KR19990075295A - Speech recognition language learning device - Google Patents

Abstract

The present invention relates to a speech recognition language learning apparatus, and since the conventional language learner is a device based on a player that has a video or audio tape attached, the rewinding operation of the part set by the user is repeated for repeated learning, and the output audio is output. In order to listen to and follow the sound, the built-in tape must be silenced for a certain period of time. To control the speed, if the playback speed is changed, the sound level of the playback audio is also changed, which is inconvenient for listening. There is a problem that it is difficult to expect a systematic learning effect because of the language learning.

Accordingly, the present invention stores the data according to various language learning in the CD-ROM and performs the repetitive learning and recording functions by processing the MPEG data of the video and audio data stored in the CD-ROM according to the user's selection. In order to improve the language listening ability by outputting, the user's voice is sampled and stored, and the user's voice is recognized through comparison with previously stored data. It was possible to judge whether the sentence was appropriate and correct pronunciation, so that the objective ability evaluation by self-learning was possible.

Description

Speech recognition language learning device

The present invention relates to a speech recognition language learning apparatus, and more particularly, to repeatedly play, change a playback speed, and record an image and audio data related to language learning stored in a CD-ROM according to a user's selection. After sampling, users can check the user's voice input in the form of answers or questions according to the situation, and improve the effect of language learning by objectively checking the self-learning through increasing sentence comprehension and correcting pronunciation. It is about a language learning apparatus.

In general, the language learner is mainly composed of the method of learning by playing audio and the method of playing video tape. The characteristics of language learning include the functions of repeated playing, answering, and speed control. Depending on the learning state, it is possible to expect a great effect of learning by intensively repeating the lacking part after storing the step.

However, since the language learner released to enhance the learning effect is a device based on a video or audio tape player, the user can rewind the tape of the part set by the user for repeat learning and listen to the audio output. In order to control the speed, the built-in tape must be silenced for a certain period of time. To control the speed, if the playback speed is changed, the sound level of the playback audio is changed, which makes it difficult to hear the sound. Therefore, it is not possible for the user to expect systematic learning effects due to the language learning, and it is impossible to check the objective ability when learning the language through the learner because the pronunciation is not verified correctly when reading aloud along the output audio. There was this.

Accordingly, the present invention has been made to solve the problem of the language learner to store the data according to the various language learning in the CD-ROM, and according to the user's choice, the MPEG-processed data and video data stored in the CD-ROM repetitive learning and recording In addition to performing the function as well as outputting the audio at a variable speed to improve the language listening ability, sampling the input user's voice to listen to the output audio and whether the pronunciation is appropriate for the sentence when the user pronounces and The purpose of the present invention is to provide a speech recognition language learning device that can double the effect on language learning by objectively checking the ability of self-learning by judging correct pronunciation.

Meanwhile, in order to achieve the above object, the present invention includes a key input means for setting each language learning mode with a plurality of key buttons, and an overall learning apparatus together with signal processing of data stored according to a signal input from the key input means. Main control means for the overall control of the signal, CD-ROM signal processing means for signal-processing and outputting the video and audio data stored in the CD according to the control signal of the main control means, and the control signal of the user by the control signal of the main control means Speech recognition and processing means for recognizing the audio signal and storing it in a corresponding memory map and digitally processing the input audio signal, and a signal for processing the MPEG signal of the video and audio signal decoded according to the control signal of the main control means. Characterized in that the processing means.

1 is a block diagram showing the configuration of a language training apparatus according to the present invention.

2 is a detailed block diagram of the main control means and voice recognition and processing means shown in FIG.

3 is a detailed block diagram of the CD-ROM signal processing means and the video and audio signal processing means shown in FIG.

4 is a view showing an example of a key input means applied to the present invention.

5 is a view showing an example of an initial screen according to a key operation applied to the present invention.

6 is a view showing an example of screen movement according to the direction key operation applied to the present invention.

7 shows an example of an initial screen of asking and answering in response to a key operation;

8 is a view showing an example of a screen according to each item of asking and answering applied to the present invention.

<Description of the code | symbol about the principal part of drawing>

10: key metrics 21: main controller

22: flash memory 23: graphics controller

31: microphone amplifier 32: voice recognition processing unit

33: voice memory section 34: digital voice processing section

41 CD mechanism 42 Digital signal processor

43 CD mechanism controller 44 CD ROM decoder

45: cdrom and mpack system controller

51: m-pack image processing unit 52: m-pack audio processing unit

53: variable speed controller 54: image selection processor

55 monitor 56 voice mixing processing unit

57: speaker

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

Language learning apparatus according to the present invention is provided with a plurality of buttons as shown in Figure 1 key input means for setting the learning mode according to the language learning, and the learning mode input from the key input means 10 Main control means 20 for overall control of the language learning operation according to the video and audio output in accordance with the signal, and digitalized image and audio data processing according to the control signal of the main control means 20 for signal processing Voice recognition and processing means 30 for recognizing the voice signal of the user according to the control signal of the CD-ROM signal processing means 40 and the main control means 20 and storing it in a corresponding memory map and digitally processing the input voice signal. And video and audio signal processing means 50 for processing MPEG signals of the video and audio signals decoded and input according to the control signal of the main control means 20.

Here, the main control means 20 is set by the main controller 21 responsible for the overall control of language learning according to the key signal input from the key input means 10, and the control signal of the main controller 21 The flash memory 22 temporarily stores data and performs a flash operation, and the graphic controller 23 controls output of pre-stored graphic data according to a control signal of the main controller 21.

The voice recognition and processing means 30 includes a voice recognition processor 32 for sampling a voice signal input from the microphone amplifier 31 and converting the voice signal into data in a corresponding memory map and storing the data, and the voice recognition processor 32. And a voice memory section 33 for storing the corresponding voice data which has been recognized and processed by the digital voice processor.

In addition, the CD-ROM signal processing means 40 includes a CD mechanism 41 for reading information in the CD, and a digital signal processor for digitizing and processing the audio and video data read by the CD mechanism 41. (42), a CD-ROM decoder 44 for decoding the audio and video signals digitally processed by the digital signal processor 42, and a CD-ROM decoder 44 connected to the CD-ROM operation and the MPEG image and MPEG audio. The CD-ROM and MPAC system controller 45 which controls the overall signal processing, and the CD mechanism controller 43 which specifically controls the operation of the CD mechanism 41 according to the control signal of the CD-ROM and MPAC system controller 45. Is done.

The video and audio signal processing means 50 outputs an MPEG image processor 51 for expanding the compressed signal output from the CD-ROM decoder 44 into an original video signal, and the CD-ROM decoder 44. The MPEG audio processing unit 52 for determining and processing the compressed audio signal to be the original audio signal, and the image signal output from the MPEG image processing unit 51 and the graphic controller 23 are stored on the CD-ROM and MPEG system controller 45. The image selection processing unit 54 selectively outputs the control signal of the MPAC audio processing unit 52, and outputs the signal output from the MPEG audio processing unit 52 at a set speed according to the control signal of the CD-ROM and MPEG system controller 45 or mixed signal processing. And a voice mixing processing section 56 for outputting.

On the other hand, the key input means 10 is composed of a plurality of buttons are connected to the main body of the learning apparatus by wire or wireless, for example, as shown in Figure 4 the key input means 10 "F1" "F2" "F3" and "F4" function selection buttons, "Speed", "Repeat", "Menu", "Record", "Stop", "Run", etc. characterized action button, direction or speed control or move to the next screen Four direction buttons are provided.

Here, the function selection button is defined as shown in Table 1 below.

button function F1 Display the sentence in Korean F2 Interpret new words or idioms in sentence content and explain grammar F3 Pronounce the sentence and interpret it in English and Korean F4 Choose texts, contextual examples, cultural etiquette, etc.

The execution button defines its function as shown in Table 2 below.

button function Speed button When this button is set, a graphic is displayed at the top of the monitor screen and the speed can be adjusted by using the arrow keys. Repeat button When this button is set, the "Repeat" character is displayed at the top of the monitor, and the output sentence is repeated. Press again to continue playing the next screen. Menu button Setting this button will return to the initial display menu. Record Button When this button is set, recording graphics are created at the top of the screen, creating a situation for recording with a microphone. Stop button Setting this button stops the current operating mode. Run Button Setting this button activates a menu or function.

On the other hand, the operation of the language learning device having such a configuration will be described in detail with reference to the accompanying drawings.

First, the user sets a predetermined button, for example a "menu" button, via the key matrix 10. Then, the initial screen as shown in FIG. 5 is displayed on the screen of the monitor 55. At this time, when the "run" button is pressed while the "daily life" is set on the screen, the sub screen as shown in FIG. The screen of 55 is displayed. At this time, each context menu displayed on this sub-screen can be moved through the direction key in the key matrix 10.

That is, when a predetermined button is set through the key matrix 10 as described above, a signal corresponding thereto is input to the main controller 21. Then, the main controller 21 applies the screen display data stored therein as the control signal to the image selection processing unit 54 through the graphic controller 23, so that the screen of the monitor 55 is shown in FIGS. The illustrated graphic is output.

Meanwhile, when the user presses the “Execute” button to set “greeting” through the key matrix 10 on the screen illustrated in FIG. 6, the main controller 21 sends a predetermined control signal to the CD-ROM / MPac system controller 45. Is applied. Then, the CD-ROM / MPAC system controller 45 applies a predetermined control signal to the CD mechanism controller 42, so that the video data and audio data stored in the CD, which is an optical recording medium, in accordance with the control signal of the CD mechanism controller 42, Detected by the drive of the CD mechanism 41, the detected image data and audio data are digitally signaled and input to the CD-ROM decoder 44 via the digital signal processor 42. Therefore, the CD-ROM decoder 44 decodes the input data and transmits the data to the MPEG image processor 51 and the MPEG audio processor 52 according to the control signal of the CD-ROM / MPAC system controller 45. Is authorized. Therefore, the MPEG image processing unit 51 extends the input image data into an original signal, and the processed image signal is applied to the monitor 55 via the image selection processing unit 54. At the same time, the MPEG voice processing unit 52 extends the input voice data into an original signal and applies it to the voice mixing processing unit 56. In this case, the voice mixing processor 56 outputs the voice through the speaker 57 according to the speed set by the variable speed controller 53.

Therefore, the user can listen to the voice according to the language along with the video screen corresponding to the greeting situation, wherein the video screen output to the monitor 55 is displayed according to the user selects a function selection button. At this time, while the user is listening to the audio and video by the language data currently output from the CD-ROM and set the "speed" button of the key metrics 10, and then set to move in the vertical direction through the "direction button". The corresponding signal is then applied to the main controller 21. This main controller 21 applies a predetermined control signal to the CD-ROM / MPac system controller 45. Then, according to the control signal of the CD-ROM system controller 45, the variable speed controller 53 inputs a constant variable signal to the voice synthesis mixing unit 56. Then, the voice synthesis mixer 56 is currently signal-processed through the MPEG voice processor 52, and the input voice signal is adjusted by the variable signal. At the same time, the video signal processed by the MPEG image processing unit 54 is also adjusted and output to correspond to the audio output speed.

On the other hand, when the "Repeat" button in the key matrix 10 is pressed once during the output of the audio and video as described above, the "Repeat" character is displayed on the screen of the monitor 55 and the currently executed sentence is repeated. Press this button again to cancel the repeat mode. When the menu button is pressed during the output of the audio and video, the controller returns to the initial screen shown in FIG. 5 according to the control signal of the main controller 21.

In the initial screen shown in FIG. 5 as described above, when the user moves to "Ask and Answer" through the direction buttons and sets the "Execute" button, a signal corresponding thereto is applied to the main controller 21 and a predetermined signal according to the graphic is displayed. Is applied to the controller 23. The graphic controller 23 then applies the pre-stored screen information to the image selection processor 54 as the control signal. Therefore, the image selection processing section 54 processes the input screen information by the selection signal processing and outputs it through the monitor 55 so that the screen as shown in FIG. 7 is displayed on the monitor 55. On the screen, "practice", "recording", "answer", "question" and "evaluation" items are displayed. In this case, the user must set the "Recording" item first because the user needs to store the standard sampling data for voice recognition before executing the "answer" or "ask".

Accordingly, the operation for storing the voice standard sampling data of the user is as follows. First, when the "record" button of the key matrix 10 is pressed while "recording" is set on the screen, the predetermined signal is input to the main controller 21. At this time, the main controller 21 applies a predetermined control signal to the CD-ROM / MPac controller 45 according to the input signal. Therefore, the CD-ROM / MPac system controller 45 applies a predetermined control signal to the CD mechanism controller 43 so that the CD mechanism 41 is driven to read the data stored in the CD, and the read data is a digital signal processor. Digital signal processing at 42 is applied to the CD-ROM decoder 44. Therefore, the CD-ROM decoder 44 decodes the input data and separates the data signal into video data and audio data according to the control signal of the CD-ROM / MPac controller 45. Input to processing unit 52, respectively. Therefore, the MPEG image processing unit 51 processes the input signal into the original image and applies it to the image selection processing unit 54 so that the screen shown in FIG. 8A is displayed on the monitor 55. At the same time, the MPEG audio processing unit 52 extends the input audio signal into an original signal and applies it to the voice mixing processing unit 56. At this time, the voice mixing processing unit 56 processes the input voice signal, and finally a predetermined voice message is output through the speaker 57.

That is, for example, the voice message "This is a menu for the computer to generate standard sampling data for the user's voice recognition before entering the ask and answer menu" is output. In addition, the CD-ROM signal processing means 40 outputs data corresponding to a pre-stored sentence according to a control signal of the main controller 21 over a predetermined time. At this time, since the possible time of voice recording is displayed according to the output sentence on the screen of the monitor 55, the user inputs the voice through the microphone not shown during the period. The user's voice is amplified by the microphone amplifier 31 and the amplified voice signal is applied to the voice recognition processor 32. At this time, the voice signal applied to the voice recognition processor 32 is converted into data at a set sampling frequency and the data is stored in the voice memory unit 33 having a predetermined map pattern. If it matches the signal according to the user's voice is stored in the voice memory unit 33, but if it does not match the original pronunciation of the output sentence, the corresponding sentence is output again from the CD-ROM signal processing means 40 again Can give recording opportunities.

On the other hand, if the user's voice corresponding to the pronunciation of the sentence output as described above to store a certain amount of data in the voice memory unit 33, the mode corresponding to the "recording" is terminated. When the "Practice" item is set in the screen output shown in FIG. 7, the data read from the CD-ROM signal processing means 40 is signal-processed by the signal processing means 50 according to the control signal of the main controller unit 21. The speaker 57 and the monitor 55 are displayed on the screen of the voice message and the video, and the output voice message is "a sufficient mastery of the sentence and pronunciation practice before entering the Ask and Answer menu." After that, one sentence is outputted with a subtitle and the voice of the corresponding sentence is output to the speaker so that the user can perform the pronunciation practice.

That is, in this operation, when the user's sampled voice data is stored in the voice memory unit 33, the " answer " is set in the menu shown in FIG. Press the "Execute" button is the corresponding signal is input to the main control means (20). Then, the main control means 20 applies a predetermined control signal to the CD-ROM signal processing means 40, so that the predetermined video data and audio data output from the CD-ROM signal processing means 40, as described above, are converted into a video and an image. The audio and video screens are signal-processed into audio and video signals through the audio signal processing unit 50 and output to the speaker 57 and the monitor 55. That is, the speaker 57 outputs a voice message "Please answer questions". Thereafter, the question sentence stored in the CD-ROM signal processing means 40 is output through the speaker 57, and at the same time, as shown in FIG. 8B, the question sentence is also displayed and a screen to which the sentence to be answered is flashed. Therefore, when the user hears the voice output and the text display of the sentence and inputs the voice corresponding to the answer within a predetermined time through the microphone, the voice signal is amplified through the microphone amplifier 31 and then input into the voice recognition processor 32. do. Then, the voice recognition processor 32 analyzes the corresponding voice data pre-stored in the voice memory unit 33 and the input voice signal according to the control signal of the main controller 21. At this time, if the input voice signal matches the corresponding voice data, the correct answer is displayed on the screen or a constant sound is output through the speaker.

On the other hand, such a "answer" mode is usually in conjunction with the "inquiry" mode, when the user sets the "inquiry" item, unlike the "answer" mode, the speaker 57 first outputs the voice corresponding to the answer At the same time, a sentence corresponding to the answer is displayed on the screen. In this case, by outputting the voice corresponding to the appropriate sentence, the user can determine whether the voice input of the user is correct through input / output of data by the above-described operation.

As described above, the present invention includes a language data embedded in the CD-ROM, a device for outputting the same, and a means for recognizing a user's voice. By repeating the learning, you can select the speed control function, recording function, etc. according to the user's key operation can improve the listening ability according to the speed completion and can expect the effect of pronunciation correction.

Claims (6)

Key input means 10 for setting a mode according to language learning and a plurality of buttons, and main for controlling the overall language learning operation according to the video and audio output according to the signal input from the key input means 10 Control means 20, CD-ROM signal processing means 40 for digitizing and processing the video and audio data of the video stored according to the control signal of the main control means 20, and control of the main control means 20 Voice recognition and processing means 30 for recognizing the voice signal of the user according to the signal and storing it in a corresponding memory map and digitally processing the input voice signal, and decoded and input according to the control signal of the main control means 20. Speech recognition language learning device, characterized in that consisting of video and audio signal processing means for processing the video and audio signal MPEG. 2. The main controller (20) according to claim 1, wherein the main control means (20) is a main controller (21) which is responsible for the overall control of language learning according to the key signal input from the key input means (10), and the control of the main controller (21). And a flash memory 22 which temporarily stores data set by a signal and performs flash operation, and a graphics controller 23 which controls output of pre-stored graphic data according to a control signal of the main controller 21. Cognitive Language Learning Device. 2. The speech recognition and processing unit (30) according to claim 1, wherein the speech recognition and processing means (30) includes a speech recognition processing unit (32) for sampling a speech signal input from the microphone amplifier 31, converting the speech signal into data in a corresponding memory map, and storing the data. A voice memory unit 33 storing corresponding voice data recognized by the recognition processor 32, and a digital voice processor 34 for signal processing the PCM coded voice signal output from the microphone amplifier 31. Speech recognition language learning device, characterized in that made. 2. The CD-ROM signal processing means (40) according to claim 1, wherein the CD-ROM signal processing means (40) digitizes the CD mechanism 41, which operates to read information in the CD, and the audio and video data read by the CD mechanism 41. A digital signal processor 42, a CD-ROM decoder 44 which decodes audio and video signals digitally processed by the digital signal processor 42, and a CD-ROM decoder 44, which is connected to the CD-ROM decoder 44 to operate and And a CD mechanism controller for controlling the operation of the CD mechanism 41 in accordance with the control signal of the CD ROM and the MPAC system controller 45 and the CD ROM and MPAC system controller 45. Speech recognition language learning device, characterized in that consisting of (43). The image and audio signal processing unit (50) of claim 1, further comprising: an MPEG image processing unit (51) for expanding the compressed signal output from the CD-ROM decoder (44) into an original video signal, and the CD-ROM decoder. The MPEG audio processing unit 52 for determining and processing the compressed audio signal output from the 44 as the original audio signal, and the image signals output from the MPEG image processing unit 51 and the graphic controller 23 are stored on the CD-ROM and MPEG. The image selection processor 54 selectively outputs the control signal of the system controller 45 and the signal output from the MPEG audio processor 52 at a speed set according to the control signal of the CD-ROM and MPEG system controller 45. Speech recognition language learning device, characterized in that composed of a speech mixing processing unit 56 for outputting or outputting a mixed signal processing. The method according to claim 1, wherein the key input means (10) is composed of a plurality of buttons connected to the main body of the learning apparatus in a wired or wireless manner, and the buttons select the functions of "F1", "F2", "F3" and "F4". Buttons, "Speed", "Repeat", "Menu", "Record", "Stop", "Run", etc., and the action buttons, and four-way direction button to move to the next screen or adjust the speed Speech recognition language learning device, characterized in that.