JPH06348297A - Pronunciation trainer - Google Patents

Abstract

PURPOSE:To provide a pronunciation trainer in which a trainee observes the shape of his mouth and the position of his tongue while he is pronunciating a word and compares the observed shape and position against the correct shape of a mouth and the correct position of a tongue. CONSTITUTION:The trainer is provided with a pronunciation data storage section 17 which beforehand stores the data relative to the shape of a mouth and the position of a tongue corresponding to the phonemes included in voices, a voice recognition section 12 which detects phonemes from inputted voices, a mouth shape predicting section 13 which predicts the shape of the mouth and the position of the tongue based on the detected phonemes, a prediction result holding section 14 which temporarily holds the prediction results and a mouth shape display section 15 which displays the predicted shape of the mouth and the position of the tongue and the correct versions of the above.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pronunciation practice device used for pronunciation practice in English conversation and the like.

[0002]

2. Description of the Related Art Conventionally, as a pronunciation practice device, for example,
There is a pronunciation practice system called "Speak English" (Digitec Co., Ltd.).

As shown in FIG. 5, this pronunciation practice system is a system for learning practical English conversation in a game format. While actually repeating dialogues with a system using a computer, the learner's utterance is referred to as a native speaker. By comparison, the pronunciation of the learner is scored and displayed.

While listening to the correct (sample) pronunciation by voice, at the same time, the movement of the cross-section of the mouth corresponding to the correct pronunciation is displayed on the display device to see how to move the mouth or tongue. You can

[0005]

However, in the case where the correct pronunciation cannot be achieved by the method described above,
Since the learner himself cannot see the shape of the mouth and the position of the tongue when pronouncing, it is not clear how to correct the shape of the mouth and the position of the tongue in order to pronounce correctly.

That is, there is a problem that it is difficult to objectively judge how the shape of the mouth and the position of the tongue are different from the correct shape of the mouth and the position of the tongue.

In consideration of such problems of the conventional pronunciation practice, the present invention enables the learner to see the mouth shape and tongue position pronounced by the learner, and to compare with the correct mouth shape and tongue position. The purpose is to provide a pronunciation training device that can

[0008]

SUMMARY OF THE INVENTION According to the present invention, storage means for storing in advance pronunciation generation information for pronouncing mouth shape, tongue position, etc. corresponding to phonological information included in speech, and phonological elements from input speech. A pronunciation practice device including a phoneme information detecting means for detecting information, and a reading display means for reading the stored pronunciation generation information based on the detected phoneme information and displaying the read pronunciation generation information. is there.

[0009]

In the present invention, the storage means corresponds to the phoneme information,
The pronunciation generation information for pronouncing the mouth shape, tongue position, etc. is stored in advance, the phonological unit information detecting unit detects the phonological unit information from the input voice, and the reading display unit is stored based on the detected phonological unit information. The pronunciation generation information is read out and the pronunciation generation information is displayed.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings showing its embodiments.

FIG. 1 is a diagram of a computer system constituting a pronunciation practice apparatus according to an embodiment of the present invention. That is, the computer system constituting the pronunciation practice device includes a central processing unit 1 for performing information processing of the entire pronunciation practice device, a storage device 2 for storing information processing programs and data, and a program for pronunciation practice. , An external storage device 3 in which data and the like are stored, a keyboard 4 for inputting data and commands, a display device 5 for displaying processing results, and voice recognition of voice input from a microphone 7. The voice recognition device 6 for Further, the program stored in the external storage device 3 is loaded into the storage device 2, and the central processing unit 1 executes the pronunciation practice process in accordance with the program.

FIG. 2 is a block diagram showing the functions of the pronunciation practice apparatus of the embodiment. That is, the pronunciation practice apparatus displays the practice data storage unit 16 in which the data of the pronunciation practice questions (sentences, words, etc.) are stored, and the data stored in the practice data storage unit 16 is displayed as a practice question for the learner. A problem display unit 11, a voice recognition unit 12 for inputting and recognizing a voice of a learner who is pronounced by looking at the displayed exercise and recognizing a phoneme contained in the voice, based on the detected phoneme, The mouth shape estimator 13 that estimates the mouth shape / tongue position at that time, and the mouth shape / tongue position corresponding to the phoneme as shown in FIG. A pronunciation data storage unit 17 for storing data (FIG. 3 shows an example, and there are many data other than those shown in the figure), and an estimation result holding unit for temporarily holding the estimation result by the mouth shape estimation unit 13. Part 14, guess result It is constituted by mouth shape display unit 15 to be displayed on the CRT screen or the like.

The estimation result holding unit 14 is provided in the storage device 2 of FIG. 1, and the practice data storage unit 16 and the pronunciation data storage unit 17 are provided in the external storage device 3 of FIG. Further, the pronunciation data storage unit 17 constitutes a storage unit, the voice recognition unit 12 constitutes a phoneme information detection unit, and the mouth shape estimation unit 13 and the mouth shape display unit 15 constitute a read display unit.
The estimation result holding unit 14 constitutes a holding means.

Next, the operation of the pronunciation practice apparatus of the above embodiment will be described.

First, the pronunciation training program stored in the external storage device 3 is loaded into the storage device 2, and the program is started.
Exercises (sentences, words, etc.) stored in
“Right” is presented to the learner by the display device 5.
The learner inputs a voice from the microphone 7 for the exercise. In this case, the pronunciation “right” of the exercise is pronounced. The voice uttered by the learner is converted into a phoneme code by the voice recognition unit 12 and output to the mouth shape estimation unit 13. Based on this phoneme code, the mouth shape estimation unit 13 refers to the phoneme stored in the pronunciation data storage unit 17 and the relationship data of the mouth shape and the position of the tongue, and the mouth shape / inferred from the input phoneme / Determine the tongue position. The shape of the mouth and the position of the tongue at this time are shown in FIG. Further, FIG. 4A shows the correct mouth shape and tongue position, which are the reference pronunciation generation information of the presented exercise “right”.

Next, the estimated mouth shape and tongue position are temporarily stored in the estimation result holding unit 14, and the estimation result is displayed as a moving image in real time on the display device 5. That is, the screen is sequentially switched from the top to the bottom of FIG. In this example, the “r” of “right” uttered by the learner is estimated as the pronunciation of “l”.

In addition, in the above display, if there is an incorrect mouth shape / tongue position in phonological units compared to the correct one, "open mouth more sideways", "tongue up" Instructions such as "Do not attach to the chin" and the correct mouth shape and tongue position and the estimated mouth shape and tongue position for that part are displayed, so that the learner can compare errors.
Here, the above-mentioned instruction may be performed by voice using voice synthesis.

Further, by using the estimation result stored in the estimation result storage unit 14, it is possible to view the estimated mouth shape repeatedly or to display it at a slower display speed after displaying it in real time.

Further, the correct (ideal) mouth shape and tongue position as shown in FIG. 4A and the estimated mouth shape and tongue position as shown in FIG. 4B are included. From the same time, to make it easier for the learner to compare, the wrong mouth shape and tongue position can be displayed by changing the color and displaying them in a superimposed manner for comparison. It is also possible.

In the above embodiment, the shape of the mouth and the position of the tongue have been described as an example of the pronunciation generation information. However, the present invention is not limited to this, and may include other elements related to pronunciation such as breathing in and out. Of course good.

Further, in the above embodiment, the word is used as an example of the problem data which is the predetermined symbol, but the present invention is not limited to this.
For example, it may be a sentence, a phonetic symbol, or the like.

Further, in the above-mentioned embodiment, the estimation result holding unit 14 which is the holding means is configured to store the estimation result of the mouth shape and the tongue position which is the pronunciation generation information, but instead of this,
You may make it memorize | store the phoneme information of the input speech. Alternatively, both the phoneme information and the estimation result may be stored.

Further, in the above embodiment, the display of the estimation result and the like is displayed as a moving image in real time. However, the present invention is not limited to this. For example, the display speed may be slowed down as described above, or a still image may be displayed. You may do it.

In the above embodiment, the pronunciation of English has been described as an example, but it goes without saying that the type of language is not limited to this.

Further, in the above-mentioned embodiment, the computer system is used as software, but instead of this, dedicated hardware having the same function may be used.

[0026]

As is apparent from the above description, the present invention provides phonological information detecting means for detecting phonological information from input speech, and stored pronunciation generation information based on the detected phonological information. Since it has read-out display means for reading out and displaying the read out pronunciation generation information, the learner can see the shape of the mouth and the position of the tongue, and the correct shape of the mouth and the position of the tongue. It has the advantage of being able to compare

[Brief description of drawings]

FIG. 1 is a diagram of a computer system that constitutes a pronunciation practice apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing functions of the embodiment.

FIG. 3 is a diagram showing relational data between phoneme and mouth shape / tongue position in the embodiment.

FIG. 4 (a) is a diagram showing a display example of a correct mouth shape and tongue position in practice according to the embodiment, and FIG. 4 (b).
FIG. 8 is a diagram showing a display example of the mouth shape and tongue position estimated from the utterance at that time.

FIG. 5 is a diagram showing an example of a conventional pronunciation practice apparatus.

[Explanation of symbols]

 1 Central Processing Unit 2 Storage Device 5 Display Device 6 Speech Recognition Device 7 Microphone 12 Speech Recognition Unit 13 Mouth Guessing Unit 14 Guessing Result Holding Unit 15 Mouth Shape Displaying Unit 17 Pronunciation Data Storage Unit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Reference number within the agency FI technical display location G10L 9/14 301 Z 9379-5H (72) Inventor Tadada Tagawa Toranomon 1-chome, Minato-ku, Tokyo No. 12 Oki Electric Industry Co., Ltd. (72) Inventor Yukari Many 3-8-10 Uchiyama, Chikusa-ku, Nagoya, Aichi Oki Techno Systems Laboratory Co., Ltd.

Claims (5)

[Claims] 1. Storage means for storing in advance pronunciation generation information for pronouncing mouth shape, tongue position, etc. corresponding to phonological information included in speech, and phonological information detection for detecting the phonological information from input speech. A pronunciation practice apparatus comprising: a means and a read-out display means for reading the stored pronunciation generation information based on the detected phoneme information and displaying the read pronunciation generation information. 2. The input voice is generated by a speaker for a predetermined symbol for preset sounding, and reference pronunciation generation information of the predetermined symbol is displayed on the read display means. The pronunciation practice device according to claim 1, wherein 3. The pronunciation practice device according to claim 2, wherein the read-out display means performs a comparative display of the displayed pronunciation generation information and the reference pronunciation generation information. 4. The holding means for holding the phoneme information of the input voice and / or the read pronunciation generation information corresponding to the phoneme information, is provided. The pronunciation practice device described in 3. 5. The pronunciation practice apparatus according to claim 1, wherein the pronunciation generation information is displayed as a moving image.