CN106856095A - The voice quality evaluating system that a kind of phonetic is combined into syllables - Google Patents
The voice quality evaluating system that a kind of phonetic is combined into syllables Download PDFInfo
- Publication number
- CN106856095A CN106856095A CN201510904230.5A CN201510904230A CN106856095A CN 106856095 A CN106856095 A CN 106856095A CN 201510904230 A CN201510904230 A CN 201510904230A CN 106856095 A CN106856095 A CN 106856095A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- module
- tone
- phoneme
- phonetic
- 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
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/93—Discriminating between voiced and unvoiced parts of speech signals
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification
- G10L17/06—Decision making techniques; Pattern matching strategies
- G10L17/08—Use of distortion metrics or a particular distance between probe pattern and reference templates
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/90—Pitch determination of speech signals
Abstract
The invention provides the voice quality evaluating system that a kind of phonetic is combined into syllables, the system includes:Front end processing block, pressure alignment module, evaluating pronunciation module, tone evaluation and test module and evaluation and test output module;The front end processing block, end-point detection, the extraction of acoustic feature and regular are carried out for combining voice into syllables to the phonetic being input into;The pressure alignment module, for the use of force alignment techniques being several phonemes section the phonetic segmentation after front end processing block treatment, between obtaining at the beginning of each phoneme and deadline;The evaluating pronunciation module, for calculating every section of posterior probability of phoneme using wellness of pronouncing;The tone evaluation and test module, it is identified for the tone to every section of phoneme, and judge whether recognition result is consistent with just intonation, provide the assessed value of tone quality, the evaluation and test output module, for carrying out synthesis to the evaluation result that sound evaluation and test module and tone evaluation and test module are exported, overall evaluation result is obtained, and feed back to user.
Description
Technical field
The invention belongs to computer-aided learning (CAL) field, more particularly to the voice quality evaluating system that a kind of phonetic is combined into syllables.
Background technology
Continuous improvement and international association recently as China's international status it is increasingly extensive, countries in the world are for the Chinese
The demand sharp increase of language study.Chinese is the most language of population in use in the world, is also the six of the United Nations's regulation
One of working language is planted, it is by the weight of increasing national government, educational institution, enterprise and medium
Depending on.According to statistics, the foreigner of learning Chinese increases substantially every year, and there are the 2000 of nearly 100 countries in the current whole world
Chinese is more included regular education system, the Global Learning Chinese by Duo Suo universities in professor's Chinese, the state such as English, Thailand, Indonesia
The number of language reaches nearly 30,000,000, it is contemplated that the number of Global Learning Chinese will be up to 100,000,000 people after the several years.
2 months 1958, the 5th meeting of first National People's Congress passed through《The Scheme for the Chinese Phonetic Alphabet》, it
In more than 50 years afterwards, the Chinese phonetic alphabet have passed through test of many times and reform, and just development is fixed to present " teaching material ".
The present Chinese phonetic alphabet has become learning Chinese characters, learning putonghua, culture and improves the important of reading and writing ability
Instrument.The Chinese phonetic alphabet is made up of 26 Latin alphabets, including 21 initial consonants and 39 simple or compound vowel of a Chinese syllable.
The teaching of the Chinese phonetic alphabet, be most initially according to《The Scheme for the Chinese Phonetic Alphabet》Implement, but in the process of teaching
In, vast language teaching worker has found that allowing beginner to grasp complicated spelling rules has very big difficulty.Then,
In teaching practice, the effective teaching method of the continuous Scientific exploration of teaching person updates teaching material, makes it more
Plus the characteristics of adaptation beginner, mitigate the pressure of beginner.The mode that wherein phonetic is combined into syllables is that teaching person is closed very much
The problem of note because rational spelled speech mode can help beginner it is incremental learn the Chinese phonetic alphabet well.
Constantly improved by teaching person, current phonetic spelled speech mode mainly there are four kinds:The spelling of sound two,
Sound is situated between and is merged method, Three kinds of hors d'oeuvres liaison method and directly exhales method.Wherein, the spelling of sound two is the most frequently used, the maximally effective side of combining into syllables
Formula.
The spelling of sound two refer to syllable according to initial consonant, simple or compound vowel of a Chinese syllable successively it is bright run through after whole reading again mode, such as:
B+an4=ban4, q+iang3=qiang3 etc..Typically there was only two phonetic sections of initial consonant and simple or compound vowel of a Chinese syllable in teaching now, adopt more
With the spelling of sound two, because this mode splits simply, it is easily mastered, only should be noted that initial consonant is gently short, simple or compound vowel of a Chinese syllable is heavier,
Initial consonant reader sound, simple or compound vowel of a Chinese syllable directly band adjust the tricks of the trade such as read just can be with.
Since the nineties in last century, domestic many research institutions and scholar are carried out for the voice quality evaluation and test of Chinese
Related research, has also emerged in large numbers the systems of many practicalities.But, traditional voice quality evaluation and test primarily directed to
The whole form read or read aloud paragraph, evaluation and test unit is larger, it is impossible to well to independent initial consonant or simple or compound vowel of a Chinese syllable aspect
On pronunciation advantage and disadvantage carry out detailed evaluation.
The content of the invention
It is an object of the invention to overcome current pinyin pronunciation quality assessment to exist cannot to independent initial consonant or
A kind of defect that pronunciation advantage and disadvantage in simple or compound vowel of a Chinese syllable aspect are evaluated, it is proposed that the voice quality evaluation and test system that phonetic is combined into syllables
System, the system evaluates and tests technology and tone evaluation and test technology, Neng Gou under the framework of speech recognition using voice quality
Evaluated respectively on single initial consonant, simple or compound vowel of a Chinese syllable and whole read layer face, effectively aided in the user learning Chinese phonetic alphabet.
To achieve the above object, the invention provides the voice quality evaluating system that a kind of phonetic is combined into syllables, the system
Including:Front end processing block, pressure alignment module, evaluating pronunciation module, tone evaluation and test module and evaluation and test output mould
Block;
The front end processing block, the extraction of end-point detection, acoustic feature is carried out for combining voice into syllables to the phonetic being input into
And it is regular;
The pressure alignment module, for being cut using the voice after forcing alignment techniques that the front end processing block is processed
It is divided into several phonemes section, between obtaining at the beginning of each phoneme and deadline;And the phoneme section input that will be obtained
The evaluating pronunciation module and tone evaluation and test module;
The evaluating pronunciation module, for calculating every section of posterior probability of phoneme using wellness of pronouncing;As measurement sound
Plain voice quality is estimated;By value output to the evaluation and test output module;
The tone evaluation and test module, is identified for the tone to every section of phoneme, and judges recognition result with correct sound
Whether consistent adjust, provide the assessed value of tone quality, by value output to the evaluation and test output module;
The evaluation and test output module, it is comprehensive for being carried out to the evaluation result that sound evaluation and test module and tone evaluation and test module are exported
Close, obtain overall evaluation result, and feed back to user, allow users to be carried out targetedly according to evaluation result
Correct and practise.
In above-mentioned technical proposal, the end-point detection of the front end processing block is that the phonetic for detecting input combines voice into syllables
At the beginning of between point and end time point.
In above-mentioned technical proposal, the wellness of pronouncing of the utilization in the evaluating pronunciation module calculates every section of posteriority of phoneme
The detailed process of probability is:
Phoneme q is calculated using following formulaiPosterior probability:
Wherein, O is phonetic feature sequence;p(O|qi) it is in phoneme qiDown it was observed that the probability of O;J is system institute
There is the number of phoneme;NF (O) is the segment length of phonetic feature sequence.
In above-mentioned technical proposal, the detailed process being identified to the tone of every section of phoneme in the tone evaluation and test module
For:According to force alignment module obtain each phoneme section at the beginning of between and deadline, calculate simple or compound vowel of a Chinese syllable fundamental frequency sequence
Row, and extract tone feature;Then tone feature is classified using mixed Gauss model, obtains the knowledge of tone
Other result.
Advantage of the invention is that:
1st, system of the invention is used as a kind of computer-assisted instruction instrument, can Real-time Feedback user pronounce at each
Details in aspect.In the engineering of language learning, learner is not in the case where needing teacher to pay close attention to regard to energy
Access on oneself voice quality quality Detailed Feedback, for different study contents, can word, initial consonant,
Each different aspect such as simple or compound vowel of a Chinese syllable and tone and level obtain different feedbacks;
2nd, whether just the pronunciation that system of the invention time tells learner oneself in single initial consonant, simple or compound vowel of a Chinese syllable and whole read layer
Really, the quality of voice quality, such learner is just able to know that the quality of oneself pronunciation, changes the pronunciation of oneself,
Improve the level of pronunciation;
3rd, system of the invention is based on computer-aided learning (CAL) technology, can not only save substantial amounts of teacher resource, also
Can allow learner's flexible arrangement learning time of oneself and study place.
Brief description of the drawings
Fig. 1 is the general frame figure of the voice quality evaluating system that phonetic of the invention is combined into syllables.
Specific embodiment
The present invention will be further described in detail with specific embodiment below in conjunction with the accompanying drawings.
As shown in figure 1, the voice quality evaluating system that a kind of phonetic is combined into syllables, the system includes:Front end processing block,
Force alignment module, evaluating pronunciation module, tone evaluation and test module and evaluation and test output module;
The front end processing block, the extraction of end-point detection, acoustic feature is carried out for combining voice into syllables to the phonetic being input into
And it is regular;The end-point detection to detect input voice at the beginning of between point and end time point, so that the later stage is accurate
True is processed voice accordingly;In the present embodiment, the acoustic feature of use is to perceive linear prediction cepstrum coefficient
Coefficient (PLP), this feature simulates processing feature of the human ear to voice to a certain extent, and robustness is preferable.
The pressure alignment module, for being cut using the voice after forcing alignment techniques that the front end processing block is processed
It is divided into several phonemes section, between obtaining at the beginning of each phoneme and deadline;And the phoneme section input that will be obtained
The evaluating pronunciation module and tone evaluation and test module;
For the accuracy of alignment, in grammer network, each phoneme corresponding filling sound in parallel is used for
Absorb the ambiguous phoneme of pronunciation.Need to use corresponding acoustic model and pronunciation word during alignment is forced
Allusion quotation.
The evaluating pronunciation module, for using wellness (Goodness of Pronunciation, GOP) of pronouncing) meter
Calculate every section of posterior probability of phoneme;As estimating for measurement phoneme pronunciation quality;Value output is defeated to the evaluation and test
Go out module;
Phoneme q is calculated using following formulaiPosterior probability:
Wherein, O is phonetic feature sequence;p(O|qi) it is in phoneme qiDown it was observed that the probability of O;J is system institute
There is the number of phoneme;NF (O) is the segment length of phonetic feature sequence.
Phoneme posterior probability is a kind of good measure of similarity between description phonetic speech section to be measured and acoustic model.
Posterior probability is difficult to be occurred drastically to change with the change of the personal feature or sound channel of learner, and robustness is strong,
The similitude between the pronunciation of learner and RP is preferably reflected, is to use a kind of most common hair at present
Sound quality is estimated.
The tone evaluation and test module, for according to force alignment module obtain each phoneme section at the beginning of between and cut-off when
Between, the fundamental frequency sequence of simple or compound vowel of a Chinese syllable is calculated, and extract tone feature;Then tone feature is carried out using mixed Gauss model
Classification, obtains the recognition result of tone;Finally judge whether recognition result is consistent with just intonation, provide tone matter
The assessed value of amount, by value output to the evaluation and test output module;
The evaluation and test output module, it is comprehensive for being carried out to the evaluation result that sound evaluation and test module and tone evaluation and test module are exported
Close, export overall evaluation result and feed back to user, allow users to targetedly be corrected according to evaluation result
And exercise.
Performance evaluation is carried out to system of the invention below, the related evaluation metricses and term used are situated between first
Continue:
1st, positive sample:Refer to tone that learner reads aloud or pronounce completely the same with content of text to be evaluated, that is, send out
Sound or the right-on data of tone;
2nd, negative sample:Refer to tone that learner reads aloud or pronounce it is completely inconsistent with content of text to be evaluated, i.e.,
Pronunciation or tone incorrect data completely;
3rd, accuracy:It is a kind of evaluation metricses, is referred to and positive sample is evaluated and tested into the ratio of positive sample or negative sample is commented
Survey into the ratio of negative sample.During experiment statisticses, appraisal result is classified as positive sample more than 70 points, will commented
Point result is classified as negative sample less than 60 points;
4th, data volume:Tone, pronunciation each 300 of positive and negative samples, wherein syllable test data with combine test number into syllables
According to not being same batch of data.
It is compared with the result of voice quality evaluating system of the invention with traditional voice quality evaluating system, whole reading
Syllable mode comparative result is shown in Table 1:
Table 1
Positive sample | Negative sample | |
Tone | 94.6% | 86.6% |
Pronunciation | 83.4% | 88.9% |
The comparative result that sound two spells mode is shown in Table 2:
Table 2
It is can be found that by contrasting both the above form:In syllable aspect, evaluation and test performance of the invention is commented with traditional
Performance is surveyed almost to be consistent.On the premise of the effective evaluation and test that ensure that syllable aspect, in consonant, vowel aspect,
Evaluation and test performance of the present invention for positive sample is consistent with the evaluation and test property retention for syllable, is commented in the pronunciation of negative sample
Performance has declined in survey, it may be possible to which, because the female pronunciation of single sound is shorter, identification has certain difficulty.But
It is that, by the analysis to actual experiment and data, this auxiliary information still reflects learner's to a certain extent
Pronunciation situation.Therefore, system of the invention can not only react the voice quality of syllable aspect, it is also possible to effective
React the voice quality of single initial consonant and simple or compound vowel of a Chinese syllable aspect.
Claims (4)
1. a kind of voice quality evaluating system that phonetic is combined into syllables, the system includes:Front end processing block, pressure are right
Neat module, evaluating pronunciation module, tone evaluation and test module and evaluation and test output module;Characterized in that,
The front end processing block, the extraction of end-point detection, acoustic feature is carried out for combining voice into syllables to the phonetic being input into
And it is regular;
The pressure alignment module, for being cut using the voice after forcing alignment techniques that the front end processing block is processed
It is divided into several phonemes section, between obtaining at the beginning of each phoneme and deadline;And the phoneme section input that will be obtained
The evaluating pronunciation module and tone evaluation and test module;
The evaluating pronunciation module, for calculating every section of posterior probability of phoneme using wellness of pronouncing;As measurement sound
Plain voice quality is estimated;By value output to the evaluation and test output module;
The tone evaluation and test module, is identified for the tone to every section of phoneme, and judges recognition result with correct sound
Whether consistent adjust, provide the assessed value of tone quality, by value output to the evaluation and test output module;
The evaluation and test output module, it is comprehensive for being carried out to the evaluation result that sound evaluation and test module and tone evaluation and test module are exported
Close, obtain overall evaluation result, and feed back to user, allow users to be carried out targetedly according to evaluation result
Correct and practise.
2. the voice quality evaluating system that phonetic according to claim 1 is combined into syllables, it is characterised in that before described
The end-point detection for holding processing module be detect input phonetic combine into syllables at the beginning of voice between point and end time point.
3. the voice quality evaluating system that phonetic according to claim 1 is combined into syllables, it is characterised in that the hair
Utilization in sound evaluation and test module wellness of pronouncing calculates the detailed process of every section of posterior probability of phoneme and is:
Phoneme q is calculated using following formulaiPosterior probability:
Wherein, O is phonetic feature sequence;p(O|qi) it is in phoneme qiDown it was observed that the probability of O;J is system institute
There is the number of phoneme;NF (O) is the segment length of phonetic feature sequence.
4. the voice quality evaluating system that phonetic according to claim 1 is combined into syllables, it is characterised in that the sound
It is to the detailed process that the tone of every section of phoneme is identified in tune evaluation and test module:
According to force alignment module obtain each phoneme section at the beginning of between and deadline, calculate simple or compound vowel of a Chinese syllable fundamental frequency sequence
Row, and extract tone feature;Then tone feature is classified using mixed Gauss model, obtains the knowledge of tone
Other result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510904230.5A CN106856095A (en) | 2015-12-09 | 2015-12-09 | The voice quality evaluating system that a kind of phonetic is combined into syllables |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510904230.5A CN106856095A (en) | 2015-12-09 | 2015-12-09 | The voice quality evaluating system that a kind of phonetic is combined into syllables |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106856095A true CN106856095A (en) | 2017-06-16 |
Family
ID=59132545
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510904230.5A Pending CN106856095A (en) | 2015-12-09 | 2015-12-09 | The voice quality evaluating system that a kind of phonetic is combined into syllables |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106856095A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110010123A (en) * | 2018-01-16 | 2019-07-12 | 上海异构网络科技有限公司 | English phonetic word pronunciation learning evaluation system and method |
CN112331180A (en) * | 2020-11-03 | 2021-02-05 | 北京猿力未来科技有限公司 | Spoken language evaluation method and device |
CN112397091A (en) * | 2019-08-16 | 2021-02-23 | 庞帝教育公司 | Chinese speech comprehensive scoring and diagnosing system and method |
CN113035238A (en) * | 2021-05-20 | 2021-06-25 | 北京世纪好未来教育科技有限公司 | Audio evaluation method, device, electronic equipment and medium |
CN113421467A (en) * | 2021-06-15 | 2021-09-21 | 读书郎教育科技有限公司 | System and method for assisting in learning pinyin spelling and reading |
CN115440103A (en) * | 2022-09-27 | 2022-12-06 | 广州优谷信息技术有限公司 | Reading evaluation method, system, device and storage medium |
CN116403604A (en) * | 2023-06-07 | 2023-07-07 | 北京奇趣万物科技有限公司 | Child reading ability evaluation method and system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315733A (en) * | 2008-07-17 | 2008-12-03 | 安徽科大讯飞信息科技股份有限公司 | Self-adapting method aiming at computer language learning system pronunciation evaluation |
CN101661675A (en) * | 2009-09-29 | 2010-03-03 | 苏州思必驰信息科技有限公司 | Self-sensing error tone pronunciation learning method and system |
CN101739868A (en) * | 2008-11-19 | 2010-06-16 | 中国科学院自动化研究所 | Automatic evaluation and diagnosis method of text reading level for oral test |
CN101826263A (en) * | 2009-03-04 | 2010-09-08 | 中国科学院自动化研究所 | Objective standard based automatic oral evaluation system |
CN101887725A (en) * | 2010-04-30 | 2010-11-17 | 中国科学院声学研究所 | Phoneme confusion network-based phoneme posterior probability calculation method |
CN102163428A (en) * | 2011-01-19 | 2011-08-24 | 无敌科技(西安)有限公司 | Method for judging Chinese pronunciation |
-
2015
- 2015-12-09 CN CN201510904230.5A patent/CN106856095A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315733A (en) * | 2008-07-17 | 2008-12-03 | 安徽科大讯飞信息科技股份有限公司 | Self-adapting method aiming at computer language learning system pronunciation evaluation |
CN101739868A (en) * | 2008-11-19 | 2010-06-16 | 中国科学院自动化研究所 | Automatic evaluation and diagnosis method of text reading level for oral test |
CN101826263A (en) * | 2009-03-04 | 2010-09-08 | 中国科学院自动化研究所 | Objective standard based automatic oral evaluation system |
CN101661675A (en) * | 2009-09-29 | 2010-03-03 | 苏州思必驰信息科技有限公司 | Self-sensing error tone pronunciation learning method and system |
CN101887725A (en) * | 2010-04-30 | 2010-11-17 | 中国科学院声学研究所 | Phoneme confusion network-based phoneme posterior probability calculation method |
CN102163428A (en) * | 2011-01-19 | 2011-08-24 | 无敌科技(西安)有限公司 | Method for judging Chinese pronunciation |
Non-Patent Citations (1)
Title |
---|
张珑: "《发音质量自动评测技术》", 30 June 2015 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110010123A (en) * | 2018-01-16 | 2019-07-12 | 上海异构网络科技有限公司 | English phonetic word pronunciation learning evaluation system and method |
CN112397091A (en) * | 2019-08-16 | 2021-02-23 | 庞帝教育公司 | Chinese speech comprehensive scoring and diagnosing system and method |
CN112331180A (en) * | 2020-11-03 | 2021-02-05 | 北京猿力未来科技有限公司 | Spoken language evaluation method and device |
CN113035238A (en) * | 2021-05-20 | 2021-06-25 | 北京世纪好未来教育科技有限公司 | Audio evaluation method, device, electronic equipment and medium |
CN113421467A (en) * | 2021-06-15 | 2021-09-21 | 读书郎教育科技有限公司 | System and method for assisting in learning pinyin spelling and reading |
CN115440103A (en) * | 2022-09-27 | 2022-12-06 | 广州优谷信息技术有限公司 | Reading evaluation method, system, device and storage medium |
CN116403604A (en) * | 2023-06-07 | 2023-07-07 | 北京奇趣万物科技有限公司 | Child reading ability evaluation method and system |
CN116403604B (en) * | 2023-06-07 | 2023-11-03 | 北京奇趣万物科技有限公司 | Child reading ability evaluation method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106856095A (en) | The voice quality evaluating system that a kind of phonetic is combined into syllables | |
CN103065626B (en) | Automatic grading method and automatic grading equipment for read questions in test of spoken English | |
CN101739869B (en) | Priori knowledge-based pronunciation evaluation and diagnosis system | |
CN102568475B (en) | System and method for assessing proficiency in Putonghua | |
CN108922563B (en) | Based on the visual verbal learning antidote of deviation organ morphology behavior | |
CN103594087B (en) | Improve the method and system of oral evaluation performance | |
Bolaños et al. | FLORA: Fluent oral reading assessment of children's speech | |
Bolanos et al. | Automatic assessment of expressive oral reading | |
CN102426834B (en) | Method for testing rhythm level of spoken English | |
CN101814066A (en) | Text reading difficulty judging device and method thereof | |
CN103985391A (en) | Phonetic-level low power consumption spoken language evaluation and defect diagnosis method without standard pronunciation | |
Zhao et al. | Automatic Chinese pronunciation error detection using SVM trained with structural features | |
Duan et al. | A Preliminary study on ASR-based detection of Chinese mispronunciation by Japanese learners | |
Gomaa et al. | Arabic short answer scoring with effective feedback for students | |
Godde et al. | Evaluation of reading performance of primary school children: Objective measurements vs. subjective ratings | |
Yoon et al. | A comparison of grammatical proficiency measures in the automated assessment of spontaneous speech | |
Zechner et al. | Automatic scoring of children’s read-aloud text passages and word lists | |
Zhao | Study on the effectiveness of the asr-based english teaching software in helping college students’ listening learning | |
CN107436863A (en) | The evaluating method and device of English discourse readability degree | |
Coto-Solano et al. | Advances in completely automated vowel analysis for sociophonetics: Using end-to-end speech recognition systems with DARLA | |
CN113205729A (en) | Foreign student-oriented speech evaluation method, device and system | |
Hussein et al. | Mandarin tone perception and production by German learners | |
CN103021226B (en) | Voice evaluating method and device based on pronunciation rhythms | |
Luo et al. | Investigation of the effects of automatic scoring technology on human raters' performances in L2 speech proficiency assessment | |
Li et al. | English sentence pronunciation evaluation using rhythm and intonation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170616 |
|
RJ01 | Rejection of invention patent application after publication |