CN110176249A - A kind of appraisal procedure and device of spoken language pronunciation - Google Patents
A kind of appraisal procedure and device of spoken language pronunciation Download PDFInfo
- Publication number
- CN110176249A CN110176249A CN201910266722.4A CN201910266722A CN110176249A CN 110176249 A CN110176249 A CN 110176249A CN 201910266722 A CN201910266722 A CN 201910266722A CN 110176249 A CN110176249 A CN 110176249A
- Authority
- CN
- China
- Prior art keywords
- word
- pronunciation
- consonant
- feature
- gop
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 239000000284 extract Substances 0.000 claims abstract description 4
- 238000000605 extraction Methods 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; 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 TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; 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
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The present invention relates to a kind of appraisal procedure of spoken language pronunciation and devices, which comprises the spoken content of text for answering result of the examinee of acquisition;Extract each word pronunciation character in the content of text;Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;Based on the quantity of orthoepic word, the spoken language pronunciation result of examinee is assessed.The spoken language of examinee is answered, the present invention can determine the quantity of orthoepic word based on the pronunciation character of each word, so that the spoken language pronunciation effect for assessing examinee is good or bad, thus the targetedly spoken language proficiency of promotion examinee.
Description
Technical field
The present invention relates to field of computer technology, more particularly to the appraisal procedure and device of a kind of spoken language pronunciation.
Background technique
As the important medium of interpersonal communication, conversational language occupies extremely important status in real life.With society
The progress of continuous development and economical globalization tendency that can be economic, people are objective to the efficiency of language learning and language assessment
Property, fairness and scale test propose increasingly higher demands.For example oral composition of Open-ended Question type in speaking test, story
Repetition and picture talk etc. are to reflect an important topic type of the ability to express of examinee's spoken language.In general, teacher is in addition in terms of content
Outside being judged, also judge the student pronunciation of words whether standard, and whole pronunciation situation.
Traditional speaking test points-scoring system is directly to learn Rating Model according to the total score labeled data of teacher's marking, is given
A total score output out.And pronunciation of the student in oral expression whether standard, whole pronunciation what state has no way of learning.
Summary of the invention
Based on this, it is necessary to aiming at the problem that pronunciation of current speaking test is difficult to assess, provide a kind of commenting for spoken language pronunciation
Estimate method and device.
A kind of appraisal procedure of spoken language pronunciation, which comprises
The spoken content of text for answering result of the examinee of acquisition;
Extract the pronunciation character of each word in the content of text;
Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;
Based on the quantity of orthoepic word, the spoken language pronunciation result of examinee is assessed.
In the present embodiment, the pronunciation character includes the acoustics likelihood feature of word, described to extract in the content of text often
The pronunciation character of a word, comprising:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By the acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
In the present embodiment, the pronunciation character includes gop feature, the pronunciation for extracting each word in the content of text
Feature, comprising:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
In the present embodiment, the pronunciation character includes the consistency feature of pronunciation, described to extract in the content of text often
The pronunciation character of a word, comprising:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as the equivalent.
In the present embodiment, the pronunciation character includes the accuracy feature of pronunciation, described to extract in the content of text often
The pronunciation character of a word, comprising:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant determines that each word medial vowel and consonant articulation are correct
Frame accuracy;
Each word medial vowel and the correct frame accuracy of consonant articulation are being determined as the pronunciation of each word just
True rate feature.
A kind of assessment device of spoken language pronunciation, described device include:
Module is obtained, the spoken content of text for answering result of the examinee for acquisition;
Extraction module, for extracting the pronunciation character of each word in the content of text;
Determining module determines whether the pronunciation of each word is correct for the pronunciation character based on each word;
Evaluation module assesses the spoken language pronunciation result of examinee for the quantity based on orthoepic word.
In the present embodiment, the pronunciation character includes the acoustics likelihood feature of word, and the extraction module is used for:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By the acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
In the present embodiment, the pronunciation character includes gop feature, and the extraction module is used for:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
In the present embodiment, the pronunciation character includes the consistency feature of pronunciation, and the extraction module is used for:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as the equivalent.
In the present embodiment, the pronunciation character includes the accuracy feature of pronunciation, and the extraction module is used for:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant determines that each word medial vowel and consonant articulation are correct
Frame accuracy;
Each word medial vowel and the correct frame accuracy of consonant articulation are being determined as the pronunciation of each word just
True rate feature.
In the present invention, after the spoken content of text for answering result of the examinee of acquisition, the content of text can be extracted
In each word pronunciation character;Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;Just based on pronunciation
The quantity of true word assesses the spoken language pronunciation result of examinee.It is answered accordingly, for the spoken language of examinee, the present invention can be based on every
The pronunciation character of a word, determines the quantity of orthoepic word, thus the spoken language pronunciation effect for assessing examinee be it is good or bad, from
And targetedly promote the spoken language proficiency of examinee.
Detailed description of the invention
Fig. 1 is the flow chart of the appraisal procedure of the spoken language pronunciation of an embodiment;
Fig. 2 is the structure chart of the assessment device of the spoken language pronunciation of an embodiment.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Fig. 1 is the flow chart of the appraisal procedure of the spoken language pronunciation of an embodiment.As shown in Figure 1, this method comprises:
Step 110, the spoken content of text for answering result of the examinee of acquisition;
Step 120, the pronunciation character of each word in content of text is extracted;
Step 130, the pronunciation character based on each word determines whether the pronunciation of each word is correct;
Step 140, the quantity based on orthoepic word assesses the spoken language pronunciation result of examinee.
In the present invention, after the spoken content of text for answering result of the examinee of acquisition, the content of text can be extracted
In each word pronunciation character;Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;Just based on pronunciation
The quantity of true word assesses the spoken language pronunciation result of examinee.It is answered accordingly, for the spoken language of examinee, the present invention can be based on every
The pronunciation character of a word, determines the quantity of orthoepic word, thus the spoken language pronunciation effect for assessing examinee be it is good or bad, from
And targetedly promote the spoken language proficiency of examinee.
Wherein, the spoken result of answering of examinee can be by the audio file etc. of voice acquisition system acquisition.
Content of text can be speech recognition system from the spoken word content answered and extracted in result.It is appreciated that this
Embodiment is not limited to the spoken acquisition modes for answering result and content of text.
In the present embodiment, pronunciation character can be acoustics likelihood feature, gop feature, the consistency feature of pronunciation, pronunciation
At least one of accuracy feature etc..
In one implementation of the present embodiment, pronunciation character includes the acoustics likelihood feature of word, is extracted every in content of text
A word pronunciation character, comprising:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
Wherein it is possible to count the acoustics Likelihood Score of word in content of text, and consider the difference of vowel, consonant, specifically may be used
To be based on following data:
The frame of word grade is averaged likelihood score (mean, max, min)
The frame of word medial vowel is averaged likelihood score (mean, max, min)
The frame of consonant is averaged likelihood score (mean, max, min) in word
Word medial vowel number, consonant number, phone total number
In one implementation of the present embodiment, pronunciation character includes gop feature, and it is special to extract each word pronunciation in content of text
Sign, comprising:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
Wherein, the phone grade gop marking of word in content of text is counted, and considers the difference of vowel, consonant, it can be specific
It is as follows:
The mean, max, min of the gop marking of word medial vowel;
The mean, max, min of the gop marking of consonant in word;
The mean, max, min of the gop marking of phone in word.
In one implementation of the present embodiment, pronunciation character includes the consistency feature of pronunciation, is extracted every in content of text
A word pronunciation character, comprising:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as equivalent.
Wherein it is possible to count the consistency of the vowel of fa, rec of input word $ word_i $, consonant, all in content of text
(unit: frame)
All pronunciation consistency:
agrall=sum(HitFramesphonex)/NumFramesAll;
Vowel consistency:
agrvowels=sum(HitFramesvowelsx)/NumFramesVow;
Consonant consistency:
agrconsonants=sum(Hitframesconsonants)/NumFramesCon;
HitFrames_x indicates the number that the consistent frame of phone is corresponded in rec and fa;NumFramesX indicates word's
All or part of frame number.
In one implementation of the present embodiment, pronunciation character includes the accuracy feature of pronunciation, is extracted each in content of text
Word pronunciation character, comprising:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant determines that each word medial vowel and consonant articulation are correct
Frame accuracy;
Each word medial vowel and the correct frame accuracy of consonant articulation are determined as to the accuracy feature of the pronunciation of each word.
Count the accounting (unit: a of the consistent number of vowel, consonant, all of fa, rec of input word in content of text
Number):
The accuracy of all pronunciations:
CountsAccall=sum(Hitphonex)/NumCountsAll;
The accuracy of vowel:
CountsAcclvowels=sum (Hitvowelsx)/NumCountsVow;
The accuracy of consonant:
CountsAccconsonants=sum (Hitconsonants)/NumCountsCon;
Hit_x indicates to be 1 if fa is more than or equal to some threshold value (0.5) with the consistency of the corresponding phone of rec, otherwise
It is 0;After NumCountsX indicates the phone number of some or all of word, input word in content of text can be counted
Fa, rec accuracy (unit: frame) of vowel, consonant, all:
The frame accuracy of all pronunciations:
FramesAccall=sum (FramesHitPhonex)/NumFramesAll;
The frame accuracy of vowel:
FramesAccvowels=sum (FramesHitVowelsx)/NumFramesVow;
The frame accuracy of consonant:
FramesAccconsonants=sum (FramesHitConsonants)/NumFramesCon;
Frames_HitX is indicated
Otherwise frame number of the phone in fa is 0;NumFramesX indicates the frame number of some or all of word
It, can when determining whether the pronunciation of each word is correct based on the pronunciation character of each word in one embodiment of this implementation
To use Multilayer Perception classifier (MLP) algorithm.Mainly consider that its model complexity is controllable, supports the training of high-volume data.
It can will be in the above acoustics likelihood feature, gop feature, the consistency feature of pronunciation, accuracy feature of pronunciation etc.
At least one be input in MLP model, so that it is determined that whether the pronunciation of each word correct.
Quantity later based on orthoepic word, it can be estimated that the spoken language pronunciation result of examinee.Specifically, can be based on
It is divided by with total word quantity, obtains a ratio, can determine examinee's sheet based on the ratio by the quantity of orthoepic word
The secondary spoken overall marking answered.Ratio and overall marking are proportional, and therefore, overall marking is based on whole section of spoken Open-ended Question type
The measurement of the word accuracy of sample, i.e. accuracy is higher, then shows that the whole pronunciation situation of the sample is better.
Fig. 2 is the structure chart of the assessment device of the spoken language pronunciation of an embodiment.As shown in Fig. 2, the device includes:
Module 210 is obtained, the spoken content of text for answering result of the examinee for acquisition;
Extraction module 220, for extracting the pronunciation character of each word in content of text;
Determining module 230 determines whether the pronunciation of each word is correct for the pronunciation character based on each word;
Evaluation module 240 assesses the spoken language pronunciation result of examinee for the quantity based on orthoepic word.
In the present invention, after the spoken content of text for answering result of the examinee of acquisition, the content of text can be extracted
In each word pronunciation character;Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;Just based on pronunciation
The quantity of true word assesses the spoken language pronunciation result of examinee.It is answered accordingly, for the spoken language of examinee, the present invention can be based on every
The pronunciation character of a word, determines the quantity of orthoepic word, thus the spoken language pronunciation effect for assessing examinee be it is good or bad, from
And targetedly promote the spoken language proficiency of examinee.
In one implementation of the present embodiment, pronunciation character includes the acoustics likelihood feature of word, and extraction module is used for:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
In one implementation of the present embodiment, pronunciation character includes gop feature, and extraction module is used for:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
In one implementation of the present embodiment, pronunciation character includes the consistency feature of pronunciation, and extraction module is used for:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as equivalent.
In one implementation of the present embodiment, pronunciation character includes the accuracy feature of pronunciation, and extraction module is used for:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant determines that each word medial vowel and consonant articulation are correct
Frame accuracy;
Each word medial vowel and the correct frame accuracy of consonant articulation are determined as to the accuracy feature of the pronunciation of each word.
The realization process of apparatus above of the present invention and the realization process of above method are identical, are specifically referred to above method
The detailed process of embodiment, the present invention are no longer specifically described.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of appraisal procedure of spoken language pronunciation, which is characterized in that the described method includes:
The spoken content of text for answering result of the examinee of acquisition;
Extract the pronunciation character of each word in the content of text;
Based on the pronunciation character of each word, determine whether the pronunciation of each word is correct;
Based on the quantity of orthoepic word, the spoken language pronunciation result of examinee is assessed.
2. the method according to claim 1, wherein the pronunciation character includes the acoustics likelihood feature of word, institute
It states and extracts word pronunciation character each in the content of text, comprising:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By the acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
3. the method according to claim 1, wherein the pronunciation character includes gop feature, described in the extraction
The pronunciation character of each word in content of text, comprising:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
4. the method according to claim 1, wherein the pronunciation character includes the consistency feature of pronunciation, institute
State the pronunciation character for extracting each word in the content of text, comprising:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as the equivalent.
5. the method according to claim 1, wherein the pronunciation character includes the accuracy feature of pronunciation, institute
State the pronunciation character for extracting each word in the content of text, comprising:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant is determining each word medial vowel and the correct frame of consonant articulation just
True rate;
Each word medial vowel and the correct frame accuracy of consonant articulation are determined as to the accuracy of the pronunciation of each word
Feature.
6. a kind of assessment device of spoken language pronunciation, which is characterized in that described device includes:
Module is obtained, the spoken content of text for answering result of the examinee for acquisition;
Extraction module, for extracting the pronunciation character of each word in the content of text;
Determining module determines whether the pronunciation of each word is correct for the pronunciation character based on each word;
Evaluation module assesses the spoken language pronunciation result of examinee for the quantity based on orthoepic word.
7. device according to claim 7, which is characterized in that the pronunciation character includes the acoustics likelihood feature of word, institute
Extraction module is stated to be used for:
The frame of vowel and consonant based on each word is averaged likelihood score, determines the acoustics Likelihood Score of each word;
By the acoustics Likelihood Score to the acoustics likelihood feature that should be used as each word.
8. device according to claim 6, which is characterized in that the pronunciation character includes gop feature, the extraction module
For:
Obtain the gop marking of each word medial vowel and consonant;
It is given a mark based on the gop of each word medial vowel and consonant and determines the gop marking of corresponding each word;
Gop feature by the gop marking of each word as equivalent.
9. device according to claim 6, which is characterized in that the pronunciation character includes the consistency feature of pronunciation, institute
Extraction module is stated to be used for:
Determine the number of each word medial vowel and the consonant frame consistent with standard pronunciation;
The consistency feature that the number for the frame that each word medial vowel is consistent with consonant is pronounced as the equivalent.
10. device according to claim 6, which is characterized in that the pronunciation character includes the accuracy feature of pronunciation, institute
Extraction module is stated to be used for:
Obtain the number of each word medial vowel and the correct frame of consonant;
Number based on each word medial vowel and the correct frame of consonant is determining each word medial vowel and the correct frame of consonant articulation just
True rate;
Each word medial vowel and the correct frame accuracy of consonant articulation are determined as to the accuracy of the pronunciation of each word
Feature.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910266722.4A CN110176249A (en) | 2019-04-03 | 2019-04-03 | A kind of appraisal procedure and device of spoken language pronunciation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910266722.4A CN110176249A (en) | 2019-04-03 | 2019-04-03 | A kind of appraisal procedure and device of spoken language pronunciation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110176249A true CN110176249A (en) | 2019-08-27 |
Family
ID=67689392
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910266722.4A Pending CN110176249A (en) | 2019-04-03 | 2019-04-03 | A kind of appraisal procedure and device of spoken language pronunciation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110176249A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110648690A (en) * | 2019-09-26 | 2020-01-03 | 广州三人行壹佰教育科技有限公司 | Audio evaluation method and server |
CN110782921A (en) * | 2019-09-19 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Voice evaluation method and device, storage medium and electronic device |
CN112614510A (en) * | 2020-12-23 | 2021-04-06 | 北京猿力未来科技有限公司 | Audio quality evaluation method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN104485115A (en) * | 2014-12-04 | 2015-04-01 | 上海流利说信息技术有限公司 | Pronunciation evaluation equipment, method and system |
CN108682437A (en) * | 2018-05-18 | 2018-10-19 | 网易乐得科技有限公司 | Information processing method, device, medium and computing device |
CN108711319A (en) * | 2018-05-24 | 2018-10-26 | 李炜 | A kind of international professional Chinese teaching method and system |
CN109545243A (en) * | 2019-01-23 | 2019-03-29 | 北京猎户星空科技有限公司 | Pronunciation quality evaluating method, device, electronic equipment and storage medium |
-
2019
- 2019-04-03 CN CN201910266722.4A patent/CN110176249A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN104485115A (en) * | 2014-12-04 | 2015-04-01 | 上海流利说信息技术有限公司 | Pronunciation evaluation equipment, method and system |
CN108682437A (en) * | 2018-05-18 | 2018-10-19 | 网易乐得科技有限公司 | Information processing method, device, medium and computing device |
CN108711319A (en) * | 2018-05-24 | 2018-10-26 | 李炜 | A kind of international professional Chinese teaching method and system |
CN109545243A (en) * | 2019-01-23 | 2019-03-29 | 北京猎户星空科技有限公司 | Pronunciation quality evaluating method, device, electronic equipment and storage medium |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110782921A (en) * | 2019-09-19 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Voice evaluation method and device, storage medium and electronic device |
CN110782921B (en) * | 2019-09-19 | 2023-09-22 | 腾讯科技(深圳)有限公司 | Voice evaluation method and device, storage medium and electronic device |
CN110648690A (en) * | 2019-09-26 | 2020-01-03 | 广州三人行壹佰教育科技有限公司 | Audio evaluation method and server |
CN112614510A (en) * | 2020-12-23 | 2021-04-06 | 北京猿力未来科技有限公司 | Audio quality evaluation method and device |
CN112614510B (en) * | 2020-12-23 | 2024-04-30 | 北京猿力未来科技有限公司 | Audio quality assessment method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110782921B (en) | Voice evaluation method and device, storage medium and electronic device | |
CN109545244A (en) | Speech evaluating method, device, electronic equipment and storage medium | |
US8226416B2 (en) | Method and apparatus for reading education | |
Richards | Conversation | |
CN110176249A (en) | A kind of appraisal procedure and device of spoken language pronunciation | |
JP2005321817A (en) | Method and apparatus for obtaining combining information from speech signals for adaptive interaction in teaching and testing | |
KR20010074705A (en) | Automated language assessment using speech recognition modeling | |
JP2009503563A (en) | Assessment of spoken language proficiency by computer | |
Uchanski et al. | Automatic speech recognition to aid the hearing impaired: prospects for the automatic generation of cued speech. | |
CN101393694A (en) | Chinese character pronunciation studying device with pronunciation correcting function of Chinese characters, and method therefor | |
Inoue et al. | A Study of Objective Measurement of Comprehensibility through Native Speakers' Shadowing of Learners' Utterances. | |
CN110164422A (en) | A kind of the various dimensions appraisal procedure and device of speaking test | |
CN107910059A (en) | A kind of language ability obstacle hierarchy system and its implementation | |
CN109584906A (en) | Spoken language pronunciation evaluating method, device, equipment and storage equipment | |
US11138379B2 (en) | Determination of transcription accuracy | |
Cámara Arenas et al. | Automatic pronunciation assessment vs. automatic speech recognition: A study of conflicting conditions for L2-English | |
KR100997255B1 (en) | Language learning system of simultaneous interpretation type using voice recognition | |
KR102407055B1 (en) | Apparatus and method for measuring dialogue quality index through natural language processing after speech recognition | |
CN113327615B (en) | Voice evaluation method, device, equipment and storage medium | |
CN115099222A (en) | Punctuation mark misuse detection and correction method, device, equipment and storage medium | |
Davies | Language assessment in call centres: The case of the customer service representative | |
CN114708854A (en) | Voice recognition method and device, electronic equipment and storage medium | |
Jun et al. | Factors in Raters’ Perceptions of Comprehensibility and Accentedness | |
US20210304628A1 (en) | Systems and Methods for Automatic Video to Curriculum Generation | |
Nath | Towards naturally responsive spoken dialog systems by modelling pragmatic-prosody correlations of discourse markers |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190827 |