KR101566013B1 - Method and system for improvement of the pronunciation accuracy and expressiveness during language beginners reading an e-book aloud - Google Patents

Abstract

The present invention relates to a method for improving the accuracy and expressiveness of utterances of an e-book reader of a language, and more particularly, it relates to a method and apparatus for recognizing utterance utterances of a language primer by recognizing a sentence utterance By analyzing the speech signal, it is possible to inform the beginner of the language by the response signal of how much the utterance accuracy of the utterance sentence is, to induce repetitive and intensive correct utterance utterance until reaching the utterance utterance accuracy, The present invention relates to a method and a system for providing improved utterance accuracy and expressive power through read aloud of an electronic book to help naturally enable a user to read an electronic book.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and system for improving the accuracy of speech and expressive power by reading an e-book,

The present invention is to improve the accuracy and expressiveness of utterance when reading an e-book of a language primer. As the voice recognition module is linked to the e-book implementation module, it recognizes the utterance of the utterance of the language primer and recognizes the voice signal of the recognized utterance By analyzing the sentence, it is possible to inform the beginner of the language by the response signal of how much the utterance accuracy of the utterance sentence is, and to induce the repetitive and intensive correct utterance utterance until reaching the utterance utterance accuracy. And more particularly, to a method and system for providing improved speaking accuracy and expressive power by reading an e-book to help a user read a book.

Recently, digital devices capable of driving electronic books such as electronic book terminals, smart phones, and tablet computers have become widespread, and the number of users who subscribe to electronic books using such digital devices is gradually increasing.

However, in the case of e-books using digital devices, when reading an e-book from a particle of an introductory language, the meaning of a book for acquiring the meaning of the text and the meaning of the expression exercise as a language can be used. However, There is no way to measure and evaluate sentence pronunciations, so that even if an e-book is read repeatedly, meaningful sentences can be learned and proper intonation and pronunciation can not be learned.

In addition, although an e-book has been proposed in which a simple reaction is performed by recognizing the pronunciation of a sentence in the conventional language, the read sentence, the unread sentence, the appropriately pronounced sentence, It is necessary to extract the data of usage statistical data related to the utterance utterance such as spoken language accuracy of the language beginner in order to analyze the data to help the language learning. There is a problem that reading e-books is not effective.

Korean Patent Publication No. 2012-0042298

Accordingly, the present invention has been made to solve the above-mentioned problems of the prior art, and it is an object of the present invention to provide an electronic book implementation module for a learner's voice recognition module to measure and evaluate electronic book sentences or word utterances of a language- And an object of the present invention is to provide a method and system for providing accurate pronunciation and expressive power by reading an electronic book that can analyze statistical data that helps language learning, such as initialization accuracy of a beginner.

According to an aspect of the present invention, there is provided a system for providing improved accuracy of speech and expressive power through read aloud of an electronic book, including an electronic book text implemented on an electronic book, an element recognizing that a user is speaking when the text is uttered, An electronic book content including an element for recognizing completion of speech recognition and an element for recognizing a level of speech recognition accuracy; An electronic book implementing module installed in a digital device capable of inputting and outputting a voice signal and a video signal and having a function of storing and processing data and implementing the electronic book contents on the digital device; And a read aloud sound recognition module installed in the digital device and interworking with the electronic book implementation module to receive a voice signal when the user utters a text and analyze and evaluate a user's utterance accuracy through the voice recognition module Feature.

According to another aspect of the present invention, there is provided a method for providing improved utterance accuracy and expressive power through read aloud of an electronic book, comprising the steps of: inputting utterance speech signals of sentences or words of uttered electronic books; An analysis and discrimination step of analyzing and discriminating the speaking accuracy in the speech recognition module of the speech voice signal; A storing step of transmitting the presence or absence of a reaction according to the firing accuracy to the control unit and storing the firing accuracy data in the statistical database; An instruction step of causing the control unit to call and output a content corresponding to the speaking accuracy from a content database and output a display of a speech completion sentence; And a display step of displaying a statistical display related to a spoken sentence display, a spoken accuracy indication, and a spoken word in an electronic book implementation module so as to be realized on a digital device.

As described above, the method and system for improving the speaking accuracy and the expressive power by reading the e-book of the present invention can help improve the accuracy of speech by measuring and evaluating sentence or word utterance only by read- In order to improve the accuracy of speaking, it is effective to improve the speaking accuracy and the expressive power of the beginner in terms of the effectiveness of the language learning by acquiring the meaning of the language and acquiring the expressive power through the process of repeated speaking.

In addition, by providing the statistical data extracted from the accumulated speech data of the first language learner, it is possible to recognize the improved state of the current sentence speech accuracy of the first language learner so that the continuous and intensive speech practice is performed, Can be induced naturally.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram showing the configuration of a system for providing improved speaking accuracy and expressive power through read aloud of an electronic book according to the present invention; FIG.
2 is a schematic diagram showing a configuration of a read aloud voice recognition module which is an embodiment of the present invention;
3 is a diagram showing an embodiment of an electronic book implemented in a digital device through the present invention.
FIG. 4 is a schematic flow diagram illustrating a method for providing speaking accuracy and expressive power enhancement through e-book reading according to the present invention; FIG.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram showing a configuration of a system for providing improved speech accuracy and expressive power through read aloud of an electronic book according to the present invention, and FIG. 2 is a schematic diagram showing a configuration of a readable voice recognition module as an embodiment of the present invention.

1, the present invention includes a voice input unit 110, a voice recognition module 120, a controller 130, an electronic book implementation module 140, a statistics database 150, a content database 160, And an output unit 170.

The voice input unit 110 refers to an input module such as a microphone that is provided in a digital device in which the system of the present invention can be implemented and enables voice and sound input. Quot;) corresponds to a configuration in which a speech voice signal generated according to reading is received.

Here, the digital device may be any one of an electronic book exclusive terminal, a smart phone, and a tablet PC through which the electronic book application program can be run through the system of the present invention, but is not limited thereto.

The read voice recognition module 120 receives the speech voice signal received from the voice input unit 110 and analyzes the voice speech signal to calculate the accuracy of the voice recognition accuracy and the level of the voice recognition accuracy.

The reader voice recognition module 120 recognizes a sentence utterance of the user's e-book, processes the recognized utterance utterance voice signal, and if the recognized utterance sentence is included in the contents of the corresponding e-book, Identify which part of the sentence.

The sentence recognition of the reader / writer speech recognition module 120 can be provided through comparative analysis with e-book contents previously stored in the content database 160 described below.

2, the read voice recognition module 120 includes a voice input processor 121, an end point detector 122, a phoneme separator 123, a phoneme scoring unit 124, An endpoint re-detection unit 125a, a rejection processing unit 125b, a quantitative evaluation unit 125c, and an utterance score determination unit 126. [

The speech input processor 121 receives the speech voice signal of the user transmitted from the speech input unit 110 and converts the speech voice signal data inputted for a predetermined time into data of the voice feature vector type, (122).

The voice input processor 121 converts the voice signal data received from the voice input by the user into data of a voice feature vector form for use in a lower module of the voice recognition module 120, To the end point detection unit 122. The end point detection unit 122 detects the end point of the voice feature vector data.

Here, the voice feature vector data refers to conversion data for each interval by dividing input voice signal data at predetermined intervals. The predetermined interval is an interval having a time unit of 10 msec, 16 msec, 20 msec, etc., and the voice feature vector values included in the interval are extracted by dividing the data by a fraction of this time unit interval. For example, if a signal of 8 kHz is input at intervals of 10 msec, there are 80 data samples in 10 msec intervals, and the 80 data samples are extracted as one voice feature vector data.

The end point detection unit 122 extracts a starting point and an end point, that is, a valid part of the utterance speech signal, from the speech speech signal converted into the speech feature vector data from the speech input processing unit 121, To the phoneme separator (123).

That is, the end point detector 122 detects the start and end points of the voice feature vector data transmitted from the voice input processor 121, for example, an EPD (End-Point-Detection) module. The end point detector 122 transmits the voice feature vector data of the valid part between the detected start point and end point to the phoneme separator 123.

The end point detector 122 detects voice feature vector data transmitted from the end point re-detector 125a to detect a start point and an end point from the voice feature vector data transmitted from the voice input processor 121 And detects the start point and the end point again.

Meanwhile, the phoneme separator 123 detects the start and end points transmitted from the end point detector 122 and compares the values of the valid speech feature vector data between the start and end points with the reference phoneme model to determine a phoneme- The start address and the end address of the voice feature vector data are recorded and the number of intervals determined as phonemes is recorded, and as a result, the voice feature vector data are separated in units of phonem (minimum phonetic unit). The speech characteristic texture data separated by the phoneme-by-phoneme unit is transmitted to the phoneme-by-phoneme scoring unit 124.

The phoneme-by-phoneme scoring unit 124 compares phonetic feature vector data separated by phonemes in the phoneme separator 123 with a scoring-based phoneme model to calculate a score, And transmits the phoneme-by-phoneme score data to the trust score determining unit 125.

If the average value of the voice feature vector data transmitted from the phoneme-by-phoneme scoring unit 124 and the score data of the phoneme matched by the matched phoneme score data are within the reliability range of the reference phoneme model And transmits the voice feature vector data to the quantitative evaluation unit 125c.

If the reliability of the reference phoneme model is lacking, the reliability score determiner 125 transmits the average value of the score data to the end point detector 122 as described above.

If the average value of the score data for each phoneme to be matched is significantly lacking in the reliability of the reference phoneme model, the trust score determining unit 125 transmits the reliability score to the rejection processing unit 125b.

That is, if it falls within the range of the reliability, it is discriminated as a properly separated phoneme. If it is a lack of reliability, it is discriminated as a non-properly separated phoneme.

The reliability range of the trust score discrimination unit 125 can be used by extracting the statistic value by an average value of the phoneme score or the like.

The endpoint re-detection unit 125a performs phoneme separation for the re-detection of the start point and the end point of the voice feature vector data (determined as a non-properly separated phoneme) corresponding to the lack of reliability transmitted from the trust score discrimination unit 125 (123) and phoneme-by-phoneme scoring unit (124) and transmits the data to the endpoint detector (122).

Then, the rejection processing unit 125b terminates the processing of the voice feature vector data (determined to be not a phoneme) that corresponds to a significant lack of reliability transmitted from the trust score determining unit 125. [

The quantitative evaluation unit 125c compares the voice feature vector data within the reliability range transmitted from the reliability score determination unit 125 with the speech evaluation model made based on the statistical data evaluated and calculated by the experts The quantitative accuracy of the utterance of the corresponding text is calculated.

Finally, the utterance score determiner 126 replaces the voice feature vector data and the quantitative utterance accuracy value transmitted from the quantitative evaluator 125c with utterance scores set according to the criteria of the application used.

The ignition accuracy value and the utterance score thus calculated are stored in the statistical database 150 by the process of the read aloud voice recognition module 120 and the electronic book implementation module 140 The user can confirm the utterance accuracy of the user's utterance.

The e-book implementation module 140 is a kind of application software for implementing an electronic book on a digital device. The e-book implementation module 140 includes text and electronic information for an electronic book stored in the content database 160 So that the basic images for constituting the book are displayed on the digital device.

In addition, the electronic book implementation module 140 is interlocked with the read-aloud voice recognition module 120 by the control unit 130 so that the user can read various kinds of electronic books and respond in various ways on the digital device correspondingly .

That is, the electronic book implementation module 140 recognizes elements that can recognize that the user is speaking when the user utters the text, elements that recognize the completion of the speech recognition, and recognizes that the speaking accuracy is above or below a certain level The contents corresponding to the voice recognition are called from the contents database 160 storing various contents including the contents of the contents, and the function of displaying the electronic book is displayed so that the elements corresponding to the utterance are reflected.

The calling function of the electronic book implementation module 140 is performed from the control unit 130 that substantially interlocks the respective components.

The statistical database 150 stores the number of sentences and sentences of the text utterance recognition of the electronic book implemented in the electronic book implementation module 140 and the utterance accuracy of the utterance recognition completion sentences by the reader speech recognition module 120, And stores the statistical data such as the completion date and time of the utterance recognition completion of the utterance recognition sentence. The controller 130 calls and outputs the data corresponding thereto at the request of the controller 130.

Also, the statistical database 150 may include accumulation data on the pronunciation history of the user. Here, the accumulation data refers to cumulative data on the analysis result of the speech recognition module 120 for the user's utterance, and can be displayed as a digital device through the electronic book implementation module 140 described above Through this accumulation data, it is possible to deduce the improvement of the learning ability including the user's speaking accuracy.

In particular, the statistical database 150 stores statistical data of all users who read the e-book, not only the user but also the statistical data of all users who read the e-book, It becomes possible to provide statistical information on a specific range of a specific range of the utterance accuracy according to the number of completed sentences, and it becomes possible to display the utterance accuracy statistics according to the number of completed utterance recognition sentences of the user.

The controller 130 calculates the statistics of each time period including day, week, and month based on the statistical data of all users, stores the statistics of the time period in the statistical database 150 again, Let the user display it on the e-book.

Accordingly, the user can easily check the statistical data of the user's pronunciation accuracy as well as the statistical data of the user's own pronunciation, thereby comparing the level of the pronunciation accuracy of the user.

The content database 160 includes text and basic images constituting the electronic book, factors displayed on the digital device, for example, elements that can recognize that the user is speaking when he utteres the text, , And elements that recognize that the utterance accuracy is above or below a certain level. In addition, the content database 160 stores contents including images, sound effects, and the like for providing visual and auditory effects when a user reads an electronic book.

The content database 160 may also be displayed in conjunction with the implementation of the electronic book by interfacing with the electronic book implementation module 140 and calling and outputting corresponding contents according to the request of the controller 130. [

The output unit 170 outputs a sound effect, an image, and an image of a digital device. The output unit 170 is, for example, an output module such as a speaker and a display of a digital device. The output unit 170 provides any one of a sound effect, an image, and a composite output corresponding to a voice signal input from the microphone of the voice input unit 110, and the output data is output from the content database 160 And outputs the extracted data.

Meanwhile, the controller 130 controls interlocking between the read aloud voice recognition module 120 and the electronic book implementation module 140, which are elements for implementing the present invention. The interoperability between the electronic book implementation module 140 and the statistical database 150 and the interoperability between the electronic book implementation module 140 and the content database 160 are controlled.

In addition, the control unit 130 performs a function of interlocking between the respective constitutions, and calls and outputs contents corresponding to the utterance accuracy in the contents database 160 to display the utterance completion sentences, display the utterance accuracy, And so on to be implemented together with other contents in the e-book through the e-book implementation module 140.

The interlocking function of the controller 130 may be a request made by a user's operation or an automatic signal generated by inter-constituting data processing.

3 is a diagram showing an embodiment of an electronic book implemented in a digital device through the present invention.

As shown in FIG. 3, when an electronic book implemented by the electronic book implementation module 140 of the present invention starts to be uttered in the electronic book sentence of the user, the pre-utterance recognition text 311 is divided into The shape of the text is changed to the speech recognition completion text 322. [

In addition, the position of the ignition recognition completion text indicates the end point of the part where the speech recognition is completed by the indicator 323 so that the user can check it, so that the initial part of the language recognizes the already ignited part.

In particular, the output unit 170 described above is output as a speaker, an image, or the like. For example, as shown in FIG. 3, a speech recognition score score correlating reaction image 324 and a speech recognition score score correlating reaction sound (325), it is possible to recognize a form of a certain score or more of the accuracy of utterance of a language primer so as to be easier and more interesting. 3, not shown here, is a firing input volume indicator.

In addition, a history bar 326 for recognizing the speech recognition state can be implemented at the bottom of the e-book, so that it is possible to recognize the flow of the spoken accuracy of all or a part of the electronic book currently being uttered by the language beginner.

As described above, according to the present invention, the user can perceive the accuracy of the text utterance of the user in real time by the system and method flow according to the present invention in the process of uttering the electronic book text implemented in the electronic book implementation module 140 Although not shown in the figure, the e-book implementation module 140, in response to a user's request, sets a relative position of the user's utterance accuracy And the absolute position (the average number of cumulative speech recognition completed sentences and the average of the utterance accuracy of the user) can be recognized.

4 is a schematic flow chart illustrating a method for providing speech accuracy and expressive power enhancement through e-book reading according to the present invention. Referring to FIG. 4, a description will be given of a method for providing improved speaking accuracy and expressive power by reading an electronic book according to the present invention.

First, a speech voice signal of a sentence or a word of an electronic book uttered by the user is input. When the user reads the sentence of the electronic book through the digital device, the speech input unit 110 inputs the speech voice signal of the user, and transmits the speech voice signal to the speech recognition module 120. (S100)

The portable speech recognition module 120 receives the speech speech signal and processes the speech speech signal to analyze and determine the speech accuracy. The analysis and the determination of the utterance accuracy are implemented by the configuration of the voice recognition module 120 of the voice recognition system 120 described above, and a detailed description thereof will be omitted in step S200.

In this way, the speech accuracy data according to the user's electronic book utterance is stored in the statistical database 150 and accumulated as a basic information for recognizing the improved state of the utterance accuracy by providing the user's utterance statistical data. S300)

In addition, the control unit 130 commands and outputs the content corresponding to the speaking accuracy from the content database 160 to output the speaking completion sentence, the speaking accuracy, and the statistical display related to the speech, 140 in order to implement the electronic book together with other contents (S400)

The electronic book implementation module 140 implements an electronic book by implementing a command of the control unit 130 so as to realize a display of a spoken sentence, a display of a speaking accuracy, and a display of statistics related to a speech.

Here, in the display process on the electronic book including the above-mentioned spoken sentence display, spoken accuracy display, and statistical display related to the spoken word, it is possible to display a sentence or a word already recognized, display a sentence or word that has not yet been recognized as a spoken word, Abnormal Speech Recognition A sentence or word can be displayed, or a sentence or word recognized less than a certain utterance accuracy can be displayed.

This is realized by interlocking the voice recognition module 120 and the electronic book implementation module 140 described above and is displayed on the electronic book by calling and outputting various factors previously stored in the contents database 160 will be.

In particular, when a sentence or word already recognized as a speech is recognized, the sentence in the content database 160 is compared with the speech-based sentence recognized by the read-aloud speech recognition module 120 and scored by the quantitative evaluation, The phonetic unit of a sentence is output in a form different from the character before the speech recognition. The speech recognition unit 120 recognizes the utterance position of a sentence or a word uttered by the user in accordance with the function of the speech recognition module 120 .

In addition, according to an embodiment of the present invention, when the user utters a sentence or a word included in the e-book plot, the user may further display the utterance statistics of the user and other users.

In the statistical display step, it is calculated based on the statistical data of the user previously stored in the statistical database 150 described above and the statistical data of the user of the electronic book. The accuracy of the speaking accuracy according to the number of utterance- Of the total number of utterances of the corresponding e-book, as well as the utterance accuracy according to the number of completed utterance sentences.

In addition, the statistics of the utterance accuracy according to the number of utterance completion sentences of the user are displayed. In particular, such various statistical data are displayed in a time period including day, week, and month according to the user's request.

Here, in displaying the speech accuracy statistics according to the number of utterance recognition completed sentences of the user, data on the utterance accuracy of the utterance sentence recognized by the read-aloud speech recognition module 120 and scored by the quantitative evaluation, In the data on the number of sentences, the utterance accuracy, the average utterance accuracy, and the cumulative utterance completion sentence number are displayed.

Accordingly, the user can easily recognize the current speaking ability of the user in the speaking practice through the reading of the electronic book.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification, but should be defined by the claims.

110: voice input unit 120: voice recognition module
130: controller 140: electronic book implementation module
150: statistics database 160: content database
170: output unit 121: audio input processing unit
122: end point detection unit 123: phoneme separation unit
124: Phoneme scoring unit 125: Trust score determining unit
125a: End point re-detection unit 125b: Rejection processing unit
125c: Scalability evaluation unit 126: Speech score determination unit

Claims (10)

An electronic book text including an electronic book text implemented on an electronic book, an element recognizing that the user is speaking when the text is uttered, an element recognizing completion of speech recognition and an element recognizing a level of the speech recognition accuracy, ;
An electronic book implementing module installed in a digital device capable of inputting and outputting a voice signal and a video signal and having a function of storing and processing data and implementing the electronic book contents on the digital device; And
And a reading voice sound recognition module installed in the digital device and interfacing with the electronic book implementation module to receive a voice signal when the user utters the text and to analyze and evaluate the user's speaking accuracy through the voice recognition module To improve the accuracy of speaking and expressive power by reading e-books.
The method according to claim 1,
Statistical data such as the utterance accuracy of the uttered utterance recognition sentence by the voice recognition module, the date and time of completion of utterance recognition completion of the utterance recognition sentence, and accumulation data on the pronunciation history of the user are stored, A statistical database output by the statistical database; And
A text and an image constituting the electronic book, and elements capable of recognizing that the user is speaking when the user utters the text as expressed on the digital device, elements recognizing the completion of the speech recognition, A content database storing contents including elements for recognizing that the electronic book is a character string, a character string, a character string, a character string, a character string, a character string, and a character string.
The method according to claim 1,
The read voice recognition module includes:
A speech input processor for inputting speech speech signals transmitted from the speech input unit at predetermined intervals and extracting and converting the speech speech data into data of a speech feature vector type,
An endpoint detector for detecting a start point and an end point of a speech feature vector transmitted from the speech input processor,
A phoneme separator for separating phoneme feature vector data determined as a phoneme among valid phoneme feature vector data between a start point and an end point detected at the start point and end point transmitted from the end point detector,
A phoneme-by-phoneme scoring unit for calculating a score by comparing voice feature vector data separated by phonemic units in the phoneme separator to a scoring-based phoneme model;
A reliability score discrimination unit for discriminating whether the average value of the voice feature vector data transmitted by the phoneme scoring unit and the score data of each phoneme matching among the phoneme score data matched therewith is included in the reliability range of the predetermined reference phoneme model
A quantitative evaluation unit for comparing the voice feature vector data within the reliability range transmitted from the reliability score discrimination unit with a speech evaluation model using statistical values calculated by an utterance evaluation expert to calculate a quantitative speaking accuracy value of the uttered text;
And an utterance score discriminator for replacing the voice feature vector data and the quantitative utterance accuracy value transmitted from the quantitative evaluator with an utterance score so as to comply with a criterion of a using application. Providing system.
The method of claim 3,
Wherein the voice input processing unit comprises:
Wherein the utterance voice signal data is divided by 10 msec to 20 msec, and the feature vector values included in the interval are divided by the utterance time.
The method of claim 3,
The read voice recognition module includes:
An endpoint re-detection unit which receives the voice feature vector data corresponding to the lack of reliability from the reliability score discrimination unit and removes data recorded in the phoneme separator and phoneme scoring unit from the voice feature vector data and transmits the removed voice feature vector data to the end point detector;
Further comprising a reject processing unit for receiving the voice feature vector data that is determined not to be a phoneme by the reliability score determination unit and terminating the processing of the voice feature vector data. Providing system.
An input step of inputting a speech voice signal of a sentence or word of an electronic book uttered by the user;
An analysis and discrimination step of analyzing and discriminating the speaking accuracy in the speech recognition module of the speech voice signal;
A storing step of transmitting the presence or absence of a reaction according to the firing accuracy to the control unit and storing the firing accuracy data in the statistical database;
An instruction step of causing the control unit to call and output a content corresponding to the speaking accuracy from a content database and output a display of a speech completion sentence; And
And displaying the statistical display related to the utterance sentence, the utterance accuracy, and the utterance in the electronic book implementation module so as to be realized on the digital device.
The method according to claim 6,
In the display step
Steps to display sentences or words that have already been ignited
Steps to display sentences or words that have not yet been ignited
Displaying a sentence or word recognized as a speech recognition error over a constant speaking accuracy; and
The method comprising the steps of: displaying a sentence or a word recognized as a speech less than a predetermined utterance accuracy; and providing a speech accuracy and an expressive power improvement through an e-book read aloud.
8. The method of claim 7,
The step of displaying a sentence or a word already recognized as a speech is
The phoneme-unit sentence recognized by the read-aloud speech recognition module and scored by the quantitative evaluation is compared with the sentence of the content database to output the phoneme-unit letter of the sentence, And recognizes the utterance position of a sentence or a word of the person himself / herself.
The method according to claim 6,
Further comprising a statistical display step of displaying the utterance statistics of the user and other users when the user utters a sentence or a word contained in the e-book plot,
The statistical display step
A step of displaying the average statistic of speaking accuracy according to the number of utterance recognition sentences of all users who read the e-book
A certain range of utterance accuracy according to the number of utterance recognition sentences of all users who read the e-book, a step of displaying statistics of the upper group
Displaying the speech accuracy statistics according to the number of utterance recognition completion sentences of the user, and
And displaying each statistic in a time period including a day, a week, and a month. The method as claimed in claim 1, further comprising:
10. The method of claim 9,
Wherein the step of displaying the utterance accuracy statistics according to the number of utterance recognition completion sentences of the user
In the data on the utterance accuracy of the utterance sentence recognized by the voice recognition module and scored by the quantitative evaluation, and the data on the number of uttered utterance completion sentences, the utterance accuracy, average utterance accuracy, cumulative utterance And the number of completed sentences is displayed so that the speaking ability of the user himself / herself can be recognized in the speaking practice through reading the electronic book of the user.

Images