JP2014059330A - Tone display control device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a tone display control device in which when in language learning by way of oral reading, voice reading a text is input by a user, it can be easily confirmed via visual observation whether a tone of the voice is appropriately changed as to the text.SOLUTION: When a model voice reading out a sentence is input by directly reading out by a teacher or the like, or by selecting from a database and outputting from a voice output part, the model voice data is changed into a text by way of voice recognition, a tone at each word end part is determined, a model text Tm, in which a letter at a word end of each text-changed word is discriminated by a form according to a tone thereof, is generated, and displayed in a model display area 17M.When a read-out voice by a user (learner) is input while following the model voice and the text Tm, a user text Tu, which is generated by discriminating a word end part of each word by means of the voice recognition text-making processing and tone determination processing as mentioned above, is displayed in a comparative manner in a user display area 17U adjoining the model display area 17M.

Description

  The present invention relates to a tone display control device for learning, for example, a tone of a language that is not a native language, and a control program therefor.

  In the learning of example sentences such as foreign languages, there is a widespread learning in which a learner reads a model voice data by referring to the model voice data recorded by a teacher or recorded.

  In addition, a language learning device is also conceivable in which a model speech waveform and a speech waveform read out by a learner are displayed side by side at positions corresponding to example sentences, so that the difference can be confirmed and learned (for example, Patent Documents). 1).

JP 2000-250401 A

  In the conventional language learning device, the difference between the exemplary speech waveform and the learner's speech waveform can be confirmed, and learning can be performed so that the learner's own speech waveform is similar to the exemplary speech waveform. However, although the strength and weakness can be easily visually recognized from the speech waveform, there is a problem that it is not clear from the speech waveform whether the read-out speech of the learner is correct or wrong. For example, some learners may not pronounce the word even if they intend to raise the ending of the word, but may pronounce it strongly.

  The present invention has been made in view of such a problem, and when a voice is input to a text, it is possible to easily confirm visually whether the sound has been appropriately changed in tone for the text. It is an object to provide a tone display control device and a control program therefor.

  The tone display control apparatus according to the present invention includes a display unit, voice input means for inputting a sentence-reading voice, voice recognition means for recognizing the voice inputted by the voice input means, and converting the voice into text. Tone determination means for determining the tone of the voice input by the means for each word made into text by the speech recognition means, text data converted to text by the speech recognition means, and for each word determined by the tone determination means Tone reduction text generation means for generating tone reduction text data obtained by identifying each word in a form according to the tone, and tone reduction text data generated by the tone reduction text generation means And a tone reduction text display control means to be displayed on the display unit.

  A program according to the present invention is a program for controlling a computer of an electronic device having a display unit, and the computer inputs voice input means for inputting a text-to-speech voice, and voice input by the voice input means. Speech recognition means for recognizing text, sound tone determination means for determining the tone of the voice input by the voice input means for each word textified by the voice recognition means, and text converted by the voice recognition means Tone reduction text generation means for generating tone reduction text data by identifying each word in a form corresponding to the tone based on the text data and the tone of each word determined by the tone determination means, and this tone reduction text Tone reduction text display system for displaying the tone reduction text data generated by the generation means on the display unit. Means, is characterized in that to function as a.

  According to the present invention, when a voice is input for a text, it is possible to easily visually confirm whether or not the voice is appropriately changed in tone.

The block diagram which shows the structure of the electronic circuit of touchscreen PDA10 which concerns on embodiment of the tone display control apparatus of this invention. The front view which shows the external appearance structure of touchscreen PDA10 on which the language learning assistance screen G was displayed. The figure explaining the basic operation | movement accompanying the language learning assistance function of the said touchscreen PDA10. The flowchart which shows the language learning assistance process of the said touchscreen PDA10. The flowchart which shows the speech recognition & tone detection process accompanying the language learning assistance process of the said touchscreen PDA10. The figure which shows the display operation (the 1) in the learning text display window 17T accompanying the language learning assistance process of the said touchscreen PDA10. The figure which shows the display operation (the 2) in the learning text display window 17T accompanying the language learning assistance process of the said touchscreen PDA10. The figure which shows the display operation in the learning text display window 17T accompanying the reading aloud of Chinese by the language learning assistance process of the said touchscreen PDA10.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an electronic circuit of a touch panel PDA 10 according to an embodiment of a tone display control device of the present invention.

  This tone display control device is configured as a touch panel PDA (personal digital assistants) 10 equipped with a reading aloud learning support function described below, or as a PC (personal computer), a mobile phone, a portable game machine, or the like. .

  The PDA 10 is configured by a computer that reads a program recorded on various recording media or a transmitted program and whose operation is controlled by the read program. The electronic circuit includes a CPU (central processing unit). ) 11 is provided.

  The CPU 11 is a device control program stored in advance in the memory 12, a device control program read from the external recording medium 13 such as a ROM card into the memory 12 via the recording medium reading unit 14, or a Web on the Internet N. The operation of each part of the circuit is controlled in accordance with a device control program read into the memory 12 from the server (in this case, a program server) via the communication control unit 18.

  The device control program stored in the memory 12 is an input signal corresponding to a user operation from the touch panel type color display unit 17, an audio signal input via the audio input unit 15, or a communication control unit 18. In response to a communication signal with each Web server on the connected Internet N or a connection communication signal with a memory card (recording medium) 13 such as an EEPROM, RAM, or ROM connected externally via the recording medium reading unit 14 Is activated.

  As described above, the CPU 11 is connected to the memory 12, the recording medium reading unit 14, the audio input unit 15, the audio output unit 16, the touch panel type color display unit 17, the communication control unit 18, and the like.

  Examples of the device control program stored in the memory 12 include a system program that controls the overall operation of the PDA 10, web servers on the Internet N via the communication control unit 18, and a user PC (Personal Computer) (not shown). In addition to storing communication programs for data communication, various application programs stored in advance or downloaded from the outside, such as mail function, recording / playback function, recording / playback function, document processing function, schedule management function, etc. Is memorized.

  Furthermore, in order to display example sentence text for language learning, and to identify and display the text itself in a form corresponding to the tone (increase / decrease) in response to input of voice data corresponding to the example sentence text. Language learning support program 12a, and a word speech recognition processing module (program) for executing speech recognition processing, word sampling processing, tone analysis processing, phrase detection processing, and the like of the speech data in accordance with execution of the language learning support program 12a 12b is stored.

  Further, the memory 12 is provided with an exemplary sentence example database 12c that stores a plurality of exemplary sentence examples for language learning as a combination of text and speech data, and is necessary for the execution of the language learning support program 12a. As a work area, model voice memory 12d, user voice memory 12e, voice recognition text memory (model / user) 12f, word recognition completion time sampling memory (model / user) 12g, word-by-word ending waveform registration table (model / user) 12h, word-by-word ending tone registration table (example / user) 12i, word-by-word tone reduction text memory (example / user) 12j, practice word setting memory 12k, and the like are secured.

  In the exemplary speech memory 12d, for example, speech data input from the speech input unit 15 as a teacher reads out example sentences, or example sentences selectively read from the model example sentence database 12c according to user operations. Is stored (see (1) in FIG. 3).

  The user voice memory 12e stores, for example, voice data (user voice data) input from the voice input unit 15 by the learner based on the text of the example sentence as an example and its voice ((1 in FIG. 3). )reference).

  In the voice recognition text memory (exemplary / user) 12f, the model voice data and user text data are converted into text by recognizing the waveforms of the model voice data and the user voice data according to the word voice recognition processing module 12b. Is stored (see (2) in FIG. 3).

  In the word recognition completion time sampling memory (exemplary / user) 12g, along with the voice recognition of the model voice data and the user voice data, the recognition completion times t1, t2 for each word from the input start time of the voice data. .., Tn are sampled and stored by a timer or the like (see (3) in FIG. 3).

  In the word-by-word ending waveform registration table (exemplary / user) 12h, the recognition completion times t1, t2,..., Tn based on the recognition completion times t1, t2,. , Tn, the waveform of the speech data immediately before tn is registered as the waveform of the ending part for each word (see (4) in FIG. 3).

  In the word-by-word ending tone registration table (exemplary / user) 12i, the tone of the ending portion for each word determined by pitch analysis of the waveform of the ending portion for each word (see, for example, Japanese Patent Laid-Open No. 2000-250401). (Up adjustment part or down adjustment part) is registered (see (5) in FIG. 3). In addition, in the word-by-word ending tone registration table (exemplary / user) 12i, phrase break information (phrases 1, 2,...) Detected according to the contents of the speech-recognized text data is also registered.

  In the tone reduction text memory for each word (exemplary / user) 12j, the color corresponding to the tone of the tail part for each word (up tone part: red / down tone part: for the example and user text data). In addition to (blue), tone-delimited reduced text data generated by adding a delimiter ap at a position corresponding to the phrase delimiter information is stored (see (6) in FIG. 3).

  The practice word setting memory 12k compares the model tone-delimited reduced text data stored in the word-by-word tone-reduced text memory (model / user) 12j with the user's tone-delimited reduced text data, and tone information thereof. If there is a difference in (color at the end of the word), the difference word is stored as a practice target word (see FIGS. 6C and 6D).

  FIG. 2 is a front view showing an external configuration of the touch panel PDA 10 on which the language learning support screen G is displayed.

  In the language learning support screen G displayed on the touch panel display unit 17 in accordance with the learning support processing by the language learning support program 12a, the model voice input mode for inputting the model voice data corresponding to the example sentence is set. ] Key 19m, operated when selecting the model voice data from the model example sentence database 12c and inputting it in the model voice input mode [example sentence] key 19i, for inputting user voice data corresponding to the example sentence [User] key 19u set in the user voice input mode, [Voice input] key 19v operated when inputting the voice of the model or the user from the voice input unit 15, voice data of the model (teacher) corresponding to the example sentence [Practice] to set to practice mode when the tone of the user and the tone of the voice data of the user (learner) are different Over 19p is displayed.

  The model tone-delimited reduced text data Tm stored in the word-by-word tone-reduction text memory (model / user) 12j is displayed in the model display area 17M of the learning text display window 17T, and the user's tone-delimited reduced text data. Tu is displayed in the user display area 17U of the learning text display window 17T.

  In the touch panel PDA 10 configured as described above, the CPU 11 controls the operation of each part of the circuit in accordance with instructions described in the language learning support program 12a and the word speech recognition processing module 12b, and the software and hardware cooperate with each other. The functions described in the following operation explanation are realized.

  Next, a specific operation associated with the language learning support function by the touch panel PDA 10 having the above configuration will be described.

  FIG. 3 is a diagram for explaining a basic operation associated with the language learning support function of the touch panel PDA 10.

  FIG. 4 is a flowchart showing language learning support processing of the touch panel PDA 10.

  FIG. 5 is a flowchart showing a voice recognition & tone detection process associated with the language learning support process of the touch panel PDA 10.

  FIG. 6 is a diagram showing a display operation (part 1) in the learning text display window 17T accompanying the language learning support processing of the touch panel PDA 10.

  FIG. 7 is a diagram showing a display operation (part 2) in the learning text display window 17T accompanying the language learning support processing of the touch panel PDA 10.

  When the language learning support program 12a is started in response to a user operation, the language learning support screen G (see FIG. 2) in which the [exemplary] key 19m to the [practice] key 19p are arranged is displayed on the touch panel display unit 17. The

  In order to input a reading voice as a model for an arbitrary example sentence to be read aloud, the [model] key 19m is touched to set the model voice input mode (step S1 (Yes)), then [voice input] When the key 19v is touched (step S2 (Yes)) and, for example, the reading voice of the example sentence “Get ready to say goodbye to the old, hello to the new” by the teacher is sequentially input from the voice input unit 15 (step S3) As shown in (1) of FIG. 3, the inputted model voice data is stored in the model voice memory 12d, and is sequentially subjected to voice recognition processing and tone detection processing according to the word voice recognition processing module 12b ( Step SA).

  On the other hand, after the [exemplary] key 19m is touched to set the exemplary voice input mode (step S1 (Yes)), when the [example sentence] key 19i is touched (step S4 (Yes)), ( As shown in 1), the voice data of the model sentences (for example, “Get ready to say goodbye to the old, hello to the new”) selected according to the user operation are sequentially read from the model sentence database 12c. Are output from the audio output unit 16 and stored in the exemplary audio memory 12d (step S5). Then, in accordance with the word speech recognition processing module 12b, the speech recognition processing and tone detection processing are performed sequentially (step SA).

  The model voice data stored in the model voice memory 12d is sequentially input to the word voice recognition processing module 12b (step A1), and the model text data is converted into text as shown in (2) of FIG. “Get ready to say goodbye to the old, hello to the new” is stored in the speech recognition text memory (exemplary) 12f (step A2).

  Further, along with the voice recognition of the model voice data, the recognition completion times t1, t2,..., Tn for each word from the voice data input start time are sampled by a timer or the like, as shown in (3) of FIG. Thus, the recognition completion time for each word is stored in the sampling memory (exemplary) 12g (step A3).

  Then, based on the recognition completion times t1, t2,..., Tn for each word of the sampled exemplary speech data, the waveform of the exemplary speech data immediately before each recognition completion time t1, t2,. It is extracted as a waveform of the ending portion for each word and is registered in the ending waveform registration table (exemplary) 12h for each word as shown in (4) of FIG. 3 (step A4).

  Then, the waveform of the ending part for each word of the exemplary speech registered in the word-by-word ending waveform registration table (exemplary) 12h is subjected to pitch analysis, and the tone (ascending tone part or descending tone part) of the ending part is determined. As shown in (5) of FIG. 3, it is registered in the word-by-word ending tone registration table (exemplary) 12i (step A5).

  At this time, phrase delimiter information (phrases 1, 2,...) Is detected according to the content of the voice-recognized model text data “Get ready to say goodbye to the old, hello to the new.”. It is registered in the tone registration table (example) 12i (step A6).

  Then, with respect to the model text data “Get ready to say goodbye to the old, hello to the new.” Stored in the voice recognition text memory (model) 12f, as shown in (6) of FIG. A color (ascending tone part: red / descending tone part: blue) corresponding to the tone of the tail part of each word registered in the word-by-word ending tone registration table (exemplary) 12i is added, and the phrase delimiter information (sentence) 1, 2, ...) An example tone-delimited reduced text data Tm “Get (red) ready (blue) to say / goodbye (blue) to the old (blue), hello” (Red) to the (red) new (blue). "Is generated and stored in the tone-reduction text memory (example) 12j for each word. Then, the exemplary tone-delimited reduction text data Tm generated and stored in the word-by-word tone reduction text memory (exemplary) 12j is sequentially displayed in accordance with the input of the exemplary voice data as shown in FIG. Then, it is displayed in the model display area 17M of the learning text display window 17T (step S6).

  Thereafter, in accordance with the input of the exemplary speech data, as shown in FIG. 6B, the exemplary tone-delimited reduced text data Tm “Get (red) ready () is displayed in the exemplary display area 17M of the learning text display window 17T. (Blue) to say / goodbye (blue) to the old (blue), / hello (red) to the (red) new (blue). " In order to input the reading voice, the [user] key 19u is touched to set the user voice input mode (step S7 (Yes)).

  Then, when the [speech input] key 19v is touched (step S8 (Yes)), and the reading voice of the model tone-delimited reduced text data Tm is sequentially input from the voice input unit 15 (step S9). The input user's voice data is stored in the user voice memory 12e as shown in (1) of FIG. 3, and similar to the processing for the model voice data (see (2) to (6) of FIG. 3). ), Voice recognition processing and tone detection processing are sequentially performed in accordance with the word voice recognition processing module 12b (step SA).

  Then, the user's text data “Get ready to say goodbye to the old, hello to the new.” Stored in the speech recognition text memory (user) 12f is registered in the word-by-word ending tone registration table (user) 12i. A color corresponding to the tone of the tail part of each word (ascending tone part: red / descending tone part: blue) is added, and a delimiter ap at a position corresponding to the phrase delimiter information (sentences 1, 2,...) Tone / goodbye (blue) to the old (blue), hello (red) to the (red) new (blue). Is generated and stored in the word-by-word tone reduction text memory (user) 12j. Then, the user's tone delimiter reduction text data Tu generated and stored in the word-by-word tone reduction text memory (user) 12j is input to the user's voice data as shown in FIGS. In addition, they are sequentially displayed in the user display area 17U of the learning text display window 17T (step S10).

  Here, the model tone-delimited reduced text data Tm and the user tone-delimited reduced text data Tu stored in the word-by-word tone-reduced text memory (model / user) 12j are compared, and the tone information (the end of each word) is compared. It is determined whether or not there is a difference in the color of the part (step S11).

  When it is determined that there is no difference between the tone information of the exemplary tone-delimited reduced text data Tm and the tone information of the user's tone-delimited reduced text data Tu (step S11 (No)), the tone of the user voice A message (“GOOD!” Or the like) is displayed indicating that the tone matches the tone of the model voice (step S13).

  On the other hand, in step S11, as shown in FIG. 6C, the tone information (blue) of the second word “ready (blue)” from the top of the exemplary tone-delimited reduced text data Tm and the tone of the user If it is determined that there is a difference between the tone information (red) of the second word “ready (red)” of the delimited reduction text data Tu (step S11 (Yes)), the tone information is different. A yellow marker My is added to the text portion “y” of each of the model and the user for identification (step S12).

  Here, as shown in FIG. 6D, when the [Practice] key 19p is touched (Step S14 (Yes)), the word “ready” whose tone information (color at the end of the word) has been different. Is stored in the practice word setting memory 12k as a practice target word, and the practice target word “ready” in the model tone-delimited reduction text data Tm displayed in the model display area 17M is identified by a red underline RL. It is displayed (step S15).

  Then, among the exemplary speech data stored in the exemplary speech memory 12d, the speech data of the portion corresponding to the practice target word “ready” identified and displayed by the red underline RL is the recognition completion time for each word. The corresponding word “ready” stored in the sampling memory (exemplary) 12g is read based on the sampling times t1 to t2 and output from the audio output unit 16 (step S16).

  Then, in accordance with the exemplary voice of the portion corresponding to the practice target word “ready” output from the voice output unit 16, a reading voice of the same word “ready” uttered by the user is input from the voice input unit 15. Then, the user's voice data “ready” is stored in the user voice memory 12e and is subjected to voice recognition processing and tone detection processing according to the word voice recognition processing module 12b (step SA). ).

  Then, for the user's text data “ready” stored in the speech recognition text memory (user) 12f, a color corresponding to the tone of the tail part of the word registered in the word-by-word tail tone registration table (user) 12i (Ascending tone part: red / descending tone part: blue) and a delimiter ap at a position corresponding to the phrase delimiter information (sentences 1, 2,...) (Here, one word “ready” Therefore, the user's tone delimitation reduction text data Tu “ready (blue)” is generated and stored in the word-by-word tone reduction text memory (user) 12j. Then, the user's tone-delimited reduced text data Tu “ready (blue)” generated and stored in the word-by-word tone reduction text memory (user) 12j is, as shown in FIG. In accordance with the input, a red underline RL is added and displayed in the user display area 17U of the learning text display window 17T (step S18).

  Here, the tone delimited text data Tm of the model corresponding to the practice target word “ready” and the tone delimited reduced text data Tu of the user stored in the word-by-word tone reduced text memory (model / user) 12j. If it is compared and it is determined that there is a difference in the tone information (color at the end of the word) (step S19 (Yes)), similarly to the above, the example of the difference in the tone information and the text of each user The yellow marker My is added to the portion for identification and display (step S20), and the process of redoing with the same practice target word from step S16 is performed.

  On the other hand, as shown in FIG. 6D, an exemplary tone-delimited reduced text data Tm “ready (blue)” corresponding to the practice target word “ready” and the user's tone-delimited reduced text data Tu “ready ( If it is determined that there is no difference in tone information (color at the end of the word) with “blue)” (step S19 (No)), it has been added to the text portion “y” of each model and user so far. Instead of erasing the yellow marker My, only the marker frame M is added and is identified and displayed as a history of a reading error part (step S21).

  Then, the word “ready” to be practiced currently stored in the practice word setting memory 12k is converted to the model text data “Get ready to say goodbye to the old” stored in the speech recognition text memory (example) 12f. , hello to the new. ”is determined (step S22).

  Here, when it is determined that the word “ready” to be practiced is not the entire text data “Get ready to say goodbye to the old, hello to the new.” (Step S22 ( No)), the character string of the word to be practiced is extended one word at a time up to and including the word “ready” of the original tone difference part, as shown in FIG. Then, the extended practice target word “Get ready” in the model tone-delimited reduced text data Tm displayed in the model display area 17M is identified and displayed by the red underline RL (step S23).

  Then, the reading aloud practice process in steps S16 to S23 is repeated in the same manner as shown in FIGS. 7A to 7D while the practice target words stored in the practice word setting memory 12k are extended one word at a time. Executed.

  Thereafter, in step S22, as shown in FIG. 7D, the word to be practiced is added to the entire text data “Get ready to say goodbye to the old, hello to the new”. If it is determined that it has been reached (step S22 (Yes)), the series of language learning support processing is terminated.

  Therefore, according to the language learning support function by the touch panel PDA 10 having the above-described configuration, the model voice data to be read out is selected from the voice input unit 15 directly or from the model example sentence database 12c and read from the voice output unit 16 by the teacher. When input while outputting, the voice data of the model is converted into text by voice recognition, the tone of the ending portion is determined for each word, and the ending character of each word converted to the text is in accordance with the tone The model text data Tm identified by (color, font, pattern, mark addition, etc.) is generated and displayed in the model display area 17M. Learning the voice data and text data Tm of this model, when the user (learner) reads out the speech of the same example sentence, the ending part of each word is identified by the same voice recognition text conversion processing and tone determination processing. The generated user text data Tu is displayed in comparison with the user display area 17U arranged in the model display area 17M.

  Thereby, when a user (learner) inputs a voice for a text, it can be easily confirmed visually whether or not the voice is appropriately changed in tone for the text. In addition, since the model voice and the user input voice are displayed in contrast, it is easy to visually check which part of the text matches the model voice and which part is wrong. It becomes possible.

  Further, according to the language learning support function by the touch panel PDA 10 having the above configuration, the text data Tu of the user displayed in the user display area 17U is compared with the text data Tm of the model displayed in the model display area 17M. If there is a different part (the identification form corresponding to the tone of the ending part of each word is different), the corresponding part is identified and displayed by the yellow marker My.

  Thereby, the user (learner) can more clearly confirm visually which part of the text matches the model voice and which part is wrong.

  Furthermore, according to the language learning support function using the touch panel PDA 10 having the above configuration, the word identified and displayed by the yellow marker My is set as a practice target word, and the practice target word in the exemplary text data Tm is underlined in red. While being identified and displayed by the RL, exemplary voice data of the word part is cut out and output. If the user (learner) reads out the speech of the word in accordance with the identification display of the practice target word in the model text data Tm and the voice data of the model of the corresponding word, the same speech recognition text conversion process as described above is performed. In addition, the text data Tu of the corresponding word generated by the tone determination process is displayed in the user display area 17U.

  As a result, it is possible to effectively practice reading aloud by cutting out only words whose tone of the reading voice of the user is different from that of the reading voice of the sentence.

  In the above-described embodiment, the end part of each word is converted into a form (color, font, pattern) corresponding to its tone (up / down) by means of speech recognition text-to-speech conversion processing and tone judgment processing of English example sentences. , Mark addition, etc.) to generate and display the identified text data. As shown in FIG. 8, in the case of Chinese composed of an array of kanji, each kanji is in a form corresponding to its tone (four voices) by the voice recognition text conversion process and tone determination process (one voice; Generated and displayed text data identified as red letters or red lines added / 2 voices; blue letters or blue rising lines added / 3 voices; brown letters or brown lines added / 4 voices; green letters or green descending lines added) What is necessary is just composition.

  FIG. 8 is a diagram showing a display operation in the learning text display window 17T accompanying Chinese reading aloud by the language learning support processing of the touch panel PDA 10.

  In FIG. 8, in the model display area 17M of the learning text display window 17T, the tone-delimited reduced text data Tm of the model Chinese example sentence is displayed together with the voice-recognized Pinyin notation Pin and displayed in the user display area 17U. Is displayed in contrast with the tone-delimited reduced text data Tu inputted by the user's voice following the model tone-delimited reduced text data Tm. Also in this case, as in the above-described embodiment, kanji whose tone differs between the exemplary text data Tm and the user text data Tu are identified and displayed by the yellow marker My.

  It should be noted that each processing method and database by the touch panel PDA 10 described in the embodiment of the tone display control device, that is, the language learning support process shown in the flowchart of FIG. 4 and the language learning support process shown in the flowchart of FIG. Each method such as voice recognition & tone detection processing and the exemplary sentence database 12c are programs that can be executed by a computer, such as a memory card (ROM card, RAM card, etc.), magnetic disk (floppy disk, hard disk, etc.). ), An optical disc (CD-ROM, DVD, etc.), an external recording medium 13 such as a semiconductor memory, and the like can be distributed. Then, the computer of the electronic device having the voice input / output units (15, 16) and the display unit (17) reads the program stored in the external recording medium 13 into the storage device (12), and uses the read program. By controlling the operation, it is possible to realize the language learning support function described in the embodiment and execute the same processing by the method described above.

  Further, program data for realizing each of the above methods can be transmitted on the network N in the form of a program code, and this program data is taken into the computer connected to the network N by the communication control unit 18. Thus, the language learning support function described above can also be realized.

  The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention when it is practiced. Further, each of the embodiments includes inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent elements are deleted from all the constituent elements shown in each embodiment or some constituent elements are combined in different forms, the problems described in the column of the problem to be solved by the invention If the effects described in the column “Effects of the Invention” can be obtained, a configuration in which these constituent requirements are deleted or combined can be extracted as an invention.

  Hereinafter, the invention described in the scope of claims of the present application will be appended.

[1]
A display unit;
A voice input means for inputting a sentence reading voice;
Voice recognition means for recognizing the voice input by the voice input means and converting it into text;
A tone determination unit that determines the tone of the voice input by the voice input unit for each word that is converted into text by the voice recognition unit;
Based on the text data converted into text by the voice recognition means and the tone for each word determined by the tone determination means, the tone for generating tone reduction text data by identifying each word in a form corresponding to the tone. Reduction text generation means,
Tone reduction text display control means for displaying the tone reduction text data generated by the tone reduction text generation means on the display unit;
A tone display control device comprising:

[2]
The display unit includes a model display area and a user display area that can be compared with the model display area.
The tone reduction text display control means displays the model tone reduction text data generated by the tone reduction text generation means in the model display area when the model reading voice is input by the voice input means, When the user's reading voice is input by the voice input unit, the user's tone reduction text data generated by the tone reduction text generation unit is displayed in the user display area.
The tone display control device according to [1], wherein

[3]
A tone that determines whether or not there is a difference in tone for each word determined by the tone determination unit between when the exemplary reading speech is input by the speech input unit and when the user's reading speech is input Difference judgment means,
When it is determined by the tone difference determination means that there is a difference in tone, the model tone reduction text data displayed in the model display area and the user tone reduction text data displayed in the user display area A different part identification display control means for identifying and displaying a part of the word determined to have the difference,
The tone display control device according to [2], comprising:

[4]
Audio output means for outputting audio;
When it is determined that the tone difference is determined by the tone difference determination unit, the model voice data corresponding to the word determined to have the difference is cut out from the model reading voice input by the voice input unit. Model voice output control means for outputting by the voice output means,
Generated by the tone reduction text generation means by the tone reduction text display control means when the user's reading voice is input by the voice input means after the model reading voice is output by the model voice output control means. User text redisplay control means for newly displaying the user's tone reduction text data in the user display area;
The tone display control device according to [3], comprising:

[5]
A recognition completion time extraction unit for each word for extracting a recognition completion time from the beginning of the voice input by the voice input unit for each word made into text by the voice recognition unit;
The model voice output control means inputs, by the voice input means, model voice data corresponding to a word determined to have a difference when the tone difference determination means determines that there is a difference in tone. Cut out based on the recognition completion time extraction means for each word extracted from the word-by-word recognition completion time extraction means and the recognition completion time of the previous word, and output by the voice output means,
The tone display control device according to [4], wherein

[6]
A program for controlling a computer of an electronic device having a display unit,
The computer,
A voice input means for inputting a sentence reading voice;
Voice recognition means for recognizing the voice input by the voice input means and converting it into text;
Tone determination means for determining the tone of the voice input by the voice input means for each word made into text by the voice recognition means;
Based on the text data converted into text by the voice recognition means and the tone for each word determined by the tone determination means, the tone for generating tone reduction text data by identifying each word in a form corresponding to the tone. Reduction text generation means,
Tone reduction text display control means for displaying the tone reduction text data generated by the tone reduction text generation means on the display unit;
Program to function as.

10 ... Touch panel PDA (tone display control device)
11 ... CPU
DESCRIPTION OF SYMBOLS 12 ... Memory 12a ... Language learning support program 12b ... Word voice recognition processing module 12c ... Model example sentence database 12d ... Model voice memory 12e ... User voice memory 12f ... Voice recognition text memory (model / user)
12g ... Word recognition completion time sampling memory (exemplary / user)
12h ... Word ending waveform registration table (exemplary / user)
12i ... Word-by-word tail tone registration table (exemplary / user)
12j ... Word tones reduction text memory (exemplary / user)
12k ... practice word setting memory 13 ... external recording medium 14 ... recording medium reading unit 15 ... voice input unit 16 ... voice output unit 17 ... touch panel display unit 17T ... learning text display window 17M ... model display area 17U ... user display area 18 ... Communication control unit 19m ... [exemplary] key 19i ... [example sentence] key 19u ... [user] key 19v ... [voice input] key 19p ... [practice] key G ... language learning support screen Tm ... exemplary tone-delimited reduction text data Tu ... User's tone delimiter reduction text data ap ... Delimiter My ... Yellow marker M ... Marker frame

Claims (6)

A display unit;
A voice input means for inputting a sentence reading voice;
Voice recognition means for recognizing the voice input by the voice input means and converting it into text;
A tone determination unit that determines the tone of the voice input by the voice input unit for each word that is converted into text by the voice recognition unit;
Based on the text data converted into text by the voice recognition means and the tone for each word determined by the tone determination means, the tone for generating tone reduction text data by identifying each word in a form corresponding to the tone. Reduction text generation means,
Tone reduction text display control means for displaying the tone reduction text data generated by the tone reduction text generation means on the display unit;
A tone display control device comprising: The display unit includes a model display area and a user display area that can be compared with the model display area.
The tone reduction text display control means displays the model tone reduction text data generated by the tone reduction text generation means in the model display area when the model reading voice is input by the voice input means, When the user's reading voice is input by the voice input unit, the user's tone reduction text data generated by the tone reduction text generation unit is displayed in the user display area.
The tone display control apparatus according to claim 1. A tone that determines whether or not there is a difference in tone for each word determined by the tone determination unit between when the exemplary reading speech is input by the speech input unit and when the user's reading speech is input Difference judgment means,
When it is determined by the tone difference determination means that there is a difference in tone, the model tone reduction text data displayed in the model display area and the user tone reduction text data displayed in the user display area A different part identification display control means for identifying and displaying a part of the word determined to have the difference,
The tone display control device according to claim 2, comprising: Audio output means for outputting audio;
When it is determined that the tone difference is determined by the tone difference determination unit, the model voice data corresponding to the word determined to have the difference is cut out from the model reading voice input by the voice input unit. Model voice output control means for outputting by the voice output means,
Generated by the tone reduction text generation means by the tone reduction text display control means when the user's reading voice is input by the voice input means after the model reading voice is output by the model voice output control means. User text redisplay control means for newly displaying the user's tone reduction text data in the user display area;
The tone display control apparatus according to claim 3, comprising: A recognition completion time extraction unit for each word for extracting a recognition completion time from the beginning of the voice input by the voice input unit for each word made into text by the voice recognition unit;
The model voice output control means inputs, by the voice input means, model voice data corresponding to a word determined to have a difference when the tone difference determination means determines that there is a difference in tone. Cut out based on the recognition completion time extraction means for each word extracted from the word-by-word recognition completion time extraction means and the recognition completion time of the previous word, and output by the voice output means,
The tone display control device according to claim 4. A program for controlling a computer of an electronic device having a display unit,
The computer,
A voice input means for inputting a sentence reading voice;
Voice recognition means for recognizing the voice input by the voice input means and converting it into text;
Tone determination means for determining the tone of the voice input by the voice input means for each word made into text by the voice recognition means;
Based on the text data converted into text by the voice recognition means and the tone for each word determined by the tone determination means, the tone for generating tone reduction text data by identifying each word in a form corresponding to the tone. Reduction text generation means,
Tone reduction text display control means for displaying the tone reduction text data generated by the tone reduction text generation means on the display unit;
Program to function as.

Images