CN111599234A - Automatic English spoken language scoring system based on voice recognition - Google Patents
Automatic English spoken language scoring system based on voice recognition Download PDFInfo
- Publication number
- CN111599234A CN111599234A CN202010423027.7A CN202010423027A CN111599234A CN 111599234 A CN111599234 A CN 111599234A CN 202010423027 A CN202010423027 A CN 202010423027A CN 111599234 A CN111599234 A CN 111599234A
- Authority
- CN
- China
- Prior art keywords
- module
- voice
- recognition
- scoring
- language
- 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
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B7/00—Electrically-operated teaching apparatus or devices working with questions and answers
- G09B7/02—Electrically-operated teaching apparatus or devices working with questions and answers of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student
Abstract
The invention discloses an automatic English spoken language scoring system based on voice recognition, which comprises a client, a scoring server and a cloud database, wherein the client, the scoring server and the cloud database are connected with one another; the client comprises a voice extraction module, a voice processor and a communication module; the voice extraction module records the spoken voice of the user through the recording device, converts the spoken voice into a digital signal and outputs the digital signal to the voice recognizer; the sound processor includes: the device comprises a noise reduction module, a conversion module, a feature extraction module, a noise suppression module, an identification module and a control module. Has the advantages that: the voice recognition module and the scoring module in the application extract the speech flow comprehensive characteristics, the pronunciation accuracy direction characteristics, the fluency direction characteristics and the text semantic similarity direction characteristics in the recognition result, and then compare the similarity with the standard answer according to the calculation recognition result, thereby obtaining the similarity score.
Description
Technical Field
The invention relates to the technical field of spoken language scoring, in particular to an automatic scoring system for spoken English reading based on voice recognition.
Background
With the development of computer science and technology, information technology has been widely applied to education and teaching, which enriches teaching resources, improves learning environment, and makes the learning mode of students and the teaching mode of teachers change fundamentally. On the other hand, with the development of artificial intelligence, acoustics and linguistics, the voice intelligence technology has become a novel information technology, and linguistics teaching gradually advances towards computer-aided teaching. However, there are many technical problems involved in the english spoken language scrolling, and among them, there are mainly a speech recognition technique and a natural language processing technique.
In recent years, there have been some english oral examination systems based on computer and network technologies, such as a spoken examination system of shanghai foreign language education press, a system of blue pigeons, and the like, which implement separation of examiners and examinees on sites and support organization of large-scale oral examinations. But only the examination paper of objective questions is supported in the examination paper aspect. The examination paper marking work of the subjective questions still needs to invest a large amount of manpower and material resources. Such as a repeat question among the english examination subjective questions. Therefore, in reality, the scoring task of the spoken language expression level is still completely read by manpower, and the scoring task has strong subjectivity, time tightness and high intensity, so that the scoring quality is difficult to control.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides an automatic English reading scoring system based on voice recognition.
In order to achieve the purpose, the invention adopts the following technical scheme: an automatic English spoken language reading scoring system based on voice recognition comprises a client, a scoring server and a cloud database which are connected with one another;
the client comprises a voice extraction module, a voice processor and a communication module;
the voice extraction module records the spoken voice of the user through the recording device, converts the spoken voice into a digital signal and outputs the digital signal to the voice recognizer;
the sound processor includes: the device comprises a noise reduction module, a conversion module, a feature extraction module, a noise suppression module, an identification module and a control module;
the communication module collects signals of the voice extraction module and the voice processor and transmits the signals to the scoring server;
the scoring server comprises a scoring device and a statistical uploading module; the scoring device comprises an identification module and a scoring module;
the recognition module comprises an acoustic module, a language module and a specific recognition module, the acoustic module extracts acoustic features of the user answering audio to obtain an acoustic model, the language module obtains a language model according to the question information and the training text, and the recognition module decodes the user answering audio through the acoustic model and the language model to obtain a recognition result;
the scoring module comprises an effective feature extraction module and an effective feature scoring module, the effective feature extraction module is used for extracting comprehensive speech flow features in the recognition result, the comprehensive speech flow features comprise features in pronunciation accuracy direction, features in fluency direction and features in text semantic similarity direction in oral language assessment, and the features in the text semantic similarity direction comprise semantic relevance features and grammatical structure similarity features; the effective feature extraction module is used for extracting the semantic relatedness feature;
the statistical uploading module collects the effective characteristics in the scoring module and sorts the effective characteristics according to the similarity sequence of the effective characteristics, and the statistical uploading module uploads the sorted effective characteristics to the cloud database.
In the automatic grading system for spoken English reading based on voice recognition, the noise reduction module performs noise suppression on user voice acquired in real time or other stored voice to obtain voice information after noise reduction; the noise suppression module adopts at least one of a spectrum elimination method and/or a learning similarity method and/or a noise reduction automatic encoder to perform noise suppression;
the conversion module carries out Laplace transformation on the sound information to obtain Laplace frequency spectrum information;
the feature extraction module performs two-dimensional Fourier transform on the obtained Laplace spectrum information to obtain wave number spectrum features of Laplace transform sound information data;
the noise suppression module cuts out at least 5 time slice filter factors according to the wave number spectrum of the Laplace transform sound information data to form a filter group; and performing two-dimensional Fourier transform on the time-sliced information data to obtain a filtering information data wave number spectrum, performing environmental noise suppression on the whole recorded sound data wave number spectrum by using the filtering information data wave number spectrum, and transmitting the sound data subjected to the environmental noise suppression to a scoring server.
In the above automatic scoring system for spoken english reading based on voice recognition, the recognition module performs voice recognition on an input voice being input by referring to recognition target language information including label information and pronunciation information of each recognition target language included in recognition target words registered in advance in a voice recognition dictionary, using a voice recognition engine corresponding to english which is a language set in advance as a recognition target;
the cloud database is registered with a pronunciation information conversion rule indicating correspondence between pronunciation information of words in a plurality of languages, and the pronunciation information conversion is based on the pronunciation information conversion rule of the cloud database to convert the pronunciation information of the words between the languages.
In the above-described automatic spoken english reading scoring system based on voice recognition, the control module controls the reading information conversion module to convert the reading information of the different language into the reading information of the set language when a word of the different language, which is a language different from the set language, is included in a recognition target word to which the recognition target word information is referred by the voice recognition module, and the recognition module performs voice recognition by referring to the reading information of the set language after conversion of the input voice being input and the recognition target word information of the recognition target word registered in advance in the voice recognition dictionary.
In the above automatic scoring system for spoken english reading based on voice recognition, the semantic relevance features include: calculating semantic similarity scores of each word in the recognition result and each word in the standard answer; calculating semantic similarity scores of each word in the recognition result and each sentence in the standard answers; calculating the maximum value or the average value of the semantic similarity score in each word in the recognition result and each sentence in the standard answer as the similarity score between the word and the sentence; a similarity score between the user answer and the standard answer is calculated.
In the above automatic scoring system for spoken english reading based on voice recognition, the valid feature extraction module is configured to extract the grammar structure similarity feature, and includes: respectively establishing a grammar sequence vector for each sentence of the recognition result; respectively solving the grammatical structure similarity score of each sentence in the recognition result and each sentence in the standard answers, and taking the maximum value of the grammatical structure similarity score of each sentence in the recognition result as the grammatical structure similarity score of the sentence; and calculating the grammar structure similarity characteristic between the user answer and the standard answer by weighted average of the grammar structure similarity score of each sentence in the recognition result.
In the above automatic scoring system for spoken english reading based on voice recognition, the recognition module employs a decoding system based on large-scale continuous voice recognition, the acoustic model employs a hidden markov model, the language model employs a language model based on a metagrammar, and a multi-pass decoding technique is employed during decoding, wherein the multi-pass decoding includes direct decoding, unsupervised self-adaptation based on maximum linear likelihood regression, and secondary decoding; and the effective characteristic scoring module is used for scoring and training the comprehensive characteristics of the stream to obtain a scoring model, and scoring the recognition result according to the scoring model.
Compared with the prior art, the invention has the advantages that:
1. in a high-noise environment, filtering factors of at least 5 time slices are intercepted by a noise suppression module, the filtering information data wave number spectrum is utilized to perform environmental noise suppression on the whole recorded sound data wave number spectrum, and the environmental noise is suppressed, so that accurate sound identification under high noise is realized, and subsequent scoring operation is facilitated;
2. the voice recognition module and the scoring module in the application extract the speech flow comprehensive characteristics, the pronunciation accuracy direction characteristics, the fluency direction characteristics and the text semantic similarity direction characteristics in the recognition result, and then compare the similarity with the standard answer according to the calculation recognition result, thereby obtaining the similarity score.
Drawings
Fig. 1 is a system block diagram of an automatic scoring system for spoken english reading based on voice recognition according to the present invention;
fig. 2 is a system block diagram of a scoring device in an automatic spoken english reading scoring system based on voice recognition according to the present invention.
Detailed Description
The following examples are for illustrative purposes only and are not intended to limit the scope of the present invention.
Examples
Referring to fig. 1-2, an automatic grading system for spoken english reading based on voice recognition comprises a client, a grading server and a cloud database which are connected with each other;
the client comprises a voice extraction module, a voice processor and a communication module;
the voice extraction module records the spoken voice of the user through the recording device, converts the spoken voice into a digital signal and outputs the digital signal to the voice recognizer;
the sound processor includes: the device comprises a noise reduction module, a conversion module, a feature extraction module, a noise suppression module, an identification module and a control module;
the communication module collects signals of the voice extraction module and the voice processor and transmits the signals to the scoring server;
the scoring server comprises a scoring device and a statistical uploading module; the scoring device comprises an identification module and a scoring module;
the recognition module comprises an acoustic module, a language module and a specific recognition module, the acoustic module extracts acoustic features of the user answering audio to obtain an acoustic model, the language module obtains a language model according to the question information and the training text, and the recognition module decodes the user answering audio through the acoustic model and the language model to obtain a recognition result;
the scoring module comprises an effective feature extraction module and an effective feature scoring module, the effective feature extraction module is used for extracting comprehensive speech flow features in the recognition result, the comprehensive speech flow features comprise features in pronunciation accuracy direction, features in fluency direction and features in text semantic similarity direction in oral language assessment, and the features in the text semantic similarity direction comprise semantic relevance features and grammatical structure similarity features; the effective feature extraction module is used for extracting the semantic relatedness feature;
the statistical uploading module collects the effective characteristics in the scoring module and sorts the effective characteristics according to the similarity sequence of the effective characteristics, and the statistical uploading module uploads the sorted effective characteristics to the cloud database.
The noise reduction module carries out noise suppression on user voice acquired in real time or other stored voice to obtain voice information after noise reduction; the noise suppression module adopts at least one of a spectrum elimination method and/or a learning similarity method and/or a noise reduction automatic encoder to perform noise suppression;
the conversion module carries out Laplace transformation on the sound information to obtain Laplace frequency spectrum information;
the feature extraction module performs two-dimensional Fourier transform on the obtained Laplace spectrum information to obtain wave number spectrum features of Laplace transform sound information data;
the noise suppression module cuts out at least 5 time slice filter factors according to the wave number spectrum of the Laplace transform sound information data to form a filter group; and performing two-dimensional Fourier transform on the time-sliced information data to obtain a filtering information data wave number spectrum, performing environmental noise suppression on the whole recorded sound data wave number spectrum by using the filtering information data wave number spectrum, and transmitting the sound data subjected to the environmental noise suppression to a scoring server.
The recognition module performs voice recognition on an input voice to be input by referring to recognition target language information having tag information and pronunciation information of each recognition target language included in recognition target words registered in advance in a voice recognition dictionary, using a voice recognition engine corresponding to english which is a language set in advance as a recognition target;
the cloud database is registered with a pronunciation information conversion rule indicating correspondence between pronunciation information of words in a plurality of languages, and the pronunciation information conversion is based on the pronunciation information conversion rule of the cloud database to convert the pronunciation information of the words between the languages.
The control module controls the reading information conversion module to convert the reading information of the other language into the reading information of the set language when a word of the other language, which is a language different from the set language, is included in a recognition target vocabulary referred to by the voice recognition module, and the recognition module performs voice recognition by referring to the reading information of the set language after conversion of the input voice being input and the recognition target language information of the recognition target vocabulary registered in advance in the voice recognition dictionary.
The semantic relatedness features include: calculating semantic similarity scores of each word in the recognition result and each word in the standard answer; calculating semantic similarity scores of each word in the recognition result and each sentence in the standard answers; calculating the maximum value or the average value of the semantic similarity score in each word in the recognition result and each sentence in the standard answer as the similarity score between the word and the sentence; a similarity score between the user answer and the standard answer is calculated.
The effective feature extraction module is used for extracting the grammar structure similarity feature, and comprises: respectively establishing a grammar sequence vector for each sentence of the recognition result; respectively solving the grammatical structure similarity score of each sentence in the recognition result and each sentence in the standard answers, and taking the maximum value of the grammatical structure similarity score of each sentence in the recognition result as the grammatical structure similarity score of the sentence; and calculating the grammar structure similarity characteristic between the user answer and the standard answer by weighted average of the grammar structure similarity score of each sentence in the recognition result.
The recognition module adopts a decoding system based on large-scale continuous speech recognition, the acoustic model adopts a hidden Markov model, the language model adopts a language model based on a metagrammar, a multi-pass decoding technology is adopted during decoding, and the multi-pass decoding comprises direct decoding, unsupervised self-adaption based on maximum linear likelihood regression and secondary decoding; and the effective characteristic scoring module is used for scoring and training the comprehensive characteristics of the stream to obtain a scoring model, and scoring the recognition result according to the scoring model.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical scope of the present invention and the equivalent alternatives or modifications according to the technical solution and the inventive concept of the present invention within the technical scope of the present invention.
Claims (7)
1. An automatic English spoken language scoring system based on voice recognition is characterized by comprising a client, a scoring server and a cloud database which are connected with one another;
the client comprises a voice extraction module, a voice processor and a communication module;
the voice extraction module records the spoken voice of the user through the recording device, converts the spoken voice into a digital signal and outputs the digital signal to the voice recognizer;
the sound processor includes: the device comprises a noise reduction module, a conversion module, a feature extraction module, a noise suppression module, an identification module and a control module;
the communication module collects signals of the voice extraction module and the voice processor and transmits the signals to the scoring server;
the scoring server comprises a scoring device and a statistical uploading module; the scoring device comprises an identification module and a scoring module;
the recognition module comprises an acoustic module, a language module and a specific recognition module, the acoustic module extracts acoustic features of the user answering audio to obtain an acoustic model, the language module obtains a language model according to the question information and the training text, and the recognition module decodes the user answering audio through the acoustic model and the language model to obtain a recognition result;
the scoring module comprises an effective feature extraction module and an effective feature scoring module, the effective feature extraction module is used for extracting comprehensive speech flow features in the recognition result, the comprehensive speech flow features comprise features in pronunciation accuracy direction, features in fluency direction and features in text semantic similarity direction in oral language assessment, and the features in the text semantic similarity direction comprise semantic relevance features and grammatical structure similarity features; the effective feature extraction module is used for extracting the semantic relatedness feature;
the statistical uploading module collects the effective characteristics in the scoring module and sorts the effective characteristics according to the similarity sequence of the effective characteristics, and the statistical uploading module uploads the sorted effective characteristics to the cloud database.
2. The spoken English reading automatic scoring system based on voice recognition according to claim 1, wherein the noise reduction module performs noise suppression on the user voice acquired in real time or other stored voices to obtain noise-reduced voice information; the noise suppression module adopts at least one of a spectrum elimination method and/or a learning similarity method and/or a noise reduction automatic encoder to perform noise suppression;
the conversion module carries out Laplace transformation on the sound information to obtain Laplace frequency spectrum information;
the feature extraction module performs two-dimensional Fourier transform on the obtained Laplace spectrum information to obtain wave number spectrum features of Laplace transform sound information data;
the noise suppression module cuts out at least 5 time slice filter factors according to the wave number spectrum of the Laplace transform sound information data to form a filter group; and performing two-dimensional Fourier transform on the time-sliced information data to obtain a filtering information data wave number spectrum, performing environmental noise suppression on the whole recorded sound data wave number spectrum by using the filtering information data wave number spectrum, and transmitting the sound data subjected to the environmental noise suppression to a scoring server.
3. The automatic scoring system for spoken english reading based on voice recognition according to claim 1, wherein the recognition module performs voice recognition on the input voice being input by using a voice recognition engine corresponding to english, which is a language set in advance as a recognition target, with reference to recognition target language information including label information and reading information of each recognition target language included in recognition target words registered in advance in a voice recognition dictionary;
the cloud database is registered with a pronunciation information conversion rule indicating correspondence between pronunciation information of words in a plurality of languages, and the pronunciation information conversion is based on the pronunciation information conversion rule of the cloud database to convert the pronunciation information of the words between the languages.
4. The automatic scoring system for spoken english reading according to claim 3, wherein the control module controls the reading information conversion module to convert the reading information of the other language into the reading information of the set language when a word of the other language, which is a language different from the set language, is included in the recognition target vocabulary referred to by the voice recognition module, and the recognition module performs voice recognition by referring to the reading information of the set language after conversion of the input voice being input and the recognition target language information of the recognition target vocabulary registered in advance in the voice recognition dictionary.
5. The spoken english reading automatic scoring system according to claim 1, characterized in that the semantic relatedness feature comprises: calculating semantic similarity scores of each word in the recognition result and each word in the standard answer; calculating semantic similarity scores of each word in the recognition result and each sentence in the standard answers; calculating the maximum value or the average value of the semantic similarity score in each word in the recognition result and each sentence in the standard answer as the similarity score between the word and the sentence; a similarity score between the user answer and the standard answer is calculated.
6. The spoken english reading automatic scoring system according to claim 1, wherein the valid feature extracting module is configured to extract the grammar structure similarity feature, and includes: respectively establishing a grammar sequence vector for each sentence of the recognition result; respectively solving the grammatical structure similarity score of each sentence in the recognition result and each sentence in the standard answers, and taking the maximum value of the grammatical structure similarity score of each sentence in the recognition result as the grammatical structure similarity score of the sentence; and calculating the grammar structure similarity characteristic between the user answer and the standard answer by weighted average of the grammar structure similarity score of each sentence in the recognition result.
7. The automatic scoring system for spoken English reading based on voice recognition according to claim 3, wherein the recognition module employs a decoding system based on large-scale continuous voice recognition, the acoustic model employs a hidden Markov model, the language model employs a language model based on a metagrammar, and a multi-pass decoding technique is employed in decoding, the multi-pass decoding includes direct decoding, unsupervised adaptive and secondary decoding based on maximum linear likelihood regression; and the effective characteristic scoring module is used for scoring and training the comprehensive characteristics of the stream to obtain a scoring model, and scoring the recognition result according to the scoring model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010423027.7A CN111599234A (en) | 2020-05-19 | 2020-05-19 | Automatic English spoken language scoring system based on voice recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010423027.7A CN111599234A (en) | 2020-05-19 | 2020-05-19 | Automatic English spoken language scoring system based on voice recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111599234A true CN111599234A (en) | 2020-08-28 |
Family
ID=72183586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010423027.7A Pending CN111599234A (en) | 2020-05-19 | 2020-05-19 | Automatic English spoken language scoring system based on voice recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111599234A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112767932A (en) * | 2020-12-11 | 2021-05-07 | 北京百家科技集团有限公司 | Voice evaluation system, method, device, equipment and computer readable storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0095069A1 (en) * | 1982-05-25 | 1983-11-30 | Texas Instruments Incorporated | Electronic learning aid with sound effects mode |
| CN101739870A (en) * | 2009-12-03 | 2010-06-16 | 深圳先进技术研究院 | Interactive language learning system and method |
| WO2011042808A1 (en) * | 2009-10-09 | 2011-04-14 | Toyota Jidosha Kabushiki Kaisha | Signal separation system and signal separation method |
| CN103038816A (en) * | 2010-10-01 | 2013-04-10 | 三菱电机株式会社 | Speech recognition device |
| CN103151042A (en) * | 2013-01-23 | 2013-06-12 | 中国科学院深圳先进技术研究院 | Full-automatic oral language evaluating management and scoring system and scoring method thereof |
| CN103928023A (en) * | 2014-04-29 | 2014-07-16 | 广东外语外贸大学 | Voice scoring method and system |
| CN106353816A (en) * | 2016-08-09 | 2017-01-25 | 中国石油天然气集团公司 | Seismic acquisition footprint noise suppression method and system |
| CN107481732A (en) * | 2017-08-31 | 2017-12-15 | 广东小天才科技有限公司 | Noise-reduction method, device and terminal device in a kind of spoken test and appraisal |
| CN109493847A (en) * | 2018-12-14 | 2019-03-19 | 广州玛网络科技有限公司 | Sound recognition system and voice recognition device |
| CN109727609A (en) * | 2019-01-11 | 2019-05-07 | 龙马智芯（珠海横琴）科技有限公司 | Spoken language pronunciation appraisal procedure and device, computer readable storage medium |
| CN110379440A (en) * | 2019-07-19 | 2019-10-25 | 宁波奥克斯电气股份有限公司 | Voice de-noising method, device, voice air conditioner and computer readable storage medium |
| CN110600052A (en) * | 2019-08-19 | 2019-12-20 | 天闻数媒科技（北京）有限公司 | Voice evaluation method and device |
| CN110867193A (en) * | 2019-11-26 | 2020-03-06 | 广东外语外贸大学 | Paragraph English spoken language scoring method and system |
-
2020
- 2020-05-19 CN CN202010423027.7A patent/CN111599234A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0095069A1 (en) * | 1982-05-25 | 1983-11-30 | Texas Instruments Incorporated | Electronic learning aid with sound effects mode |
| WO2011042808A1 (en) * | 2009-10-09 | 2011-04-14 | Toyota Jidosha Kabushiki Kaisha | Signal separation system and signal separation method |
| CN101739870A (en) * | 2009-12-03 | 2010-06-16 | 深圳先进技术研究院 | Interactive language learning system and method |
| CN103038816A (en) * | 2010-10-01 | 2013-04-10 | 三菱电机株式会社 | Speech recognition device |
| CN103151042A (en) * | 2013-01-23 | 2013-06-12 | 中国科学院深圳先进技术研究院 | Full-automatic oral language evaluating management and scoring system and scoring method thereof |
| CN103928023A (en) * | 2014-04-29 | 2014-07-16 | 广东外语外贸大学 | Voice scoring method and system |
| CN106353816A (en) * | 2016-08-09 | 2017-01-25 | 中国石油天然气集团公司 | Seismic acquisition footprint noise suppression method and system |
| CN107481732A (en) * | 2017-08-31 | 2017-12-15 | 广东小天才科技有限公司 | Noise-reduction method, device and terminal device in a kind of spoken test and appraisal |
| CN109493847A (en) * | 2018-12-14 | 2019-03-19 | 广州玛网络科技有限公司 | Sound recognition system and voice recognition device |
| CN109727609A (en) * | 2019-01-11 | 2019-05-07 | 龙马智芯（珠海横琴）科技有限公司 | Spoken language pronunciation appraisal procedure and device, computer readable storage medium |
| CN110379440A (en) * | 2019-07-19 | 2019-10-25 | 宁波奥克斯电气股份有限公司 | Voice de-noising method, device, voice air conditioner and computer readable storage medium |
| CN110600052A (en) * | 2019-08-19 | 2019-12-20 | 天闻数媒科技（北京）有限公司 | Voice evaluation method and device |
| CN110867193A (en) * | 2019-11-26 | 2020-03-06 | 广东外语外贸大学 | Paragraph English spoken language scoring method and system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112767932A (en) * | 2020-12-11 | 2021-05-07 | 北京百家科技集团有限公司 | Voice evaluation system, method, device, equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105845134B (en) | Spoken language evaluation method and system for freely reading question types | |
| CN109272992B (en) | Spoken language evaluation method and device for generating spoken language evaluation model | |
| Donnelly et al. | Words matter: automatic detection of teacher questions in live classroom discourse using linguistics, acoustics, and context | |
| CN103151042B (en) | Full-automatic oral evaluation management and points-scoring system and methods of marking thereof | |
| CN101751919B (en) | Spoken Chinese stress automatic detection method | |
| US8160866B2 (en) | Speech recognition method for both english and chinese | |
| CN103761975A (en) | Method and device for oral evaluation | |
| CN112581964B (en) | Multi-domain oriented intelligent voice interaction method | |
| CN111681143A (en) | Multi-dimensional analysis method, device, equipment and storage medium based on classroom voice | |
| CN111785275A (en) | Voice recognition method and device | |
| Mohammed et al. | Quranic verses verification using speech recognition techniques | |
| Ahamad et al. | Accentdb: A database of non-native english accents to assist neural speech recognition | |
| Ahsiah et al. | Tajweed checking system to support recitation | |
| CN109300339A (en) | A kind of exercising method and system of Oral English Practice | |
| Maqsood et al. | An efficientmis pronunciation detection system using discriminative acoustic phonetic features for arabic consonants. | |
| US8145483B2 (en) | Speech recognition method for all languages without using samples | |
| CN110598041A (en) | FlACS real-time analysis method and device | |
| CN111599234A (en) | Automatic English spoken language scoring system based on voice recognition | |
| Wang et al. | Design and implementation of an English pronunciation scoring system for pupils based on DNN-HMM | |
| CN115497455B (en) | Intelligent evaluating method, system and device for oral English examination voice | |
| CN112767961B (en) | Accent correction method based on cloud computing | |
| CN117409765A (en) | Qingdao dialect voice recognition model based on Transformer | |
| CN115376547A (en) | Pronunciation evaluation method and device, computer equipment and storage medium | |
| CN114373447A (en) | Chinese-English translation question scoring method and system | |
| CN114420159A (en) | Audio evaluation method and device and non-transient storage medium |
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: 20200828 |