JP4978982B2 - Portable information terminal, character input support program and method - Google Patents

Description

  The present invention relates to a mobile information terminal capable of at least voice call and character input, such as a mobile phone terminal, and a character input support program and method for the mobile information terminal.

  2. Description of the Related Art Conventionally, for example, portable information terminals such as mobile phone terminals are required to be smaller and lighter. Therefore, in such a portable information terminal, it is difficult to take a large space for an operation device for a user interface.

  On the other hand, in recent years, as with a general personal computer, these personal digital assistants also require functions such as a character input / editing function and a function for displaying a web page by connecting to the Internet.

  In particular, in the case of a mobile phone terminal, according to a recent survey of usage conditions, the average number of communications is greater than the number of communications for email communications. From this, it can be recognized that the input / editing of characters is one of important functions in the mobile phone terminal.

  Also, recent mobile phone terminals are equipped with a so-called predictive conversion function as a standard function as a function for reducing the burden when a user inputs characters.

  Japanese Patent Application Laid-Open No. 2004-96502 discloses a technique for performing voice recognition of recorded data and converting it into document data. Therefore, if the technique described in this publication is used as, for example, a technique for assisting character input of a mobile phone terminal, a sentence to be input is recorded by voice, and the recorded data is converted into document data. It is considered that the text input operation by pressing is unnecessary.

Japanese Patent Laid-Open No. 2004-96502

  By the way, in the case of character input using the predictive conversion function described above, the predictive conversion candidates displayed on the display screen are referred to from a predictive candidate dictionary database prepared in advance, or a learning dictionary database based on past input or the like. It is supposed to be referenced from.

  For this reason, for example, in the case where an e-mail text is created using a mobile phone terminal, for example, a predictive conversion candidate displayed on the display screen by referring to the input character “A” from the dictionary database at a certain time, for example. For example, “B”, “C”, “D”, etc., the e-mail transmission destination may be any party, such as a family member, a friend, or a person with a business relationship. The candidates referred to from the dictionary database as predictive conversion candidates for the character “A” are “B”, “C”, “D”,.

  Therefore, if the predicted conversion candidates displayed on the display screen are candidates that do not match the destination of the email, the user can create another new prediction in order to create an email sentence that matches the destination. There is a need to perform operations such as displaying conversion candidates or manually inputting desired characters and words.

  On the other hand, for example, a plurality of prediction candidate dictionary databases may be provided as user dictionaries, and desired characters, words, etc. may be registered in each database, and these user dictionaries may be used by switching for each e-mail destination, for example. However, it is a very time-consuming operation to switch between a plurality of user dictionaries and a dictionary according to the other party. For example, even when a plurality of user dictionaries can be registered, In order to make the dictionaries easier to use, the dictionaries must be constantly updated, and the updating work takes a lot of time and effort, which is hardly practical.

  The present invention has been proposed in view of such a situation, and can reduce the user's work burden at the time of character input and also enables optimization of candidates referred to from the dictionary database, making it easier for the user to use. An object of the present invention is to provide a portable information terminal, a character input support program, and a method capable of realizing character input support.

The portable information terminal of the present invention includes a storage unit for storing conversation voice data, a word extraction unit for extracting words in the conversation in the conversation voice data by voice recognition of the conversation voice data, and the extracted words. A database unit that stores and registers the database together with at least one of the conversation time zone, the conversation partner and the frequency of occurrence of words in the conversation, and the user designates the partner as a process related to the conversation content. when performing a character input of an e-mail with no state, based on the words that are used in e-mail text in the input, e-mail with reference to the at least one of the time zone and the other party and the frequency of database this has a candidate search unit to search for predictive conversion candidate that matches the destination that the estimated, and a presentation section for presenting the retrieved predicted conversion candidates to the user as well as estimates of the destination Accordingly, to solve the problems described above.

Further, the character input support program of the present invention includes a storage unit that stores conversation voice data, a word extraction unit that extracts words in the conversation in the conversation voice data by voice recognition of the conversation voice data, and the extracted A database unit that registers and registers words together with at least one of the conversation time zone and the conversation partner and the frequency of occurrence of the word in the conversation, and the partner as a process related to the conversation content by the user When entering e-mail characters without specifying them, refer to at least one of the database time zone, destination, and frequency based on the words used in the e-mail text being entered. a candidate search unit to search for predictive conversion candidate that matches the destination that the estimated while estimating the destination of e-mail, Hisage to present the retrieved predictive conversion candidate indication section As a control unit, by functioning the portable information terminal computer, to solve the problems described above.

The character input support method of the present invention includes a step of storing conversation voice data, a step of extracting words in the conversation in the conversation voice data by voice recognition from the conversation voice data, and the extracted word, A step of creating a database together with at least one of the conversation time zone, the conversation partner and the frequency of occurrence of words in the conversation, and the user does not specify the partner as processing related to the conversation content When entering e-mail characters in the state, refer to at least one of the database time zone, destination, and frequency based on the words used in the e-mail text being entered . Yes retrieving the predictive conversion candidate that matches the destination that the estimated while estimating the destination, and presenting the retrieved predicted conversion candidates to the user By Rukoto, for solving the above problems.

That is, according to the present invention, the words extracted from the conversation voice data are databased together with at least one of the conversation time zone and the conversation partner and the frequency of occurrence of the words in the conversation, and related to the conversation contents. When an e-mail character is input without specifying the other party as a process to be performed , based on the words used in the input e-mail text, In order to estimate the destination of the e-mail with reference to at least one of the above and to search for a prediction conversion candidate that matches the estimated destination, a prediction candidate word that matches the conversation content can be presented.

According to the present invention, the words extracted from the conversation voice data are databased together with at least one of the conversation time zone and the conversation partner and the frequency of occurrence of the words in the conversation, and processing related to the conversation contents When a character is entered in an email without specifying the recipient , the database time zone, the recipient and the frequency are at least based on the words used in the email text being entered. Since the destination of the email is estimated with reference to one and the prediction conversion candidate suitable for the estimated destination is searched, it is possible to present the optimum prediction candidate word suitable for the conversation content. The user's work burden during input can be reduced, and character input support that is easier for the user to use is realized.

  Hereinafter, an embodiment of a portable information terminal, a character input support program, and a method according to the present invention will be described with reference to the drawings.

  In the present embodiment, as an example of the mobile information terminal of the present invention, a mobile phone terminal capable of at least voice communication using a mobile phone line is cited, but of course, the contents described here are only It is an example, and it goes without saying that the present invention is not limited to this example.

〔Overview〕
In the mobile phone terminal of the present embodiment, when the user is making a call, the voice data of each of the utterances and receptions is stored in the built-in large-capacity storage device, for example, there is a margin in the remaining battery level in the standby state The voice of the stored call voice is recognized during the idle time, and words included in the conversation content of the call voice obtained by the voice recognition are extracted, and the words are classified according to appearance frequency and time zone. A database (hereinafter referred to as a call-related database) associated with information such as the frequency distribution, other related words that are close in time, the degree of temporal proximity of these other related words, and the information on the other party. create.

  Then, the call-related database is used together with the normal prediction conversion dictionary database as a reference destination of the prediction conversion candidate when the user performs character input, and words that are candidates for prediction conversion are presented to the user.

  Thereby, in the mobile phone terminal of the present embodiment, when the user performs character input for creating an e-mail text, for example, the prediction conversion candidate optimized for the destination of the e-mail can be presented to the user. As a result, the user can create a desired sentence with fewer character input operations.

  In addition, it is also possible to add items (for example, call partner name, address, date of birth, hobby, etc.) of the phone book and address book of the mobile phone terminal to the call related database. It is also possible to construct a call-related database by combining previously input characters (for example, words in e-mail sentences that have been sent and received). In addition, when detecting a predictive conversion candidate, the currently set mode such as the current date, day of the week, schedule, manner mode, position information detected by GPS (Global Positioning System), the currently connected base station A user's current state may be estimated based on information or the like, and prediction conversion candidates corresponding to the estimation may be presented.

〔Use Case〕
As a use case of the mobile phone terminal of the present embodiment, FIG. 1 shows an example in which the user 1 has a conversation (call) for planning an Italian trip with a friend 2 using the mobile phone terminal. I will give you.

  In the use case shown in FIG. 1, the mobile phone terminal according to the present embodiment stores conversation voice data of utterances and receptions during a call for the Italian travel plan, and then, for example, a battery remaining in a standby state. Within the free time with a sufficient amount, speech recognition of the stored call voice is performed to extract words included in the conversation content of the call voice. Here, in the conversation related to the Italian travel plan as in the use case of FIG. 1, words related to travel, such as “travel”, “Italy”, “cooking”, etc. frequently appear. The telephone terminal extracts these frequently occurring words as important keywords constituting the conversation, and registers these words in the call-related database in association with the frequency, time zone, call partner, and the like.

  Then, for example, when the user 1 creates an e-mail about the Italian travel plan with the friend 2 as the transmission destination, for example, the mobile phone terminal of the present embodiment includes the call-related database together with the normal prediction conversion dictionary database. Refer to As a result, on the display screen 3 of the mobile phone terminal of the present embodiment, the above-mentioned “travel”, “Italy”, Words such as “cook” are preferentially presented one after another, and as a result, the user 1 can quickly and easily create an e-mail message tailored to the friend 2 who is the other party with less work. Is possible.

  In addition, the mobile phone terminal according to the present embodiment is not limited to the creation of e-mail texts as in the use case described above, but also various character input such as when inputting text when browsing schedules, notepads, web pages, etc. Sometimes, the above-described call-related database can be used.

  As another use case of the mobile phone terminal of the present embodiment, FIG. 2 shows a case where the user 4 has a conversation (call) related to a conference schedule with the manager 5 of the company, for example, using the mobile phone terminal. Take as an example.

  In the use case of FIG. 2, the mobile phone terminal of the present embodiment stores voice data of utterance and reception at the time of a call related to the conference schedule, and then, for example, a free space with a sufficient remaining battery level in a standby state Within the time, speech recognition of the stored call voice is performed, and words included in the conversation content of the call voice are extracted. Here, in the conversation related to the meeting schedule as in the use case of FIG. 2, words related to the meeting frequently appear, for example, “F meeting room”, “meeting”, “product E”. The mobile phone terminal extracts these frequently occurring words as important keywords constituting the conversation, and registers these words in the call-related database in association with the frequency, time zone, call partner, and the like.

  And after that, for example, when the user 4 performs character input for registering a conference schedule using the schedule book function of the mobile phone terminal, the mobile phone terminal of the present embodiment, together with the normal predictive conversion dictionary database, Browse call related databases. Thereby, for example, when the user 4 inputs the date and time in the schedule book, the above-mentioned “F conference room” is displayed on the display screen 6 of the mobile phone terminal of the present embodiment as a prediction conversion candidate optimal for the schedule. , “Meetings”, “Product E”, and the like, words related to the location and content of the conference are preferentially presented one after another. As a result, the user 4 can input a schedule related to the conference with less work. Thus, it can be created quickly and easily.

  FIG. 3 shows an example of a call related database registered in the use case of FIG.

In the example of FIG. 3, in the call-related database, as an example, “word reading kana” (indicated simply as “word” in FIG. 3), “word kanji” (in FIG. 3, simply “kanji” ”,“ Frequency of words ”(indicated simply as“ Frequency ”in FIG. 3),“ Call partner ”,“ Related word ”,“ Received (utterance) data number ”, etc. Each item is registered. In the “word reading kana”, a word reading kana obtained by voice recognition from a call voice is registered. In the “word kanji”, kanji corresponding to the reading kana of the word is registered. As shown in the example of FIG. 3, the call-related database can also handle homonyms. In the “word frequency”, “ a ”, “ b ”, “ c ”,... Indicating the frequency level at which the word appears in the call is registered, and in this embodiment, the level “ a ” is the highest. The level represents the level, and the levels are lower in the order of levels “ b ” and “ c ”. The “word frequency” is a level of “high”, “medium”, and “low” for whether the word appears frequently in the “morning”, “daytime”, or “night” time zone. Is added and registered. In “Call partner”, a call partner who uses the word during a call is registered, and is registered in association with, for example, a destination name in a telephone book or an address book in a mobile phone terminal. In the “related word”, another word that appears in association with the word in the call (for example, another word close in time) is registered. In the “speech (receipt) data number”, the file number of the utterance voice data and the file number of the reception voice data in the call are registered. In a large-capacity storage device of a mobile phone terminal, audio data is managed by the file number.

[Hardware configuration]
FIG. 4 shows a schematic internal configuration of the mobile phone terminal of the present embodiment capable of realizing the above.

  In FIG. 4, a data line is a transmission line for transmitting various data such as voice data, e-mail data, and image data.

  The control line is a transmission line for transmitting various control information such as control data from the control unit 30 configured by a CPU (Central Processing Unit).

  The antenna 32 is connected to the communication circuit 31 and is used to transmit and receive signal radio waves during telephone calls, e-mail communication, Internet connection, and the like.

  The communication circuit 31 performs frequency conversion, modulation, demodulation, and the like of transmission / reception signals.

  Here, call voice data (received voice data) received by the antenna 32 and the communication circuit 31 is sent to the voice processing unit 42, and other received data is sent to the control unit 30 and appropriately processed. After that, it is sent from the control unit 30 to each unit as necessary. The received data other than the received voice is data related to packet communication. For example, e-mail data, moving image or still image data, music data, HTML (Hyper Text Markup Language) data, program code data, etc. Etc.

  When received voice data is supplied from the antenna 32 and the communication circuit 31, the voice processing unit 42 decodes the received voice data, converts the decoded received voice data into an analog voice signal, and receives the analog received voice. An audio signal is sent to the speaker 40 or an earphone jack (not shown).

  In the case of the present embodiment, the received voice data decoded by the voice processing unit 42 is associated with information such as a telephone number that can specify a call destination or call date / time information in the mass storage device 45. Remembered. Note that the other party identification information (telephone number or the like) and the call date / time information may be stored in the memory unit 35.

  The speaker 40 amplifies the analog received voice signal supplied from the voice processing unit 42 with an internal amplifier, converts the received voice signal into an audible voice, and outputs the audible voice to the outside. As a result, the received voice is output from the speaker 40.

  The microphone 41 converts the speech voice into an analog voice signal, amplifies it by an internal amplifier, and further sends the analog speech voice signal to the voice processing unit 42.

  When the analog speech signal is supplied from the microphone 41, the speech processing unit 42 converts the speech signal into digital speech data, encodes the speech signal, and further encodes the speech data to the communication circuit 31. send. As a result, utterance voice data is transmitted from the antenna 32.

  In the case of the present embodiment, the utterance voice data digitized by the voice processing unit 42 is stored in the mass storage device 45 in association with the call partner identification information (telephone number etc.) and the call date / time information. Is done. In this case as well, the destination identification information (telephone number and the like) and call date / time information may be stored in the memory unit 35.

  The large-capacity storage device 45 includes a hard disk drive (HDD), a large-capacity nonvolatile semiconductor memory such as a flash memory, and the received voice associated with the above-described destination identification information (telephone number, etc.) and call date / time information. Data and speech data are stored. The utterance voice data and the reception voice data stored in the large-capacity storage device 45 have a small processing load on the control unit 30 as in the case where the mobile phone terminal is in a standby state, and the remaining battery level. It is read when there is a margin and sent to the control unit 30.

  At this time, the control unit 30 performs voice recognition processing on the voice data, extracts words included in the call voice obtained by the voice recognition, the frequency and time zone of the extracted words, the call partner And the like, and the call-related database is stored in, for example, a nonvolatile memory in the memory unit 35. Note that the storage destination of the call related database may be the mass storage device 45. In addition, when the free capacity of the large-capacity storage device 45 is reduced, the control unit 30 also deletes the call voice data for which the voice recognition processing or the like has been completed from the large-capacity storage device 45.

  The display unit 33 includes a display device such as a liquid crystal display and a display drive circuit, and displays images, characters, and the like on the display screen.

  The operation unit 34 includes operation elements such as a numeric keypad (keyboard), a power button, an utterance / end call button, and a jog dial, and an operation signal generator that generates an operation signal when the operation elements are operated. The user operates the operation unit 34 to make incoming / outgoing calls and e-mail texts for calls, send / receive e-mails, connect to the Internet, and the like.

  The image processing unit 43 performs decompression decoding on the compressed and encoded image data read from the built-in memory under the control of the control unit 30 and displays the decompressed decoded image data via the data line. Send to part 33. The image processing unit 43 performs compression encoding of still image and moving image data captured by a camera unit (not shown), and stores the compression encoded image data under the control of the control unit 30. Send to memory for storage.

  The memory unit 35 includes a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM is an OS (Operating System), control program codes for the control unit 30 to control each unit, various initial setting values, font data, dictionary data, ringtones, key operation sounds, various sound data for alarm sounds, Application program code for creating and editing e-mails, application program code for performing various processing on images and sounds, character input support program code according to the present embodiment, and other mobile phones It stores program codes for various applications installed in the terminal, identification information (ID) of the mobile phone terminal, and the like. In the present embodiment, the character input support program according to the present invention includes a program code for executing recording processing of voice data, a program code for executing voice recognition processing and word extraction processing, and a call related database. And a program code for executing candidate word presentation processing with reference to the call related database. This ROM includes a rewritable ROM such as NAND-type flash memory (NAND-type flash memory) or EEPROM (Electrically Erasable Programmable Read-Only Memory), e-mail data, telephone book and e-mail address book data, Call related database data according to the present embodiment, dictionary data of normal prediction conversion candidate words, learning data at the time of character input, still image and moving image data, sound data for key operation sound, alarm sound, etc. It is also possible to save various user setting values and the like. The RAM stores data as needed as a work area when the control unit 30 performs various data processing.

  The control unit 30 includes a CPU, and controls the mobile phone terminal, the call voice recording process according to the present embodiment, the voice recognition and word extraction process, the call based on the OS and various programs stored in the memory unit 35. It performs control and various calculations to realize various functions such as creation and update processing of related databases and character input processing with reference to call related databases. The voice processing unit 42 may be in charge of the voice recognition process.

  In addition, although not shown in FIG. 4, the mobile phone terminal of the present invention has a browser function, an external memory interface, an external cable connector, an infrared communication function, a short-range wireless communication function, an electronic wallet function, and a camera function. It also includes various components included in a general mobile phone terminal, such as an LED (light emitting diode) light emitting function, a GPS function, a vibrator function, a battery, and a power control function.

〔data flow〕
FIG. 5 shows a basic data flow in the main components shown in FIG. 4 when the character input support process according to the present invention is executed. FIG. 5 shows an example in which the call related database is stored in the mass storage device 45.

  In FIG. 5, first, at the time of a call, the control unit 30 is input via the microphone 41 and digitized by the voice processing unit 42 by the execution of the call voice data recording function 15 according to the character input support process of the present invention. The received speech data is stored in the speech data storage area 19 in the large-capacity storage device 45, and the received speech data received by the communication circuit 31 and decoded by the speech processing unit 42 is stored in the large-capacity storage. It is stored in the received voice data storage area 20 in the device 45.

  Next, in a non-calling state such as a standby state, and when other CPU usage is low and the remaining battery capacity is sufficient, the control unit 30 performs the voice recognition and word extraction function according to the character input support process of the present invention. 16 is executed to read out the speech voice data and the received voice data from the speech voice data storage area 19 and the reception voice data storage area 20 of the large-capacity storage device 45, and perform voice recognition processing and word extraction processing thereof. Then, the extracted words are databased together with the frequency, time zone, call partner, and other related words, sorted according to user settings, and then registered in the call related database storage area 21 of the mass storage device 45. Here, as the sorting according to the user setting, for example, a method of giving priority to frequency, giving priority to the morning time zone, giving priority to a specific calling party, or using only words in a designated conversation can be considered. Note that the call-related database may be stored in the memory unit 35 as in the example of FIG. 4 described above.

  Thereafter, when the user inputs a sentence via the keyboard of the operation unit 34 such as creation of an e-mail sentence, editing of a schedule book, writing to a memo pad, etc., the control unit 30 supports the character input support of the present invention. By executing the candidate word presentation function 17 related to the processing, words that are prediction candidates are presented to the user while referring to the call related database in the call related database storage area 21 of the mass storage device 45.

[State transition]
FIG. 6 shows a state transition diagram of the character input support process according to the present invention.

  In FIG. 6, the mobile phone terminal of this embodiment always transitions to the call recording state S3 and performs call recording processing when in the call state S2, and sequentially stores the utterance voice data and the reception voice data in the mass storage device 45. To do.

  In the mobile phone terminal according to the present embodiment, when the CPU usage rate of the control unit 30 is low and the remaining battery level is sufficient when in the standby state S1, the mobile phone terminal transitions to the word extraction state S4 and stores a large amount of memory. Perform voice recognition processing and word extraction processing of the voice data stored in the device 45, and further create a database with the appearance frequency, time zone, other related words, etc. of the word, and create or update a call related database I do.

  In addition, when the user performs a character and enters the character input state S5, the mobile phone terminal according to the present embodiment transitions to the candidate presentation state S6, and presents the predicted conversion candidate referring to the call related database on the display.

[Call recording process flowchart]
FIG. 7 shows a flowchart of a call voice data recording processing program executed by the control unit 20 in the character input support program according to the present invention.

  In FIG. 7, the control unit 20 performs an outgoing / incoming operation for a telephone call, and when a telephone conversation is started as a process of step S <b> 10, a speech voice input and digitized through the microphone 41 is processed as a process of step S <b> 11. Data, that is, voice data to be sent to the other party is recorded in the mass storage device 45.

  At the same time, the control unit 20 sequentially records the received voice data received through the communication circuit 31, that is, the voice data sent from the other party in the large capacity storage device 45 as the process of step S 12.

  Thereafter, the control unit 20 returns the mobile phone terminal to the standby state when the call is ended in step S13.

[Flowchart of voice recognition, word extraction and creation / update processing of call related database]
FIG. 8 shows a flowchart of a voice recognition process, a word extraction process, and a call related database creation / update process program executed by the control unit 20 in the character input support program according to the present invention.

  In FIG. 8, the control unit 20 reads out the speech voice data and the received voice data recorded in the large-capacity storage device 45 as a process in step S20, and performs voice recognition processing on the voice data using voice recognition technology. Further, word extraction is performed.

  Next, the control unit 20 detects the appearance frequency of the extracted word, the appearance time zone, and other related words that follow before and after as the process of step S21, and further includes the main information together with the information as the process of step S22. Create a database of words.

  Thereafter, the control unit 20 sorts the call-related database in accordance with the mode set by the user and returns to the standby mode in the standby state as the process of step S23.

[Flowchart of candidate word presentation processing]
FIG. 9 shows a flowchart of a processing program of the character input support program according to the present invention in which the control unit 20 refers to the call related database and presents a prediction candidate word when inputting characters.

  In FIG. 9, when a character is input by the user, the control unit 20 searches the call related database by using the currently input character or the already input character or word as a search key as the process of step S30. To do.

  Next, the control part 20 shows the word hit by the search as a prediction candidate word on a display according to the mode which the user has set as a process of step S31.

  Next, as a process of step S32, the control unit 20 determines whether or not the user has selected the word of the presented prediction candidate. If not, the process returns to step S30 to return to a new prediction candidate. Perform word search and presentation processing.

  On the other hand, when it is determined in step S32 that the user has selected the word of the presented prediction candidate, the control unit 20 sets the related word of the selected word as the next prediction conversion candidate as the process of step S33. Present to the user. That is, when a candidate word is selected, other words related to past calls are successively presented for the selected word.

  Next, as a process of step S34, the control unit 20 determines whether or not an instruction to end character input has been given by the user. When no instruction to end has been given, the process returns to step S32, and when an instruction to end has been given, Return to standby mode.

[Summary]
As described above, according to the mobile phone terminal of the embodiment of the present invention, the words extracted from the call voice are made into a database in association with the appearance frequency, time, call partner, and other words, and the call is made when characters are input. Since prediction conversion candidates are presented with reference to related databases, the range of data that can be referred to by the prediction conversion function is widened, and high-performance character input support can be realized.

  In addition, according to the mobile phone terminal of the present embodiment, the prediction conversion candidate is presented with reference to the word database extracted from the call voice. For example, it is related to the conversation content with the other party who made the call in the past. Predictive candidates that are optimal for creating e-mail texts and predictive candidates that are optimal for applications such as a schedule book and memo pad can be presented, reducing the burden of character input by the user, Speeding up, for example, creating texts that meet other parties, and creating texts that link conversation content with text content. In particular, when creating an e-mail to be sent to a party who has made a call in the past, you can use words and expressions that you have met with the other party, and you can use the past conversation contents and e-mail text. The discrepancy with the contents of can be reduced.

  In addition, according to the mobile phone terminal of the present embodiment, since the call related database can be sorted according to the user's setting, the user searches for a candidate from the call related database and presents a mode, for example, a frequency. It is possible to set a desired mode such as a priority mode, a mode in which the morning time zone is prioritized, a mode in which the priority party is changed, or a mode in which only words in a designated conversation are used.

  The above description of the embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and various modifications can be made according to the design and the like as long as they do not depart from the technical idea of the present invention.

  For example, the mobile information terminal of the present invention is not limited to a mobile phone terminal, and can be applied to various mobile information terminals such as a PDA equipped with a communication function.

  In the above-described embodiment, an example in which a predictive conversion candidate is presented has been described. For example, in the case where an e-mail sentence having a high degree of freedom such as an e-mail in HTML format is created, a large-capacity storage device It is also possible to cut out voice data near the keyword from the voice data stored in 45 and attach it to an e-mail.

  In the present embodiment, since the call voice data is recorded in the large-capacity storage device 45, for example, the voice of a word presented as a predictive conversion candidate during character input, or the word and the voices before and after the word. It is also possible to read out data or voice data related to the word from the mass storage device 45 and output it from the speaker 40.

  Also, for example, when creating an e-mail message, when the other party is not specified, refer to the call related database based on the word used in the e-mail message currently being entered, and It is also possible to estimate the destination and refer to the predicted conversion candidate that matches the estimated destination from the call-related database.

  In addition, although the Japanese language is taken as an example in the above-described embodiment, the present invention can of course be applied to languages other than Japanese.

It is a figure which shows an example of prediction conversion candidate presentation when a user performs the telephone call of the plan of an Italian trip with a friend using a mobile phone terminal as a use case of this invention embodiment. It is a figure which shows an example of prediction conversion candidate presentation when a user performs the telephone call regarding a meeting schedule with a supervisor of a company using a mobile phone terminal as a use case of this invention embodiment. It is a figure which shows an example of the telephone call related database registered in the use case of FIG. It is a block diagram which shows the schematic internal structure of the mobile telephone terminal of embodiment of this invention. It is a basic data flow figure in the main component of the mobile telephone terminal of this embodiment at the time of performing the character input support function concerning the present invention. It is a state transition diagram of the character input assistance process concerning this invention. It is a flowchart of the telephone call voice data recording processing program which the control part of the mobile telephone terminal of this embodiment performs. It is a flowchart of a voice recognition process, a word extraction process, and a call related database creation / update process program executed by the control unit of the mobile phone terminal of the present embodiment. It is a flowchart of the processing program which a control part of the mobile telephone terminal of this embodiment presents a prediction candidate word with reference to a telephone call related database at the time of character input.

Explanation of symbols

  1, 4 users, 2 friends, 5 boss, 3, 6 display screen, 15 call voice data recording function, 16 voice recognition and word extraction function, 17 candidate word presentation function, 19 utterance voice data storage area, 20 received voice data storage Area, 21 call-related database storage area, 30 control unit, 31 communication circuit, 32 antenna for communication, 33 display unit, 34 operation unit, 35 memory unit, 40 speaker, 41 microphone, 42 audio processing unit, 43 image processing unit , 45 Mass storage device

Claims (7)

A storage unit for storing conversation voice data;
A word extraction unit for extracting words in the conversation in the conversation voice data by voice recognition of the conversation voice data;
A database unit for registering the word extracted by the word extraction unit as a database together with at least one of the frequency of appearance of the word in the conversation and the destination of the conversation and the destination of the conversation;
When a user inputs an e-mail character without specifying a destination as a process related to the conversation content , the above-mentioned database of the database is based on the word used in the e-mail sentence being input. A candidate search unit that estimates a destination of the e-mail with reference to at least one of a time zone, the destination, and the frequency, and searches for a predictive conversion candidate that matches the estimated destination ;
A portable information terminal comprising: a presentation unit that presents a user with the predicted conversion candidate searched by the candidate search unit.   The portable information terminal according to claim 1, wherein the word extraction unit determines whether or not to execute voice recognition and word extraction of the conversation voice data based on whether or not other processing is performed and a remaining battery level.   The portable information terminal according to claim 1, wherein the candidate search unit determines a search mode when searching for a predictive conversion candidate from the database according to a user setting. Having a communication unit for communicating at least through a mobile phone line;
The portable information terminal according to claim 1, wherein the storage unit stores conversation voice data by a call through the mobile phone line. The storage unit stores conversational voice data from a call through the mobile phone line separately for an uttered voice and an incoming voice,
The word extraction unit extracts a word by recognizing the uttered voice and the received voice individually,
5. The portable information terminal according to claim 4, wherein the database unit creates a database of words individually extracted from the uttered voice and the received voice. A storage unit for storing conversation voice data;
A word extraction unit for extracting words in the conversation in the conversation voice data by voice recognition of the conversation voice data;
A database unit for registering the word extracted by the word extraction unit together with at least one of the time zone of the conversation, the destination of the conversation, and the frequency of occurrence of the word in the conversation;
When a user inputs an e-mail character without specifying a destination as a process related to the conversation content , the above-mentioned database of the database is based on the word used in the e-mail sentence being input. A candidate search unit that estimates a destination of the e-mail with reference to at least one of a time zone, the destination, and the frequency, and searches for a predictive conversion candidate that matches the estimated destination ;
As a presentation control unit that causes the presentation unit to present prediction conversion candidates searched by the candidate search unit,
A character input support program that allows a computer of a portable information terminal to function. Storing conversation voice data in a storage unit;
A step of extracting a word in the conversation in the conversation voice data by voice recognition from the conversation voice data,
A step of creating a database of the words extracted by the word extraction unit together with at least one of the conversation time zone, the conversation partner and the frequency of occurrence of the word in the conversation, and registering the database unit When,
When a user performs character input of an e-mail without specifying a destination as a process related to the conversation content, the candidate search unit uses the word used in the e-mail sentence being input based on the word , retrieving the predictive conversion candidate that matches the destination that the estimated while estimating at least one reference to the destination of the electronic mail with the time zone and the other party and the frequency of said database,
And a step of presenting the prediction conversion candidate searched by the candidate search unit to the user at the presentation unit.

Images