CN101727764A - Method and device for assisting in correcting pronunciation - Google Patents
Method and device for assisting in correcting pronunciation Download PDFInfo
- Publication number
- CN101727764A CN101727764A CN200810170987A CN200810170987A CN101727764A CN 101727764 A CN101727764 A CN 101727764A CN 200810170987 A CN200810170987 A CN 200810170987A CN 200810170987 A CN200810170987 A CN 200810170987A CN 101727764 A CN101727764 A CN 101727764A
- Authority
- CN
- China
- Prior art keywords
- pronunciation
- assisting
- correcting
- phoneme
- mistake
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention relates to a method and a device for assisting in correcting pronunciation; wherein the method includes the following steps: providing example sentences which comprises multiple reference phonemes; providing multiple mistake rules, wherein each mistake is corresponding to a reference phoneme; receiving user pronunciation; identifying multiple phonemes are carried in the user pronunciation; comparing the multiple pronunciation phonemes with the multiple mistake rules; recording the mistake times when the multiple pronunciation phonemes are in line with the multiple mistake rules.The function of the invention lies in that the learners can repeatedly produce user pronunciation according to the example sentences, then the user pronunciation is identified to be multiple pronunciation phonemes which are compared with the multiple mistake rules; the mistake times when the multiple pronunciation phonemes are in line with the multiple mistake rules are recorded, the learners and the teachers can rapidly realize the mistakes in each pronunciation according to the analyzed result from the comparison and obtain appropriate subsequent assistance and correction according to each different pronunciation mistake.
Description
Technical field
The present invention relates to a kind of method that corrects pronunciation mistakes and device, particularly a kind of method and the device that can find the assisting in correcting pronunciation of user's articulation problems rapidly.
Background technology
In the process of language learning, people usually need by various language tool book, assist it to learn a language, as e-dictionary, personal computer (Personal Computer) and personal digital assistant (Personal Digital Assist, PDA) etc., and the function of carrying out language learning via the mode that software is carried out, its sound pronunciation is that sentence or individual character are stored in the computer system in the mode of pre-recording, and plays the relief user and hears correct pronunciation.The function of recording perhaps is provided simultaneously, and user's operational computations machine system makes its operation of recording, and is available for users to listen to the sound of oneself, and operator scheme just can reach the effect of carrying a tune property of rectification oneself and language learning whereby.
As No. 200713200 patent of TaiWan, China disclosure of the Invention " the interactive apparatus and method that correct pronunciation mistakes ", utilization is analyzed with reference to sound and user's pronunciation, and generation voice parameter, become correct pronunciation according to the reference sound and with user's vocal modifications again, learn correct pronunciation so as to auxiliary user; No. 556152 patent of TaiWan, China invention bulletin " indicates user interface, the system and method for phonetic symbol to correct pronunciation mistakes " automatically, be in the user sentence word string is provided and import its relative voice signal after, each section at input audio signal indicates its corresponding phonetic symbol, continuous diaphone with section is marked with the difference of comparing out between instructor's voice signal and the learner's voice signal, as the difference of the pronunciation of each syllable, pitch, intensity, length etc., and mark in view of the above or propose to improve suggestion.
Technique scheme allows the user can't really understand the situation of mispronounce, only can be according to the correctness of individual subjective judgement language pronouncing intonation.Therefore, how to allow the instructor understand each learner's mispronounce, or allow the learner understand own enunciative mistake, and allow the instructor can be according to the result after analyzing, the learner is provided follow-up auxiliary and rectification, is an instant problem to be solved.
Summary of the invention
In view of this, for overcoming the defective of prior art, the present invention proposes a kind of method of assisting in correcting pronunciation, provides example sentence and a plurality of wrong rule earlier, wherein, the example sentence comprises a plurality of with reference to phoneme, and after this corresponding one of a wrong rule receives user's pronunciation again with reference to phoneme, and identification user pronunciation is a plurality of pronunciation phonemes, the a plurality of pronunciation phonemes of comparison and a plurality of wrong rule after identification is finished, and write down the errors number that a plurality of pronunciation phonemes meet a plurality of wrong rules.
The present invention also proposes a kind of device of assisting in correcting pronunciation, comprises: playing module, receiver module, recognition module, rule module, comparing module.Playing module comprises a plurality of example sentences with reference to phoneme in order to broadcast.Receiver module is a plurality of pronunciation phonemes via recognition module identification user pronunciation after being received in user's pronunciation that the user sent according to the example sentence.Rule module provides and the corresponding a plurality of wrong rules of example sentence, wherein, wrong rule corresponding one with reference to phoneme.Comparing module is compared a plurality of pronunciation phonemes and a plurality of wrong rule.
Effect of the present invention is, the learner can send user's pronunciation repeatedly according to the example sentence, promptly can identification user pronunciation be a plurality of pronunciation phonemes, and a plurality of pronunciation phonemes and a plurality of wrong rule are compared, and write down the errors number that a plurality of pronunciation phonemes meet a plurality of wrong rules, the learner can understand oneself by errors number and pronounce to go up incidental mistake, the instructor also can be according to the result after analyzing, grasp the mistake of each learner's pronunciation fast, and provide suitable follow-up auxiliary reaching to correct according to the different mispronounce of each learner.
Below in embodiment, be described in detail detailed features of the present invention and advantage, its content is enough to make those of ordinary skill in the art to understand technology contents of the present invention and implements according to this, and according to the disclosed content of this instructions, claim and accompanying drawing, those of ordinary skill in the art can understand purpose and the advantage that the present invention is correlated with easily.
Description of drawings
Fig. 1 is the structure calcspar of the device of assisting in correcting pronunciation of the present invention;
Fig. 2 is the process flow diagram of the method for assisting in correcting pronunciation of the present invention.
Description of reference numerals in the above-mentioned accompanying drawing is as follows:
10 load modules
20 playing modules
30 receiver modules
40 recognition module
50 rule module
60 comparing module
70 logging modles
80 output modules
90 dictionaries
Embodiment
As shown in Figure 1, for showing the device of assisting in correcting pronunciation of the present invention.
The device of assisting in correcting pronunciation of the present invention comprises: playing module 20, receiver module 30, recognition module 40, rule module 50, comparing module 60.
Whether correct receiver module 30 receives user's pronunciation that the user sent in the user sends user's pronunciation according to the example sentence of playing module 20 after, to judge user's benchmark that pronounces as follow-up.
The a plurality of wrong rule that a plurality of pronunciation phonemes of comparing module 60 comparison recognition module 40 identifications and rule module 50 provide, at this, comparing module 60 is compared each pronunciation phonemes in regular turn with corresponding wrong regular wrong phoneme.
The example sentence of above stated specification can be system and preestablishes, or can be imported in advance by the instructor, and wherein, the instructor can import the example sentences via load module 10, and it can be according to course arrangement or the different example sentence of learner's degree input when input.
The device of assisting in correcting pronunciation of the present invention more comprises: logging modle 70, output module 80.Logging modle 70 can write down the errors number that a plurality of pronunciation phonemes meet a plurality of wrong rules, and can grasp each pronunciation situation for instructor or learner at any time via output module 80 output error number of times after comparing module 60 each comparisons be finished.
The instructor can be earlier with load module 10 input example sentences " I am a boy. " during use, and the learner plays the example sentence with playing module 20 again, and sends user's pronunciation repeatedly and the exercise pronunciation according to the example sentence of playing module 20.The learner sends the user at every turn and pronounces all via receiver module 30 receptions, be a plurality of pronunciation phonemes with recognition module 40 identification users pronunciation again, and after comparing module 60 comparison is finished with logging modle 70 misregistration number of times, as user's pronunciation is a0 $m D bC0 totally 30 times, user's pronunciation is a0 Gm D bC0 totally 25 times, user's pronunciation is a0 Am D bC0 totally 10 times, then can be with its outcome record:
With reference to phoneme ($) mistake phoneme (G) mistake phoneme (A) total degree accuracy
30 times 25 times 15 times 70 times 43%
Behind output module 80 output error number of times, the learner can understand oneself by errors number and pronounce to go up incidental mistake, the instructor also can be according to the result after analyzing, grasp the mistake of each learner's pronunciation fast, easily " $ " pronounced " G " as a certain learner, can provide suitable follow-up auxiliary and rectification according to the incidental mispronounce of this learner, assist it that " G " corrected pronunciation mistakes and be " $ ".
As shown in Figure 2, the method for assisting in correcting pronunciation of the present invention comprises the following step:
Step 201: provide to comprise a plurality of example sentences with reference to phoneme.
The instructor can import the example sentences via load module 10, and it can be according to course arrangement or the different example sentence of learner's degree input when input.In addition, the example sentence can be made up of a plurality of individual character, each individual character has at least one with reference to phoneme, therefore the example sentence can be expanded into a plurality of with reference to phoneme, as the example sentence is " I am a boy. ", formed by four individual characters such as " I ", " am ", " a ", " boy ", and it is a0 $m D bC0 with reference to phoneme.
Launch in the step of example sentence, can launch the example sentences according to dictionary 90, at this, dictionary 90 be general on the market English English, dictionary, dictionary such as English-Chinese, Chinese-English, and with its data storage in storage unit, but the present invention is not as limit.
Step 202: a plurality of wrong rules are provided, wrong rule corresponding one with reference to phoneme.
Provide and the corresponding a plurality of wrong rules of example sentence with rule module 50, wrong rule corresponding one with reference to phoneme, and each wrong rule has at least one wrong phoneme, comprises wrong phonemes such as " G ", " A ", " D " as the wrong rule of corresponding " $ ".
Step 203: receive user's pronunciation.
Play the example sentence via playing module 20 in modes such as sound or images, the learner sends user's pronunciation and exercise pronunciation repeatedly according to the example sentence again, and receives the user that sent of the each institute of learners with receiver module 30 and pronounce.
Step 204: identification user pronunciation is a plurality of pronunciation phonemes.
With recognition module 40 identification users pronunciation is a plurality of pronunciation phonemes, wherein, recognition module 40 directly acoustic ratio to or acoustic model comparison carry out identification, and acoustic model preferably can be concealed markov model (Hidden Markov Model, HMM), but the invention is not restricted to this, acoustic model also can be the calibration of neural network (Neural Networks), dynamic time (Dynamic Time Warping, DTW) or sound template comparison (Template Matching).
Step 205: compare a plurality of pronunciation phonemes and a plurality of wrong rule.
In this step, comparing module 60 is compared each pronunciation phonemes in regular turn with corresponding wrong regular wrong phoneme.
Step 206: write down the errors number that a plurality of pronunciation phonemes meet a plurality of wrong rules.
Step 207: output error number of times.
After comparing module 60 comparison is finished with logging modle 70 misregistration number of times, and can be via output module 80 output error number of times, pronounce to go up incidental mistake so as to allowing the learner understand oneself, and can grasp the mistake that each learner pronounces fast, and provide suitable follow-up auxiliary and rectification according to the incidental mispronounce of each learner for the instructor.
The present invention can be applicable to e-dictionary, personal computer (Personal Computer) and personal digital assistant (Personal Digital Assist, electronic installation such as PDA), on the low cost computer that particularly can be applicable to day by day popularize, so as to allowing the learner understand own enunciative mistake, simultaneously can allow the instructor be directly acquainted with each learner's mispronounce, and the pronunciation that need not expend a large amount of each learner of time series analysis, in addition, the instructor can be according to the result after analyzing, the learner is provided follow-up auxiliary and rectification, not only allow the learner learn to have more efficient, allow the instructor directly provide rectification simultaneously, effectively save the time of teaching pronunciation at comparatively serious mispronounce.
Though technology contents of the present invention with preferred embodiment openly as above; right its is not in order to limit the present invention; those of ordinary skill in the art is not breaking away from spirit of the present invention a little improvement of doing; all should be covered by in the category of the present invention, so protection scope of the present invention is as the criterion when looking appended the scope that claim defined.
Claims (19)
1. the method for an assisting in correcting pronunciation comprises the following step:
One example sentence is provided, and this example sentence comprises a plurality of with reference to phoneme;
A plurality of wrong rules are provided, and one should corresponding one of mistake rule be somebody's turn to do with reference to phoneme;
Receive user pronunciation;
This user's pronunciation of identification is a plurality of pronunciation phonemes;
Compare described a plurality of pronunciation phonemes and described a plurality of wrong rule; And
Write down the errors number that described a plurality of pronunciation phonemes meets described a plurality of wrong rules.
2. the method for assisting in correcting pronunciation as claimed in claim 1, in the step that this example sentence is provided, comprise: it is described a plurality of with reference to phoneme launching this example sentence.
3. the method for assisting in correcting pronunciation as claimed in claim 2 in the step of launching this example sentence, is launched this example sentence according to a dictionary.
4. the method for assisting in correcting pronunciation as claimed in claim 1, wherein this example sentence comprises at least one individual character, and this individual character should be made up of with reference to phoneme at least one.
5. the method for assisting in correcting pronunciation as claimed in claim 1 wherein should the mistake rule comprise at least one wrong phoneme.
6. the method for assisting in correcting pronunciation as claimed in claim 5 in the described a plurality of pronunciation phonemes of comparison and described a plurality of wrong regular steps, is compared each this pronunciation phonemes in regular turn with this mistake phoneme of corresponding this mistake rule.
7. the method for assisting in correcting pronunciation as claimed in claim 1 in the step of this user's pronunciation of identification, is carried out identification with a direct acoustic ratio to one of them of mode and acoustic model ratio.
8. the method for assisting in correcting pronunciation as claimed in claim 7, wherein this acoustic model is selected from the group that a concealed markov model, a neural network, dynamic time calibration and sound template comparison are constituted.
9. the method for assisting in correcting pronunciation as claimed in claim 1 after the step of this errors number of record, also comprises: export this errors number.
10. the device of an assisting in correcting pronunciation comprises:
One playing module, in order to play an example sentence, this example sentence comprises a plurality of with reference to phoneme;
One receiver module is in order to be received in the user pronunciation that a user is sent according to this example sentence;
One recognition module is a plurality of pronunciation phonemes in order to this user's pronunciation of identification;
One rule module, in order to provide and the corresponding a plurality of wrong rules of this example sentence, one should corresponding one of mistake rule be somebody's turn to do with reference to phoneme; And
One comparing module is in order to compare described a plurality of pronunciation phonemes and described a plurality of wrong rule.
11. the device of assisting in correcting pronunciation as claimed in claim 10, wherein this example sentence comprises at least one individual character, and this individual character should be made up of with reference to phoneme at least one.
12. the device of assisting in correcting pronunciation as claimed in claim 11 also comprises: a dictionary, in order to store this individual character and corresponding this with reference to phoneme.
13. the device of assisting in correcting pronunciation as claimed in claim 10, wherein this recognition module is carried out identification with a direct acoustic ratio to one of them of mode and acoustic model comparison.
14. the device of assisting in correcting pronunciation as claimed in claim 13, wherein this acoustic model is selected from the group that a concealed markov model, a neural network, dynamic time calibration and sound template comparison are constituted.
15. the device of assisting in correcting pronunciation as claimed in claim 10 wherein should the mistake rule comprise at least one wrong phoneme.
16. the device of assisting in correcting pronunciation as claimed in claim 15, wherein this comparing module is compared each this pronunciation phonemes in regular turn with this mistake phoneme of corresponding this mistake rule.
17. the device of assisting in correcting pronunciation as claimed in claim 10 also comprises: a load module, play for this playing module in order to import this example sentence.
18. the device of assisting in correcting pronunciation as claimed in claim 10 also comprises: a logging modle, in order to write down the errors number that described a plurality of pronunciation phonemes meets described a plurality of wrong rules.
19. the device of assisting in correcting pronunciation as claimed in claim 18 also comprises: an output module, in order to export this errors number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810170987A CN101727764A (en) | 2008-10-21 | 2008-10-21 | Method and device for assisting in correcting pronunciation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810170987A CN101727764A (en) | 2008-10-21 | 2008-10-21 | Method and device for assisting in correcting pronunciation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101727764A true CN101727764A (en) | 2010-06-09 |
Family
ID=42448594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200810170987A Pending CN101727764A (en) | 2008-10-21 | 2008-10-21 | Method and device for assisting in correcting pronunciation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101727764A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118354A (en) * | 2015-09-14 | 2015-12-02 | 百度在线网络技术（北京）有限公司 | Data processing method for language learning and device thereof |
| CN109697975A (en) * | 2017-10-20 | 2019-04-30 | 深圳市鹰硕音频科技有限公司 | A kind of Speech Assessment Methods and device |
| CN110085261A (en) * | 2019-05-16 | 2019-08-02 | 上海流利说信息技术有限公司 | A kind of pronunciation correction method, apparatus, equipment and computer readable storage medium |
| CN112908363A (en) * | 2021-01-21 | 2021-06-04 | 北京乐学帮网络技术有限公司 | Pronunciation detection method and device, computer equipment and storage medium |
-
2008
- 2008-10-21 CN CN200810170987A patent/CN101727764A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118354A (en) * | 2015-09-14 | 2015-12-02 | 百度在线网络技术（北京）有限公司 | Data processing method for language learning and device thereof |
| CN109697975A (en) * | 2017-10-20 | 2019-04-30 | 深圳市鹰硕音频科技有限公司 | A kind of Speech Assessment Methods and device |
| CN110085261A (en) * | 2019-05-16 | 2019-08-02 | 上海流利说信息技术有限公司 | A kind of pronunciation correction method, apparatus, equipment and computer readable storage medium |
| CN110085261B (en) * | 2019-05-16 | 2021-08-24 | 上海流利说信息技术有限公司 | Pronunciation correction method, device, equipment and computer readable storage medium |
| CN112908363A (en) * | 2021-01-21 | 2021-06-04 | 北京乐学帮网络技术有限公司 | Pronunciation detection method and device, computer equipment and storage medium |
| CN112908363B (en) * | 2021-01-21 | 2022-11-22 | 北京乐学帮网络技术有限公司 | Pronunciation detection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8226416B2 (en) | Method and apparatus for reading education | |
| US8272874B2 (en) | System and method for assisting language learning | |
| CN101105939B (en) | Sonification guiding method | |
| US6424935B1 (en) | Two-way speech recognition and dialect system | |
| US8033831B2 (en) | System and method for programmatically evaluating and aiding a person learning a new language | |
| KR19990044575A (en) | Interactive language training apparatus | |
| Wang et al. | Towards automatic assessment of spontaneous spoken English | |
| CN101551947A (en) | Computer system for assisting spoken language learning | |
| CN101739870A (en) | Interactive language learning system and method | |
| US8221126B2 (en) | System and method for performing programmatic language learning tests and evaluations | |
| Karlina et al. | Designing phonetic alphabets for Bahasa Indonesia (PABI) for the teaching of intelligible English pronunciation in Indonesia | |
| CN101727764A (en) | Method and device for assisting in correcting pronunciation | |
| Felker et al. | Evaluating dictation task measures for the study of speech perception | |
| KR20140087956A (en) | Apparatus and method for learning phonics by using native speaker's pronunciation data and word and sentence and image data | |
| CN111508522A (en) | Statement analysis processing method and system | |
| US20050158696A1 (en) | [interactive computer-assisted language learning method and system thereof] | |
| KR100701270B1 (en) | Online Lecture and Evaluation System and Method of Foreign Languages | |
| CN109671308B (en) | Generation method of pronunciation mouth shape correction system | |
| TWM553866U (en) | Electronic book device for language learning based on text-to-voice technology | |
| JP7039637B2 (en) | Information processing equipment, information processing method, information processing system, information processing program | |
| TWI731493B (en) | Multi-lingual speech recognition and theme-semanteme analysis method and device | |
| TWI687804B (en) | System and method for detecting and correcting error pronunciation | |
| CN109686141B (en) | Pronunciation mouth shape correction system for language learning | |
| TW201017647A (en) | Auxiliary speech correcting device and method | |
| KR20180012233A (en) | Mobile english education method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100609 |