射关 系, 如表一所示:
Figure imgf000015_0001
Figure imgf000015_0002
步骤 304、 根据用户选择的手势输入模式, 获取与手势输入模式匹配的第 一触摸轨迹。
步骤 305、 根据预先设定的编码规则, 对该第一触摸轨迹进行编码处理, 获取与该第一触摸轨迹对应的编码。 其中, 本步骤中, 该设定的编码规则和步骤 302中对第二触摸轨迹进行编 码所釆用的编码规则是相同的,具体为将第一触摸轨迹在虚拟键盘上依次经过 的按键所对应的字符进行组合, 得到与该第一输入信息对应的编码。
步骤 306、 查询预先建立的词汇库, 获取与该编码对应的候选词汇。
本步骤中, 由于预先建立的词汇库中已经存储有该编码和词汇的映射关 系 ,根据获取的编码在词汇库中进行查找,即可获得与该编码对应的候选词汇。
本实施例中, 当用户需要输入自己的 QQ号时, 可以将终端设备上的输入 界面切换至手势输入模式,在手势输入模式下, 用户通过输入界面一笔绘制一 图形, 以触发终端设备根据该手势输入模式, 获取该触摸轨迹信息, 并对该触 摸轨迹信息进行识别, 获取其对应的编码, 例如: 识别用户触摸轨迹经过的区 域为 1478, 则编码为 1478。 则查询本地词汇库, 获取与该编码为 1478对应的 候选词汇, 即 QQ号码 12345678。
另夕卜,还需要说明的是,如果该候选词汇的数量是一个,则该候选词汇(即 QQ号码 12345678 )为用户需要输入的词汇; 如果该候选词汇的数量是多个, 则可以将多个候选词汇按照词频由大到小的顺序, 对多个候选词汇进行排序, 并将排序后的多个候选词汇显示到所述终端的显示屏上,以供用户选择其中一 个候选词汇作为用户需要输入的词汇, 最后, 该终端将用户从排列后的多个候 选词汇中选择的词汇, 显示在该终端的显示屏上的特定区域, 其中, 特定区域 是指用户当前使用的应用程序或等待文本输入的应用程序所指定的输入区域。
本实施例提供的输入处理方法,通过预先在本地词汇库中存储用户绘制简 单图形的触摸轨迹、触摸轨迹对应的编码及该编码和词汇的映射关系, 建立繁 瑣的词汇与简单图形的对应关系, 该词汇为用户常用的一些繁瑣词汇, 如用户 个人的 Q Q账号、 邮箱、 密码等。 使用时, 用户只需将输入模式切换至手势输 入模式,在输入界面一笔绘制出图形,设备识别图形的触摸轨迹信息获得触摸 轨迹信息对应的编码,根据该编码与词汇的映射关系在词汇库中查找到相应的 候选词汇, 并将该候选词汇作为用户需要输入的词汇,从而使得用户只需一次 输入便可完成较长的复杂词汇的输入, 提高了输入效率和准确率。
本实施例提供的输入处理方法, 适用于复杂繁瑣词汇的输入, 和传统的手 写输入相比, 实施例提供的输入处理方法, 不做绘制的图形和文字的识别, 而 是将绘制图形经过的触摸轨迹通过编码和词汇建立对应关系。举例来说, 当输 入 L型的图形时, 并不需要识别该图形是否为字母 L, 而是根据图形的触摸轨 迹获得相应编码, 根据编码查找对应的词汇, 不仅提高了输入效率, 还提高了 输入的准确率。
图 6为本发明输入处理方法的再一个实施例的流程图,在本实施例中, 以 所述方法的执行主体为终端设备, 例如手机、 输入模式为语音输入处理模式, 且第一输入信息为用户输入的声音信息为例, 详细介绍本实施例的技术方案。 如图 6所示, 该方法包括;
步骤 601、 接收词汇处理请求, 该词汇处理请求包含语音词汇处理模式和 待添力 p词汇; 该词汇处理请求用于添加词汇到词汇库。
在本实施例中,词汇处理请求是用户通过终端设备的输入界面输入的。 图 7为终端设备的输入界面的还一个示意图, 首先用户在输入界面上打开增加自 定义词汇的页面, 在该页面上设置有相应的按钮, 分别代表不同的词汇处理模 式, 如图 4所示, 有手势词汇处理模式和语音词汇处理模式, 用户选择语音词 汇处理模式, 则出现如图 7所示的页面, 在语音词汇处理模式下, 用户输入待 添加词汇 , 并对该待添加词汇设置备注信息, 备注为 163邮箱。
步骤 602、 根据词汇处理请求, 接收用户输入的与该语音词汇处理模式匹 配的第二声音信息; 并根据预先设定的编码规则,对该第二声音信息进行编码 处理, 生成与该第二声音信息对应的编码。
本实施例中,根据词汇处理请求,接收用户输入的与该语音词汇处理模式 匹配的第二声音信息,该第二声音信息是终端通过其上的声音传感器釆集得到 的。 接收到该第二声音信息后, 根据预先设定的编码规则, 对该第二声音信息 进行编码处理, 生成与该第二声音信息对应的编码。 具体可釆用哈希算法、 隐 马尔可夫模型 ΗΜΜ算法、 或动态时间规整 DTW算法, 对该第二声音信息进 行编码处理, 得到与该第二输入信息对应的编码。
以下将通过具体例子来说明, 如图 7所示, 在语音词汇处理模式下, 终端 设备的输入界面上可以设置有一启动发声的按钮(同时作为结束声音按钮) , 通过按住启动发声按钮, 打开终端设备上的话筒或其他语音输入设备, 用户发 声, 例如, 用户发出邮箱的声音信息, 终端设备开始釆集用户的声音信息, 用 户松开启动发声按钮, 完成一次声音信息釆集, 釆集得到的声音信息即为第二 声音信息。
在釆集得到第二声音信息后,对该第二声音信息进行编码, 生成该声音信 息对应的编码。 其中, 该编码规则是预先设定的, 该编码规则可以釆用现有技 术中常用的语音识别算法, 如哈希算法、 隐马尔可夫模型 ΗΜΜ算法、 动态时 间规整 DTW算法等, 将釆集到的声音信息转换为字符串或文本, 本发明实施 例对于编码规则不作特别限制。通过对用户输入的第二声音信息进行编码,得 到该第二声音信息对应的编码。 本实施例中, 例如第二声音信息对应的编码为 Sjdegsogea4512。
步骤 603、 建立该第二声音信息对应的编码和该待添加词汇之间的映射关 系, 并将该映射关系存储到词汇库中。
具体为, 在步骤 602中生成了第二声音信息对应的编码 Sjdegsogea4512 , 将该 编码 Sjdegsogea4512 和 步骤 601 中 输入的 待添加词 汇 Jackv.Chen@163.com建立映射关系 , 该编码 Sjdegsogea4512唯一标识该词汇 Jackv.Chen@163.com , 并将该映射关系存储到词汇库中。 本实施例中, 用户的 一条声音信息对应唯一的编码, 需要说明的是, 一个编码可以对应多个词汇, 如用户有多个邮箱号, 为了记忆方便, 用户将声音信息邮箱对应的编码和多个 邮箱号建立对应关系,只需在本地词汇库中增加该编码和对应词汇的映射关系 即可。
举例来说, 该本地词汇库中还可以存储如下形式的编码和词汇的映射关 系, 如表二所示:
Figure imgf000019_0001
Figure imgf000019_0002
其中, 该音频 1、 音频 2和音频 3中记录用户输入的声音信息。
步骤 604、 根据用户选择的语音输入模式, 获取与该语音输入模式匹配的 第一声音信息。
在语音输入模式下, 该第一声音信息由终端通过传感器釆集得到。 步骤 605、 根据预先设定的编码规则, 对该第一声音信息进行编码处理, 获取与该第一声音信息对应的编码。
本步骤中釆用的编码规则和步骤 602中釆用的编码规则是相同的,具体可 釆用哈希算法、 隐马尔可夫模型 HMM算法、 或动态时间规整 DTW算法, 对 该第一声音信息进行编码处理, 得到与该第一输入信息对应的编码。
步骤 606、 查询预先建立的词汇库, 获取与该编码对应的候选词汇。
本步骤中, 由于预先建立的词汇库中已经存储有该编码和词汇的映射关 系 ,根据获取的编码在词汇库中进行查找,即可获得与该编码对应的候选词汇。
本实施例中, 当用户需要输入自己邮箱号时,先将终端设备的输入界面切 换至语音输入模式,在语音输入模式下, 用户通过按住输入界面上的启动发声 按钮,发出声音信息,以触发终端设备根据该语音输入模式,获取该声音信息, 并对该声音信息进行识别, 获取与该声音信息对应的编码, 例如, 识别用户输 入声音信息对应的编码为 Sjdegsogea4512, 则查询本地词汇库, 获取与该编码 Sjdegsogea4512对应的 4昊选词汇, 即 Jackv.Chen(¾ 163. com。
另外,还需要说明的是,如果该候选词汇的数量是一个,则该候选词汇(即
Jackv.Chen@163.com )为用户需要输入的词汇;如果该候选词汇的数量是多个, 则可以将多个候选词汇按照词频由大到小的顺序, 对多个候选词汇进行排序, 并将排序后的多个候选词汇显示给用户,以供用户选择其中一个候选词汇作为 用户需要输入的词汇,用户从该编码对应的排序后的多个候选词汇中选择需要 输入的词汇输入。
本实施例提供的输入处理方法,通过预先在本地词汇库中存储用户简短声 音信息、声音信息对应的编码及该编码和词汇的映射关系, 建立繁瑣词汇与声 音信息的对应关系, 该词汇为用户常用的一些繁瑣词汇, 如用户邮箱、 密码、 手机号等。 使用时, 用户只需将输入模式切换至语音输入模式, 进行相应的操 作, 输入用户的声音信息, 设备识别声音信息并获取该声音信息对应的编码, 根据该编码与词汇的映射关系, 在本地词汇库中查找到相应的候选词汇。 只需 输入简短声音信息便可完成复杂较长词汇的输入, 提高了输入效率。
本实施例提供的输入处理方法, 适用于晃动, 不易手写的操作场景, 如用 户在开车过程中需要进行输入。在将输入模式切换至语音输入模式下, 用户只 需录入简短的声音信息, 便可得完成复杂词汇的输入, 方便且输入效率高。 和 传统语音输入不同的是, 本实施例提供的输入处理方法, 不做声音和文字的识 另 只是将输入的声音和词汇对应, 提高了输入的准确率。 举例来说, 当用户 输入声音信息为我的邮箱时, 程序并不对声音信息和文字 "我的邮箱"进行识 另1 J , 并不需要向用户返回文字 "我的邮箱" , 而是根据声音信号对应的编码和 词汇的映射关系返回用户的邮箱号。
另外, 还需要说明的是, 本发明还可以釆用字符输入模式, 获取用户需要 输入的词汇。 具体的, 用户可以将输入界面切换至字符输入模式, 终端设备在 在该模式下获取用户输入字符信息,并从本地词汇库中索引检索与该字符信息 对应的候选词汇,并将检索到的该字符信息对应的多个候选词汇按照词汇频率 由大到小的顺序进行排序,并将排序后的多个候选词汇通过终端设备的界面显 示给用户。 用户从返回的过个候选词汇中选择需要输入的词汇, 当终端设备获 取到该用户选择的需要输入的词汇后,还可以根据该词汇的上下文,在本地词 汇库中进行二次索引, 预测出用户下一步可能输入的词汇, 进行二次索引后, 用户可以直接从二次索引后获取的候选词汇中选择需要输入的词汇,而不需要 再次输入字符。
举例来说, 用户通过拼音输入 "ao" , 通过 "ao" 在本地词汇库中检索到 与之对应的多个候选词汇(字): 奥、 熬、 澳、 傲等, 这些候选词汇可以是根 据词频由大到小的顺序排序输出的, 用户选择候选词汇 "奥"作为用户需要输 入的词汇, 则还可以根据 "奥"在本地词汇库中进行二次索引, 从本地词汇库 中获取与 "奥" 关联出多个候选词汇, 如运、 巴、 林、 美等, 以供用户选择, 当获取到用户选择第二个字为 "运" , 此时上下文为奥运, 则可以进一步根据 "奥运" , 关联出多个候选词汇, 如奥运会、 奥运村等, 最后获取用户选择的 "奥运会" 作为用户需要输入的词汇。
图 8为本发明的输入处理装置的一个实施例的结构示意图, 如图 8所示, 本实施例提供的输入处理装置 800包括,信息获取模块 801、编码处理模块 802 及查询获取模块 803。其中,信息获取模块 801用于根据用户选择的输入模式, 获取用户输入的与该输入模式匹配的第一输入信息;该输入模式为手势输入模 式或语音输入模式。编码处理模块 802用于根据预先设定的编码规则,对该第 一输入信息进行编码处理,得到与该第一输入信息对应的编码。 查询获取模块 803用于查询预先建立的词汇库, 获取与该编码对应的候选词汇。
本实施例的输入处理装置 800可以执行图 1所示方法实施例的技术方案, 其实现原理相类似, 此处不再赘述。
在本实施例中,根据用户选择的输入模式, 获取与输入模式匹配的第一输 入信息, 并根据预先设定的编码规则对该第一输入信息进行编码处理,得到与 该第一输入信息对应的编码, 最后查询预先建立的词汇库, 获取与该编码对应 的候选词汇,根据不同的输入模式获取第一输入信息, 并根据该第一输入信息 获取其对应的编码,再通过编码获取用户需要输入的词汇,从而实现了复杂词 汇的快速准确的输入, 有效地提高了输入效率。
进一步的 ,在本发明的另一个实施例中 ,在上述图 8所示实施例的基础上, 所述输入处理装置还包括: 触摸屏和声音传感器, 且所述触摸屏上显示有虚拟 键盘, 以供用户进行触摸操作, 所述声音传感器用于釆集声音; 相应地, 该信 息获取模块 801具体用于当该输入模式为手势输入模式时,获取用户在输入处 理装置的虚拟键盘上的第一触摸轨迹; 编码处理模块 802具体用于, 当信息获 取模块 801 获取到的第一输入信息为用户在输入处理装置的虚拟键盘上的第 一触摸轨迹时,根据预先设定的编码规则,将该第一触摸轨迹在虚拟键盘上依 次经过的按键所对应的字符进行组合, 得到与第一输入信息对应的编码。
该信息获取模块 801具体还用于当该输入模式为语音输入模式时,通过输 入处理装置的声音传感器釆集得到第一声音信息。编码处理模块 802具体用于 当信息获取模块 802获取的第一输入信息为第一声音信息时,根据预先设定的 编码规则, 例如釆用哈希算法、 隐马尔可夫模型 HMM算法、 或动态时间规整 DTW算法, 对该第一声音信息进行编码处理, 得到与该第一输入信息对应的 编码。
查询获取模块 803查询预先建立的词汇库,获取与所述编码对应的候选词 汇之后, 还用于: 当获取到的编码对应的候选词汇的数量为多个时, 按照词频 由大到小的顺序,对该多个候选词汇进行排列, 并将排列后的多个候选词汇显 示给用户。接收用户输入的编码对应的词汇, 编码对应的词汇为用户从排列后 的多个候选词汇中选择的词汇。
图 9为本发明的输入处理装置的又一个实施例的结构示意图,如图 9所示, 本实施例提供的输入处理装置 900包括: 接收模块 901、 信息获取模块 902、 编码处理模块 903、 词汇存储模块 904、 查询获取模块 905、 排序处理模块 906 及显示模块 907。 其中, 接收模块 901用于接收词汇处理请求, 该词汇处理请 求用于请求添加词汇到词汇库中,该词汇处理请求包含词汇处理模式和待添加 词汇, 该词汇处理模式为手势词汇处理模式或语音词汇处理模式; 信息获取模 块 902用于根据词汇处理请求,接收用户输入的与词汇处理模式匹配的第二输 入信息; 编码处理模块 903用于根据预先设定的编码规则,对该第二输入信息 进行编码处理, 生成与该第二输入信息对应的编码; 词汇存储模块 904用于建 立该第二输入信息对应的编码和该待添加词汇之间的映射关系,并将该映射关 系存储到词汇库中。信息获取模块 902还用于根据用户选择的输入模式, 获取 用户输入的与该输入模式匹配的第一输入信息;该输入模式为手势输入模式或 语音输入模式。编码处理模块 903还用于根据预先设定的编码规则,对该第一 输入信息进行编码处理, 得到与该第一输入信息对应的编码; 查询获取模块 905用于查询预先建立的词汇库, 获取与该编码对应的候选词汇。
在本实施例中,接收模块 901接收的词汇处理请求是由用户通过终端输入 界面输入的,词汇处理请求用于请求添加词汇到词汇库, 该词汇处理请求包括 含词汇处理模式和待添加词汇,该词汇处理模式为语音词汇处理模式或者手势 词汇处理模式。待添加的词汇为词汇库中不存在的词汇, 该词汇可以是单个词 语、 短语、 句子、 长串的数字和字母等字符; 例如: QQ号码 8945145182; 邮 箱 yagneos235@hotmail.com.cn。
信息获取模块 902用于根据所述词汇处理请求,接收用户输入的与该词汇 处理模式匹配的第二输入信息。信息获取模块 902具体用于当该词汇处理模式 为手势词汇处理模式时,获取用户在所述输入处理装置的虚拟键盘上的第二触 摸轨迹, 当该词汇处理模式为语音词汇处理模式时,通过输入处理装置的声音 传感器釆集得到第二声音信息。
编码处理模块 903 , 用于根据预先设定的编码规则, 对该第二输入信息进 行编码, 生成与该第二输入信息对应的编码。 在本实施例中, 对于不同的第二 输入信息, 釆用不同的编码方式, 该编码规则是预先定义的。 其中, 编码处理 模块 903具体用于当信息获取模块 902获取的第二输入信息为用户在输入处理 装置的虚拟键盘上的第二触摸轨迹时,将第二触摸轨迹在虚拟键盘上依次经过 的按键所对应的字符进行组合,得到与第二输入信息对应的编码; 当信息获取 模块 902获取的第二输入信息为第二声音信息时, 釆用哈希算法、 隐马尔可夫 模型 HMM算法、或动态时间规整 DTW算法,对第二声音信息进行编码处理, 得到与第二输入信息对应的编码。该第二输入信息对应的编码唯一标识该第二 输入信息。
词汇存储模块 904 用于建立该第二输入信息对应的编码和该待添加词汇 之间的映射关系, 并将该映射关系存储到词汇库中。 本实施例中, 词汇存储模 块 904将编码处理模块 903 生成的第二输入信息对应的编码和接收模块 901 接收到的待添加词汇建立映射关系,并将该映射关系存储到预先设定的词汇库 中。 这样, 便建立了用户输入信息和词汇的映射关系, 并将该映射关系保存到 本地词汇库中。 当用户输入再次输入信息时, 可以识别用户的输入信息并获取 该输入信息对应的编码。
本实施例中,信息获取模块 902还用于根据用户选择的输入模式, 获取用 户输入的与该输入模式匹配的第一输入信息 ,该输入模式为手势输入模式或语 音输入模式。信息获取模块 902具体用于当该输入模式为手势输入模式时, 获 取用户在输入处理装置的虚拟键盘上的第一触摸轨迹;当该输入模式为语音输 入模式时 , 通过输入处理装置的声音传感器釆集得到第一声音信息。
在信息获取模块 902获取到用户输入的第一输入信息之后,编码处理模块 903根据预先设定的编码规则, 对该第一输入信息进行编码处理, 得到与该第 一输入信息对应的编码。编码处理模块 903具体用于当获取的第一输入信息为 用户在输入处理装置的虚拟键盘上的第一触摸轨迹时,将第一触摸轨迹在虚拟 键盘上依次经过的按键所对应的字符进行组合,得到与第一输入信息对应的编 码; 其中, 该字符可以为数字或者字母, 即可以将第一触摸轨迹在虚拟键盘上 依次经过的按键所对应的数字进行组合得到编码,或者将第一触摸轨迹在虚拟 键盘上依次经过的按键所对应的字母进行组合得到编码。
编码处理模块 903还用于当获取的第一输入信息为第一声音信息时,釆用 哈希算法、 隐马尔可夫模型 HMM算法、 或动态时间规整 DTW算法, 对第一 声音信息进行编码处理, 得到与第一输入信息对应的编码。
查询获取模块 905查询预先建立的词汇库, 获取与编码对应的候选词汇。 具体来说,查询获取模块 905根据编码处理模块 903得到的与第一输入信 息对应的编码, 查找预先建立的词汇库, 该词汇库中存储有该编码和对应的词 汇的映射关系, 通过编码与词汇的映射关系, 获取与该编码对应的候选词汇。 另夕卜,还需要说明的是,如果查询获取模块 905获取到的该候选词汇的数量是 一个, 则该候选词汇为用户需要输入的词汇; 如果该候选词汇的数量是多个, 排序处理模块 906按照词频由大到小的顺序,对多个候选词汇进行排列, 并将 排列后的多个候选词汇显示到输入处理装置的显示屏上以供用户选择;显示模 块 907 , 用于将用户从排列后的多个候选词汇中选择的词汇, 显示在输入处理 装置的显示屏上的特定区域。
本实施例提供的输入处理装置,可以用于执行本发明任一方法所示实施例 的技术方案, 其实现原理和技术效果类似, 此处不再赘述。
图 10 为本发明的输入处理装置的另一个实施例的结构示意图。 如图 10 所示, 本实施例的输入处理装置 700包括: 至少一个处理器 701、 存储器 702 和总线。处理器 701和存储器 702通过总线连接并完成相互间的通信。 所述总 线可以是工业标准体系结构 (Industry Standard Architecture, 简称为 ISA ) 总 线、 外部设备互连(Peripheral Component, 简称为 PCI )总线或扩展工业标准 体系结构(Extended Industry Standard Architecture, 简称为 EISA )总线等。 所 述总线可以分为地址总线、 数据总线、 控制总线等。 为便于表示, 图 10中仅 用一条粗线表示, 但并不表示仅有一根总线或一种类型的总线。 其中:
存储器 702用于存储可执行程序代码, 该程序代码包括计算机操作指令。 存储器 702可能包含高速随机存取存储器 RAM ( Random Access Memory ) , 也可能还包括非易失性存储器 (non-volatile memory ) , 例如至少一个磁盘存 储器。
在一个实施例中,处理器 701通过读取存储器 702中存储的可执行程序代 码来运行与该可执行程序代码对应的程序, 以用于:
根据用户选择的输入模式,获取用户输入的与输入模式匹配的第一输入信 息, 该输入模式为手势输入模式或语音输入模式。
根据预先设定的编码规则,对该第一输入信息进行编码处理,得到与该第 一输入信息对应的编码。 查询预先建立的词汇库, 获取与该编码对应的候选词汇。
具体地, 在一个实施例中, 所述输入处理装置还包括: 触摸屏和声音传感 器, 且该触摸屏上显示有虚拟键盘, 以供用户进行触摸操作, 该声音传感器用 于釆集声音; 相应地, 根据预先设定的编码规则, 对该第一输入信息进行编码 处理, 得到与该第一输入信息对应的编码, 具体包括: 当获取的该第一输入信 息为用户在该终端的虚拟键盘上的第一触摸轨迹时,将该第一触摸轨迹在虚拟 键盘上依次经过的按键所对应的字符进行组合,得到与该第一输入信息对应的 编码; 当获取的该第一输入信息为该第一声音信息时, 釆用哈希算法、 隐马尔 可夫模型 HMM算法、 或动态时间规整 DTW算法, 对该第一声音信息进行编 码处理, 得到与该第一输入信息对应的编码。
进一步地, 在一个实施例中, 处理器 701通过读取存储器 702中存储的可 执行程序代码来运行与该可执行程序代码对应的程序, 还用于:
接收词汇处理请求, 该词汇处理请求用于请求添加词汇到该词汇库; 该词 汇处理请求包含词汇处理模式和待添加词汇;该词汇处理模式为手势词汇处理 模式或语音词汇处理模式。
根据该词汇处理请求,接收用户输入的与词汇处理模式匹配的第二输入信 息; 并根据预先设定的编码规则, 对该第二输入信息进行编码, 生成与该第二 输入信息对应的编码;
建立该第二输入信息对应的编码和该待添加词汇之间的映射关系,并将该 映射关系存储到该词汇库中。
本领域普通技术人员可以理解:实现上述各方法实施例的全部或部分步骤 可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取 存储介质中。 该程序在执行时, 执行包括上述各方法实施例的步骤; 而前述的 存储介质包括: ROM, RAM,磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是: 以上各实施例仅用以说明本发明的技术方案, 而非对其 限制; 尽管参照前述各实施例对本发明进行了详细的说明, 本领域的普通技术 人员应当理解: 其依然可以对前述各实施例所记载的技术方案进行修改, 或者 对其中部分或者全部技术特征进行等同替换; 而这些修改或者替换, 并不使相 应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims

权 利 要 求
1、 一种输入处理方法, 其特征在于, 包括:
终端根据用户选择的输入模式 ,获取所述用户输入的与所述输入模式匹配 的第一输入信息; 所述输入模式为手势输入模式或语音输入模式;
所述终端根据预先设定的编码规则, 对所述第一输入信息进行编码处理, 得到与所述第一输入信息对应的编码;
所述终端查询预先建立的词汇库, 获取与所述编码对应的候选词汇。
2、 根据权利要求 1所述的输入处理方法, 其特征在于, 所述获取所述用 户输入的与所述输入模式匹配的第一输入信息, 包括:
当所述输入模式为手势输入模式时,获取用户在所述终端的虚拟键盘上的 第一触摸轨迹; 或者, 当所述输入模式为语音输入模式时, 通过所述终端的声 音传感器釆集得到第一声音信息。
3、 根据权利要求 2所述的输入处理方法, 其特征在于, 所述根据预先设 定的编码规则,对所述第一输入信息进行编码处理,得到与所述第一输入信息 对应的编码, 包括:
当获取的所述第一输入信息为用户在所述终端的虚拟键盘上的第一触摸 轨迹时,将所述第一触摸轨迹在所述虚拟键盘上依次经过的按键所对应的字符 进行组合, 得到与所述第一输入信息对应的编码;
当获取的所述第一输入信息为所述第一声音信息时, 釆用哈希算法、 隐马 尔可夫模型 H固算法、或动态时间规整 DTW算法,对所述第一声音信息进行编 码处理, 得到与所述第一输入信息对应的编码。
4、 根据权利要求 1-3任一所述的输入处理方法, 其特征在于, 还包括: 所述终端接收词汇处理请求 ,所述词汇处理请求用于请求添加词汇到所述 词汇库; 所述词汇处理请求包含词汇处理模式和待添加词汇; 所述词汇处理模 式为手势词汇处理模式或语音词汇处理模式;
所述终端根据所述词汇处理请求,接收所述用户输入的与词汇处理模式匹 配的第二输入信息; 并根据所述预先设定的编码规则,对所述第二输入信息进 行编码, 生成与所述第二输入信息对应的编码;
所述终端建立所述第二输入信息对应的编码和所述待添加词汇之间的映 射关系, 并将所述映射关系存储到所述词汇库中。
5、 根据权利要求 4所述的输入处理方法, 其特征在于, 所述接收用户输 入的与所述词汇处理模式对应的第二输入信息, 包括:
当所述词汇处理模式为手势词汇处理模式时,获取用户在所述终端的虚拟 键盘上的第二触摸轨迹; 或者,
当所述词汇处理模式为语音词汇处理模式时,通过所述终端的声音传感器 釆集得到第二声音信息。
6、 根据权利要求 5所述的输入处理方法, 其特征在于, 根据所述预先设 定的编码规则,对所述第二输入信息进行编码, 生成与所述第二输入信息对应 的编码, 包括:
当获取的所述第二输入信息为用户在所述终端的虚拟键盘上的第二触摸 轨迹时,将所述第二触摸轨迹在所述虚拟键盘上依次经过的按键所对应的字符 进行组合, 得到与所述第二输入信息对应的编码;
当获取的所述第二输入信息为所述第二声音信息时, 釆用哈希算法、 隐马 尔可夫模型 H固算法、或动态时间规整 DTW算法,对所述第二声音信息进行编 码处理, 得到与所述第二输入信息对应的编码。
7、 根据权利要求 1-6任一所述的输入处理方法, 其特征在于, 在所述查 询预先建立的词汇库,获取与所述编码对应的候选词汇之后,所述方法还包括: 当获取到的所述编码对应的候选词汇的数量为多个时,按照词频由大到小 的顺序,对所述多个候选词汇进行排列, 并将排列后的多个候选词汇显示到所 述终端的显示屏上以供用户选择;
将用户从所述排列后的多个候选词汇中选择的词汇,显示在所述终端的显 示屏上的特定区域。
8、 一种输入处理装置, 其特征在于, 包括:
信息获取模块, 用于根据用户选择的输入模式, 获取所述用户输入的与所 述输入模式匹配的第一输入信息;所述输入模式为手势输入模式或语音输入模 式;
编码处理模块, 用于根据预先设定的编码规则,对所述第一输入信息进行 编码处理, 得到与所述第一输入信息对应的编码;
查询获取模块, 用于查询预先建立的词汇库, 获取与所述编码对应的候选 词汇。
9、 根据权利要求 8所述的输入处理装置, 其特征在于, 所述信息获取模 块具体用于当所述输入模式为手势输入模式时,获取用户在所述输入处理装置 的虚拟键盘上的第一触摸轨迹; 或者,
所述信息获取模块具体用于当所述输入模式为语音输入模式时,通过所述 输入处理装置的声音传感器釆集得到第一声音信息。
10、 根据权利要求 9所述的输入处理装置, 其特征在于,
所述编码处理模块具体用于当所述信息获取模块获取到的所述第一输入 信息为所述输入处理装置的虚拟键盘上的第一触摸轨迹时,将所述第一触摸轨 迹在所述虚拟键盘上依次经过的按键所对应的字符进行组合,得到与所述第一 输入信息对应的编码;
或者,
所述编码处理模块具体用于当所述信息获取模块获取到的所述第一输入 信息为所述第一声音信息时, 釆用哈希算法、 隐马尔可夫模型 H固算法、 或动 态时间规整 DTW算法,对所述第一声音信息进行编码处理,得到与所述第一输 入信息对应的编码。
11、 根据权利要求 9所述的输入处理装置, 其特征在于, 所述输入处理装 置还包括: 接收模块和词汇存储模块, 其中,
所述接收模块, 用于接收词汇处理请求, 所述词汇处理请求用于请求添加 词汇到所述词汇库; 所述词汇处理请求包含词汇处理模式和待添加词汇; 所述 词汇处理模式为手势词汇处理模式或语音词汇处理模式;
所述信息获取模块还用于根据所述词汇处理请求,接收所述用户输入的与 所述词汇处理模式匹配的第二输入信息;
所述编码处理模块还用于根据所述预先设定的编码规则,对所述第二输入 信息进行编码处理, 生成与所述第二输入信息对应的编码;
所述词汇存储模块,用于建立所述第二输入信息对应的编码和所述待添加 词汇之间的映射关系, 并将所述映射关系存储到所述词汇库中。
12、 根据权利要求 11所述的输入处理装置, 其特征在于, 所述信息获取 模块具体用于, 当所述词汇处理模式为手势词汇处理模式时, 获取用户在所述 输入处理装置的虚拟键盘上的第二触摸轨迹; 或者, 所述信息获取模块具体用于当所述词汇处理模式为语音词汇处理模式时, 通过所述输入处理装置的声音传感器釆集得到第二声音信息。
1 3、 根据权利要求 12所述的输入处理装置, 其特征在于, 所述编码处理 模块具体用于,当所述信息获取模块获取的所述第二输入信息为用户在所述输 入处理装置的虚拟键盘上的第二触摸轨迹时,将所述第二触摸轨迹在所述虚拟 键盘上依次经过的按键所对应的字符进行组合,得到与所述第二输入信息对应 的编码; 或者,
所述编码处理模块具体用于当所述信息获取模块获取的所述第二输入信 息为所述第二声音信息时, 釆用哈希算法、 隐马尔可夫模型丽 M算法、 或动态 时间规整 DTW算法,对所述第二声音信息进行编码处理,得到与所述第二输入 信息对应的编码。
14、 根据权利要求 8-1 3任一所述的输入处理装置, 其特征在于, 还包括: 排序处理模块, 用于当获取到的所述编码对应的候选词汇的数量为多个 时, 按照词频由大到小的顺序, 对所述多个候选词汇进行排列, 并将排列后的 多个候选词汇显示到所述输入处理装置的显示屏上以供用户选择;
显示模块, 用于将用户从所述排列后的多个候选词汇中选择的词汇,显示 在所述输入处理装置的显示屏上的特定区域。
PCT/CN2013/080480 2012-12-31 2013-07-31 输入处理方法和装置 WO2014101410A1 (zh)

Priority Applications (8)

Application Number Priority Date Filing Date Title
KR1020137033362A KR101586890B1 (ko) 2012-12-31 2013-07-31 입력 처리 방법 및 장치
KR1020167000682A KR20160011230A (ko) 2012-12-31 2013-07-31 입력 처리 방법 및 장치
AU2013270485A AU2013270485C1 (en) 2012-12-31 2013-07-31 Input processing method and apparatus
EP13795149.7A EP2765473A4 (en) 2012-12-31 2013-07-31 INPUT PROCESSING AND DEVICE
RU2014129212/08A RU2589873C2 (ru) 2012-12-31 2013-07-31 Способ и устройство обработки ввода
JP2014553610A JP5860171B2 (ja) 2012-12-31 2013-07-31 入力処理方法及び装置
BR112013034056A BR112013034056A2 (pt) 2012-12-31 2013-12-03 método e aparelho de processamento de entradas
US14/097,467 US20140184514A1 (en) 2012-12-31 2013-12-05 Input processing method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210590206.5 2012-12-31
CN201210590206.5A CN103064530B (zh) 2012-12-31 2012-12-31 输入处理方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/097,467 Continuation US20140184514A1 (en) 2012-12-31 2013-12-05 Input processing method and apparatus

Publications (1)

Publication Number Publication Date
WO2014101410A1 true WO2014101410A1 (zh) 2014-07-03

Family

ID=48107181

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/080480 WO2014101410A1 (zh) 2012-12-31 2013-07-31 输入处理方法和装置

Country Status (10)

Country Link
US (1) US20140184514A1 (zh)
EP (1) EP2765473A4 (zh)
JP (1) JP5860171B2 (zh)
KR (2) KR20160011230A (zh)
CN (1) CN103064530B (zh)
AU (1) AU2013270485C1 (zh)
BR (1) BR112013034056A2 (zh)
RU (1) RU2589873C2 (zh)
TW (1) TWI552008B (zh)
WO (1) WO2014101410A1 (zh)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064530B (zh) * 2012-12-31 2017-03-08 华为技术有限公司 输入处理方法和装置
KR102063766B1 (ko) * 2013-09-17 2020-01-08 엘지전자 주식회사 이동 단말기 및 그것의 제어방법
JP6342734B2 (ja) * 2014-07-10 2018-06-13 大和製衡株式会社
TWI634473B (zh) * 2014-10-29 2018-09-01 富智康(香港)有限公司 掌上型電子裝置的輸入法自動切換系統及方法
CN104932712A (zh) * 2015-06-25 2015-09-23 小米科技有限责任公司 文本输入方法和装置
US20170177205A1 (en) * 2015-12-22 2017-06-22 Mediatek Inc. Input method for recognizing user input according to vector combination derived from single-stroke input and associated non-transitory machine readable medium
CN105700448B (zh) * 2016-01-29 2018-06-08 中国联合网络通信集团有限公司 远程监控方法和远程监控系统
US10565493B2 (en) 2016-09-22 2020-02-18 Salesforce.Com, Inc. Pointer sentinel mixture architecture
JP2018072568A (ja) * 2016-10-28 2018-05-10 株式会社リクルートライフスタイル 音声入力装置、音声入力方法及び音声入力プログラム
CN107680589B (zh) * 2017-09-05 2021-02-05 百度在线网络技术(北京)有限公司 语音信息交互方法、装置及其设备
CN107885348A (zh) * 2017-11-27 2018-04-06 牡丹江师范学院 一种移动通信终端设备的日语输入装置和方法
CN108170353A (zh) * 2017-12-04 2018-06-15 阿里巴巴集团控股有限公司 终端设备执行指令的方法及装置
CN110580103A (zh) * 2018-06-11 2019-12-17 阿里巴巴集团控股有限公司 数据处理方法、装置和机器可读介质
US11557278B2 (en) * 2020-12-10 2023-01-17 Google Llc Speaker dependent follow up actions and warm words
CN112783337B (zh) * 2020-12-31 2024-04-12 科大讯飞股份有限公司 输入法候选项的排序方法及相关模型的训练方法
CN113157319A (zh) * 2021-04-27 2021-07-23 上海申铁信息工程有限公司 一种phm系统的规则编码装置及方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101393493A (zh) * 2007-09-19 2009-03-25 北京三星通信技术研究有限公司 自动注册指定操作的手写笔迹的方法和装置
CN102467336A (zh) * 2010-11-19 2012-05-23 联想(北京)有限公司 电子设备及其对象选择方法
CN102520790A (zh) * 2011-11-23 2012-06-27 中兴通讯股份有限公司 一种基于图像传感模块的字符输入方法、装置及终端
CN102725712A (zh) * 2009-11-09 2012-10-10 因文森斯公司 手持计算机系统及关于人类活动的字符和命令识别的技术
CN103064530A (zh) * 2012-12-31 2013-04-24 华为技术有限公司 输入处理方法和装置

Family Cites Families (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3181409B2 (ja) * 1992-12-28 2001-07-03 株式会社東芝 情報処理装置及び情報処理方法
JPH08137611A (ja) * 1994-11-09 1996-05-31 Toshiba Corp ジェスチャイメージ登録方法及び文書作成装置
US7082391B1 (en) * 1998-07-14 2006-07-25 Intel Corporation Automatic speech recognition
US7098896B2 (en) * 2003-01-16 2006-08-29 Forword Input Inc. System and method for continuous stroke word-based text input
JP2005275652A (ja) * 2004-03-24 2005-10-06 Canon Inc 入力軌跡処理装置および入力軌跡処理方法
JP2006277314A (ja) * 2005-03-29 2006-10-12 Nec Saitama Ltd アドレス入力装置、そのアドレス入力方法およびそのアドレス入力装置をもつ電子機器
JP2007086976A (ja) * 2005-09-21 2007-04-05 Mitsubishi Electric Corp 携帯機器
KR20070052118A (ko) * 2005-11-16 2007-05-21 한국전자통신연구원 아날로그 조이스틱 컨트롤러를 이용한 문자 입력 시스템 및방법
US7698326B2 (en) * 2006-11-27 2010-04-13 Sony Ericsson Mobile Communications Ab Word prediction
WO2008069519A1 (en) * 2006-12-04 2008-06-12 Electronics And Telecommunications Research Institute Gesture/speech integrated recognition system and method
US7895518B2 (en) * 2007-04-27 2011-02-22 Shapewriter Inc. System and method for preview and selection of words
WO2009036293A1 (en) * 2007-09-12 2009-03-19 Macfarlane Scott S Highly compact keyboards
US20090182562A1 (en) * 2008-01-14 2009-07-16 Garmin Ltd. Dynamic user interface for automated speech recognition
KR20090107364A (ko) * 2008-04-08 2009-10-13 엘지전자 주식회사 이동 단말기 및 그 메뉴 제어방법
US20100114887A1 (en) * 2008-10-17 2010-05-06 Google Inc. Textual Disambiguation Using Social Connections
US9519353B2 (en) * 2009-03-30 2016-12-13 Symbol Technologies, Llc Combined speech and touch input for observation symbol mappings
US8782556B2 (en) * 2010-02-12 2014-07-15 Microsoft Corporation User-centric soft keyboard predictive technologies
KR101477530B1 (ko) * 2010-03-12 2014-12-30 뉘앙스 커뮤니케이션즈, 인코포레이티드 이동 전화의 터치 스크린과 함께 사용하기 위한 다중 모드 문자 입력 시스템
JP5627372B2 (ja) * 2010-09-28 2014-11-19 京セラ株式会社 文字入力装置、文字入力方法および文字入力プログラム
JP5503486B2 (ja) * 2010-10-07 2014-05-28 京セラ株式会社 文字入力装置、文字入力方法および文字編集プログラム
US8490008B2 (en) * 2011-11-10 2013-07-16 Research In Motion Limited Touchscreen keyboard predictive display and generation of a set of characters
US8713464B2 (en) * 2012-04-30 2014-04-29 Dov Nir Aides System and method for text input with a multi-touch screen
CN102819567B (zh) * 2012-07-17 2016-04-20 重庆优腾信息技术有限公司 一种电子设备的联系人信息检索方法和系统
US8850350B2 (en) * 2012-10-16 2014-09-30 Google Inc. Partial gesture text entry

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101393493A (zh) * 2007-09-19 2009-03-25 北京三星通信技术研究有限公司 自动注册指定操作的手写笔迹的方法和装置
CN102725712A (zh) * 2009-11-09 2012-10-10 因文森斯公司 手持计算机系统及关于人类活动的字符和命令识别的技术
CN102467336A (zh) * 2010-11-19 2012-05-23 联想(北京)有限公司 电子设备及其对象选择方法
CN102520790A (zh) * 2011-11-23 2012-06-27 中兴通讯股份有限公司 一种基于图像传感模块的字符输入方法、装置及终端
CN103064530A (zh) * 2012-12-31 2013-04-24 华为技术有限公司 输入处理方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2765473A4 *

Also Published As

Publication number Publication date
BR112013034056A2 (pt) 2017-02-07
KR20140114280A (ko) 2014-09-26
RU2014129212A (ru) 2016-02-10
RU2589873C2 (ru) 2016-07-10
KR20160011230A (ko) 2016-01-29
AU2013270485C1 (en) 2016-01-21
EP2765473A1 (en) 2014-08-13
TWI552008B (zh) 2016-10-01
TW201426362A (zh) 2014-07-01
JP2015510176A (ja) 2015-04-02
US20140184514A1 (en) 2014-07-03
AU2013270485B2 (en) 2015-09-10
JP5860171B2 (ja) 2016-02-16
CN103064530A (zh) 2013-04-24
EP2765473A4 (en) 2014-12-10
AU2013270485A1 (en) 2014-06-26
CN103064530B (zh) 2017-03-08
KR101586890B1 (ko) 2016-01-19

Similar Documents

Publication Publication Date Title
TWI552008B (zh) 輸入處理方法和裝置
US10923118B2 (en) Speech recognition based audio input and editing method and terminal device
US6864809B2 (en) Korean language predictive mechanism for text entry by a user
TWI293455B (en) System and method for disambiguating phonetic input
JP2012079252A (ja) 情報端末装置、文字入力方法および文字入力プログラム
WO2018076450A1 (zh) 一种输入方法和装置、一种用于输入的装置
US8015013B2 (en) Method and apparatus for accessing a digital file from a collection of digital files
US20070038456A1 (en) Text inputting device and method employing combination of associated character input method and automatic speech recognition method
US8554781B2 (en) Shorthand for data retrieval from a database
RU2510524C2 (ru) СИСТЕМА И СПОСОБ ВВОДА WuBi
TW200945320A (en) Voice input system and voice input method
WO2015188437A1 (zh) 一种拼音输入方法及装置
CN100517186C (zh) 基于按键和语音识别的文字输入方法及装置
CN102999639A (zh) 一种基于语音识别字符索引的查找方法及系统
TW200947241A (en) Database indexing algorithm and method and system for database searching using the same
WO2010124510A1 (zh) 一种人机界面交互系统和方法
JP2004227557A (ja) 略コード検索方法
TWM453219U (zh) 語意辨識設備
TWI271632B (en) Voice input method for sending message
KR101643563B1 (ko) 휴대 단말의 카메라를 이용한 문자 인식처리 장치 및 방법
TWI272577B (en) Character input methods and computer systems utilizing the same
EP2763053A1 (en) Shorthand for data retrieval from a database
TW200941285A (en) Song-requesting method and system thereof

Legal Events

Date Code Title Description
REEP Request for entry into the european phase

Ref document number: 2013795149

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2013795149

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2013270485

Country of ref document: AU

ENP Entry into the national phase

Ref document number: 20137033362

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2014553610

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2014129212

Country of ref document: RU

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112013034056

Country of ref document: BR

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 112013034056

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20131230