KR20010107035A - An internet english study service method for voice recognition and voice synthesis - Google Patents

Abstract

According to the present invention, when a user interactively sees the pronunciation and mouth of a native speaker who is output as a sound and a video on the word 38 or sentence 37, the analysis and evaluation of the user's English pronunciation are fed back in real time. The present invention relates to an Internet English learning service method using voice recognition and speech synthesis, which allows a three-dimensional English learning to be performed while the screen is output as a voice waveform 28 and a score 29, wherein the user has a member ID 20 and a password 21. If you enter the member authentication screen step of loading the final learning data (S1), and if the sentence (37) or the word 38 is clicked, the sound and animation screen is output, the expression collection screen step (S2) and the corresponding content is translated Selecting the word collection screen step (S3), the sentence (37) or the word (38) provides the sound and video of the native speaker's pronunciation and mouth shape while the native speaker's voice wave (28) is output and when the user pronounces for the sentence 37, the corresponding speech waveform 28 is compared with the native speaker's speech waveform 28 and fed back while the score (29) is calculated and output expression screen step ( S4) and the word practice screen step (S5), and when the user clicks on the sentence (37) or word (38) and the image and pronounces the answer, the score (29) for the output of the self-evaluation screen step (S6) And, if each word 38 is clicked, it is characterized in that it comprises a phonics screen step (S7) and the like to be output as a video screen native speaker pronounced.

Description

An internet english study service method for voice recognition and voice synthesis}

The present invention relates to an Internet English learning service method using voice recognition and voice synthesis, which can conveniently learn words and sentences interactively with a computer using voice recognition technology and voice synthesis technology on the Internet. Analysis and evaluation of the user's English pronunciation when the user interactively sees the pronunciation and mouth of the native speaker who outputs the sound or video of the word or sentence on the internet-only web browser combined with recognition and voice synthesis technology The present invention relates to an English learning service method using speech recognition and speech synthesis, which is fed back in real time, so that three-dimensional English learning can be performed while the screen is output as voice waveforms and scores.

In general, English learning methods up to now have been mainly conducted as face-to-face lectures between a single provider and single or multiple learners, but the learners have different levels of understanding as well as one-way injection learning to suit each individual learner. There was a limit that it was difficult to correctly learn English.

In addition, due to the popularization of the Internet due to the development of information and communication, educational information and cyber learning related to English are provided on the website through the Internet network. Aside from being able to be provided conveniently without being particular, there was a considerable distance from the level of learning by talking directly with native speakers.

Accordingly, the present invention has been invented in view of such circumstances, and when the user interactively watches the sound of the word or sentence provided from a web server connected to the Internet network, and the pronunciation and mouth-shaped video of the native speaker, Internet English learning service method using voice recognition and speech synthesis that enables real-time and three-dimensional English learning such as direct dialogue with native speakers by providing feedback on real time with accurate analysis and evaluation of English pronunciation. The purpose is to provide.

1 is a system configuration for implementing the Internet English learning service method using the speech recognition and speech synthesis of the present invention.

2a and 2b is a conceptual view of speech recognition technology and speech synthesis technology applied to the Internet English learning service method using the speech recognition and speech synthesis of the present invention.

3a to 3g is a homepage member authentication and each learning configuration operation screen flowchart of the Internet English learning service method using voice recognition and speech synthesis according to the present invention.

Figures 4a to 4h is an example of the initial screen of the member authentication of the Internet English learning service method using the voice recognition and speech synthesis according to the present invention and the web screen for each learning configuration step.

Explanation of symbols on the main parts of the drawings

10: web server 11: voice recognition engine

12: voice synthesis engine 13: computer (user)

20: Member ID 21: Password

22: View All 23: Help

24: Notice 25: Search

26: Bulletin Board 27: Character Introduction

28: speech waveform 29: score

30: collection of expressions 31: collection of words

32: expression exercise 33: word practice

34: self-assessment 35: phonics

36: playground 37: crest

38: words

English learning service method using the voice recognition and speech synthesis of the present invention for achieving the above object, the member ID and password registered by the user through the initial screen of the Internet-only web browser combined with voice recognition and speech synthesis technology If you type in the screen to the screen, the membership authentication screen step is automatically loaded to the final learning data, and if you click on the sentence, the sound and animation screen output, and the contents collection screen step is automatically translated into the text, and the corresponding words If you click, the sound and animation screen is displayed and the contents are automatically translated into text.Then, select the sentence and click the native speaker icon.The native speaker's voice along with the sound and video of the native speaker's pronunciation and mouth shape When the waveform is displayed and the user pronounces that sentence When the user's voice waveform is displayed on the screen and compared with the native speaker's voice waveform, feedback is automatically calculated and the corresponding score is automatically calculated and the screen is displayed. When the user selects the word and clicks a native speaker icon, the native speaker's pronunciation and mouth shape are displayed. The voice waveform of the native speaker is displayed on the screen along with the sound and the video for the user. When the user pronounces the sentence, the user's voice waveform is output on the screen and the feedback is compared with the native speaker's voice waveform and the score is automatically calculated and displayed. A practice screen step, and a self-assessment screen step in which a score is stored in a database while the user clicks on a corresponding sentence, word, and image and pronounces the corresponding answer; Clicking on each word includes a phonics screen step of outputting a video of a native speaker pronounced as a video, a song / lyric word game with a song / lyric along with a picture, and a playground screen step with other pictures. Characterized in that made.

Particularly, in the expression practice screen step, the user's voice input is recognized and recorded as a browser built-in voice recognition engine as in the word practice screen step.

Hereinafter, the principle of implementing the Internet English learning service method through the Internet-only web browser in which voice recognition and voice synthesis technology is applied according to the present invention (dedicated web browser implementation) will be described in detail with reference to the accompanying drawings.

Figure 1 shows a system configuration for implementing the Internet English learning service method using the voice recognition and speech synthesis of the present invention, Figure 2a is a voice applied to the Internet English learning service method using the voice recognition and speech synthesis of the present invention Figure 2a shows an implementation concept of the recognition technology, Figure 2a shows an implementation concept of the speech synthesis technology applied to the Internet English learning service method using the speech recognition and speech synthesis of the present invention.

As shown in the figure, the present invention according to the user to look at the sound and mouth of the native speaker's pronunciation for each sentence 37 and the word 38 provided in the corresponding learning screen of the web server 10 through the Internet network When the user's voice is input through the microphone of the computer 13 and recorded by the built-in voice recognition engine 11 of the browser, the corresponding voice waveform of the user's input voice is fed back and the screen is displayed as the native voice waveform 28 of the native speaker. At the same time, the output is compared and calculated so that the corresponding score 29 is displayed so that the user's pronunciation can be compared with the correct pronunciation of the native speaker.

In addition, the corresponding data inputted by the user connected to the web server 10 is transmitted to the web server 10 is automatically recognized by the speech recognition engine 11 (ASR: Automatic Speech Recognition) to the corresponding speech waveform ( At the same time, the text for the sentence (37) or the word (38) clicked by the user is automatically converted into sound by the voice synthesis engine 12 (TTS: Text To Speech). The system is configured to be heard and conveyed to the user through.

On the other hand, Figures 3a to 3g together with the home screen member authentication operation screen of the Internet English learning service method using the voice recognition and speech synthesis according to the present invention, the expression collection 30, the word collection (31), expression training (32) , Word practice (33), self-assessment (34), phonics (Phonics) 35, the corresponding operation screen flow for the playground (36), respectively, Figures 4a to 4h are the voice recognition and An example of the initial screen of the member authentication of the English learning service method using voice synthesis and the web screen for each step of the learning configuration are illustrated.

The voice recognition and voice of the present invention can receive pronunciation lessons and self-evaluation in real time by viewing the native speaker's sound and mouth-shaped video on the word 38 and sentence 37 through the operation screen flow and web screen as described above. Internet English learning service method using the synthesis, the user authentication screen screen (S1), the sentence 37 or a word (S1) that the final learning data is loaded when the user enters the member ID (20) and password (21) 38) If the sound and animation is displayed on the screen, the expression collection screen step (S2) and the word collection screen step (S3) and the sentence (37) or the word (38) are selected, and the native speaker's pronunciation and When the native speaker's voice waveform 28 is output along with the sound and video of the mouth and the user pronounces the sentence 37, the corresponding voice waveform 28 is compared with the native speaker's voice waveform 28 and fed back. 29 Expression practice screen step (S4) and word practice screen step (S5) is outputted and the user clicks on the sentence (37) or word (38) and the image and pronounces the answer, the score (29) is outputted The score is stored in the self-evaluation screen step (S6), and when each word 38 is clicked is made, including the phonics screen step (S7) and the like output the screen of the native speaker pronunciation animation.

First, as shown in FIGS. 3A, 4A, and 4B, in the member authentication screen step (S1), a user who has registered as a member ID 20 and the member ID 20 are displayed on the initial screen of the Internet-only web browser to which voice recognition and voice synthesis technology are applied. If you enter the password (21), the last learning data that was learned at the last visit while branching to the corresponding learning screen is automatically loaded, and if the subscriber visits the first visit 〃hello, ○○○ members. Congratulations on your first visit. ○○ ○○ in 2000 You will learn 1C. Text, is provided, and from the second visit 〃Hello, ○○○ member. ○ It is the fourth visit. ○ ○ ○ 2000 The text you will learn is 1C.

In addition, at the bottom of the homepage, additional features such as full view (22), help (23), notice (24), search (25), bulletin board (26), character introduction (27) are given, and the current learning at the top of the homepage On the basis of the screen, it is conveniently used because it is provided with operation functions such as 〃back〃, 〃forward〃, stop〃, 〃refresh〃, 〃print〃, 〃home〃.

3B and 4C, in the expression collection screen step S2, when a speech bubble of a sentence 37 such as “What's for breakfast” is clicked, the sound and animation (video) of the corresponding English sentence are output. At the same time, if you click the helper icon for an English sentence, the screen is automatically translated into the Hangul sentence for that English sentence.

In addition, the background of the animation for the sentence (37) is blinking and at the same time output sound effects such as the sound of birds chirping so that the learning motivation can be concentrated, and click the close is configured to initialize to the icon state.

3C and 4D, in the word collection screen step S3, when a word 38 such as “Lunch” is clicked, the sound and animation (video) of the corresponding English word are displayed on the screen, and at the same time, the English word is displayed. If you click on the helper icon, the screen is automatically translated into Korean sentences.

In addition, as shown in the expression collection screen step (S2), the background of the animation for the sentence 37 is blinking, and at the same time it is desirable that the learning motivation can be concentrated while outputting sound effects such as the sound of birds singing. Do.

3D and 4E, in the expression practice screen step S4, when the corresponding sentence 37 is selected and a native speaker icon is clicked at the same time, the native speaker's voice waveform along with the sound and video of the native speaker's pronunciation and mouth shape are displayed. 28 is displayed on the screen, and when the helper icon is clicked, the sentence 37 and the speech waveform 28 that are highlighted are provided.

Subsequently, when the user clicks on the user icon of the expression exercise 32 and carefully pronounces the native speaker's voice waveform 28 for the sentence 37, the user's voice waveform 28 is displayed on the screen. The score 29 is automatically calculated while being fed back compared with the speech waveform 28, and the screen is output as scores such as # 70-79 \, # 80-89 \ and # 90-100 \.

3E and 4F, in the word practice screen step S5, when the word 38 is selected and a native speaker icon is clicked at the same time, as in the expression practice screen step S4, the pronunciation and mouth of the native speaker are shown. A native speaker's voice waveform 28 is output along with the sound and video for the screen.

If the user clicks on the user icon of the word practice 33 and carefully pronounces the native speaker's voice waveform 28 for the word 38, the user's voice waveform 28 is displayed on the screen. The score 29 is automatically calculated while being fed back compared with the speech waveform 28, and the screen is output as scores such as # 70-79 \, # 80-89 \ and # 90-100 \.

In particular, in the expression practice screen step (S4) and the word practice screen step (S5), the user's voice input for the sentence 37 and the word 38 is recognized and recorded as a browser built-in voice recognition engine (11) The corresponding speech waveform 28 and the calculated score 29 are output on the screen, and at the same time, the text for the sentence 37 or word 38 clicked by the user is automatically sounded by the browser-integrated speech synthesis engine 12. As it changes, the function is configured to be delivered to the user accurately.

And as shown in Figure 3f, in the self-assessment screen step (S6) when the user clicks on the sentence 37 or the word 38 and the image and pronounces the corresponding answer, the user's voice is input to the built-in voice recognition engine (11) is automatically output to the audio waveform 28, the corresponding calculated score (29) is displayed on the screen, if there is no problem, the score data is ASP (Active Server Page) file using a script on the web server The member data and the corresponding scores (29) are managed as they are stored in the database, and by clicking the result view, the scores of each part are graphed so that users or their parents can easily view the learning achievements.

In addition, as shown in FIGS. 3G and 4G, when each word 38 is clicked in the phonics screen step S7, a video screen positioned at the center of the word 38 is provided, and the native speaker pronounces a mouth-like sound. The video is shown in front and side, and at the same time, the text is provided along with a description screen on the left side of the video screen.

In addition, as shown in Figure 4h, the playground screen step (S8) is provided with a number of playground spaces, such as to check the song / lyrics or word game provided with the song / lyrics along with the picture, and other pictures to make the intermediate play area It helps to solve the learning fatigue.

On the other hand, the above-described embodiment is a preferred embodiment for facilitating the understanding of the implementation principle of the English learning method using speech recognition and speech synthesis according to the present invention, the present invention various screen configuration without departing from the implementation principle of the present invention This is possible, and in particular, the sound for the corresponding sentence and word of the native speaker can be multiplied by the male or female variously selected and doubled the variable learning effect, and all possible implementations without departing from the implementation principle of the present invention. It is to be understood that modifications of examples and embodiments are within the scope of the present invention.

As described above, according to the English learning method using speech recognition and speech synthesis according to the present invention, a sound of a corresponding word or sentence provided from a web server connected to an Internet network, and a corresponding pronunciation and mouth-shaped video of a native speaker If you follow the conversation interactively while watching the user, the accurate analysis and evaluation of the user's English pronunciation is fed back in real time and the sound wave form and score can be realistic and three-dimensional English learning such as talking directly with the native speaker.

In addition, when viewed in the linguistic sense, users simply click on the desired words, sentences and conversations to hear the native speaker's pronunciation, as well as pronunciation words, sentences, As soon as feedback is given on the contents of conversation, a variety of English learning methods such as pronunciation score, pronunciation waveform, comparison with native speaker's pronunciation, oral structure analysis can be provided, and those who are new to English or who are not confident in English In addition, it is expected to provide convenient and effective English learning for the visually impaired and dyslexia patients as well as those who need more specialized English learning.

Claims (2)

When the user who registers through the initial screen of the web-only web browser combined with voice recognition and voice synthesis technology enters the member ID (20) and password (21), the user is automatically loaded with the final learning data by branching to the corresponding learning screen. An authentication screen step S1; When the sentence (37) is clicked, sound and animation is output on the screen, the content collection screen step (S2) that is automatically translated into the text; A word collection screen step (S3) of clicking on the word 38 and automatically translating the content into text while outputting sound and animation; If you select the sentence (37) and click the native speaker icon, the native speaker's voice waveform (28) is displayed with the sound and animation of the native speaker's pronunciation and mouth, and if the user pronounces the sentence (37) A representation exercise screen step S4 of which the voice waveform 28 of the screen is outputted and compared with the native speaker's voice waveform 28 and fed back while the corresponding score 29 is automatically calculated and outputted on the screen; If you select the word (38) and click the native speaker icon, the native speaker's voice waveform (28) is displayed on the screen along with the sound and animation of the native speaker's pronunciation and mouth, and if the user pronounces the sentence (37) And the voice waveforms 28 of the screen output and at the same time compared with the voice waveforms 28 of the native speaker, the corresponding score 29 is automatically calculated and displayed on the screen, the word practice screen step (S5); When the user clicks on the sentence 37 or the word 38 and the image and pronounces the corresponding answer, the self-evaluation screen step S6 is stored in the database while the calculated score 29 is displayed on the screen; A phonics screen step (S7) of clicking on each word 38 and outputting a video of a native speaker pronounced by the speaker; Internet English learning service method using speech recognition and speech synthesis, characterized in that the song / lyrics provided with a picture / song or lyrics game and the playground screen step (S8) provided with a picture making. The method of claim 1, wherein the expression training screen step (S4), as shown in the word training screen step (S5), the user's voice input is recognized and recorded by the built-in voice recognition engine 11, the corresponding voice waveform 28 and Internet English learning service method using the voice recognition and speech synthesis, characterized in that the output screen score (29).