CN113035197A - Speech recognition system based on artificial intelligence algorithm - Google Patents

Speech recognition system based on artificial intelligence algorithm Download PDF

Info

Publication number
CN113035197A
CN113035197A CN202110268134.1A CN202110268134A CN113035197A CN 113035197 A CN113035197 A CN 113035197A CN 202110268134 A CN202110268134 A CN 202110268134A CN 113035197 A CN113035197 A CN 113035197A
Authority
CN
China
Prior art keywords
module
speech recognition
codebook
artificial intelligence
user
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
Application number
CN202110268134.1A
Other languages
Chinese (zh)
Inventor
杜金林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Translated By Mdt Infotech Ltd Shanghai
Original Assignee
Translated By Mdt Infotech Ltd Shanghai
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Translated By Mdt Infotech Ltd Shanghai filed Critical Translated By Mdt Infotech Ltd Shanghai
Priority to CN202110268134.1A priority Critical patent/CN113035197A/en
Publication of CN113035197A publication Critical patent/CN113035197A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • 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
    • G10L15/08—Speech classification or search
    • G10L15/18—Speech classification or search using natural language modelling
    • G10L15/183—Speech classification or search using natural language modelling using context dependencies, e.g. language models
    • G10L15/19—Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
    • G10L15/197—Probabilistic grammars, e.g. word n-grams
    • 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
    • G10L15/24—Speech recognition using non-acoustical features
    • G10L15/25—Speech recognition using non-acoustical features using position of the lips, movement of the lips or face analysis
    • 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
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/223—Execution procedure of a spoken command

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Probability & Statistics with Applications (AREA)
  • Artificial Intelligence (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

本发明提供一种基于人工智能算法的语音识别系统。所述基于人工智能算法的语音识别系统包括用户界面，所述用户界面用于显示内容；语音接收模块，所述语音接收模块用于接收语音信号；语音识别模块，所述语音识别模块用于将所述语音信号进行识别；对比模块，所述对比模块用于检测解码结果的准确率；摄像模块，所述摄像模块用于提供用户的图像信号；意图判断模块，所述意图判断模块针对所述摄像模块所拍摄的图像信号和所识别出的发声，判断所述用户有无操作所述被控制装置的意图。本发明提供的基于人工智能算法的语音识别系统具有识别准确率高、可对用户操作意识进行判断的优点。The invention provides a speech recognition system based on artificial intelligence algorithm. The artificial intelligence algorithm-based voice recognition system includes a user interface, which is used for displaying content; a voice receiving module, which is used for receiving voice signals; a voice recognition module, which is used for The voice signal is recognized; the comparison module is used to detect the accuracy of the decoding result; the camera module is used to provide the image signal of the user; the intention judgment module is used for the The image signal captured by the camera module and the recognized sound are used to determine whether the user intends to operate the controlled device. The speech recognition system based on the artificial intelligence algorithm provided by the present invention has the advantages of high recognition accuracy and can judge the user's operation awareness.

Description

Speech recognition system based on artificial intelligence algorithm
Technical Field
The invention relates to the technical field of voice recognition, in particular to a voice recognition system based on an artificial intelligence algorithm.
Background
With the progress of data processing technology and the rapid spread of mobile internet, computer technology is widely applied to various fields of society, and with the progress of data processing technology, mass data is generated. Among them, voice data is receiving more and more attention. Speech recognition is a cross discipline. Over the last two decades. Speech recognition technology has made significant progress, starting to move from the laboratory to the market. It is expected that voice recognition technology will enter various fields such as industry, home appliances, communications, automotive electronics, medical care, home services, consumer electronics, etc. within the next 10 years. The application of speech recognition dictation machines in some fields is rated by the U.S. news community as one of ten major computer developments in 1997. Many experts consider the speech recognition technology to be one of ten important branch development technologies in the information technology field from 2000 to 2010. The fields to which speech recognition technology relates include: signal processing, pattern recognition, probability and information theory, sound and hearing mechanisms, artificial intelligence, and the like. Speech recognition is technically complex but more widely used than speech synthesis. The greatest advantage of speech recognition is that it makes the human-machine user interface more natural and easy to use.
With the rapid development of microelectronic technology and communication technology, embedded communication devices such as mobile phones and the like almost become articles essential for people to work and live, and the requirements of people on the functions of the embedded communication devices are higher and higher, so that the application of voice technology to the devices becomes a hotspot of research, and the accuracy rate of the existing voice functions is not high; moreover, the voice recognition is always touched by mistake, and the intention of the user to use the voice recognition cannot be judged.
Therefore, there is a need to provide a new speech recognition system based on artificial intelligence algorithm to solve the above technical problems.
Disclosure of Invention
The invention solves the technical problem of providing the voice recognition system based on the artificial intelligence algorithm, which has high recognition accuracy and can judge the operation consciousness of the user.
In order to solve the above technical problems, the speech recognition system based on artificial intelligence algorithm provided by the invention comprises:
a user interface for displaying content;
the voice receiving module is used for receiving a voice signal;
a speech recognition module for recognizing the speech signal, the speech recognition module comprising:
the device comprises a signal conversion module, a feature extraction module, an encoding module, a codebook module and an operation decoding module;
the signal conversion module is used for converting the voice signal into a digital signal;
the characteristic extraction module is used for performing framing processing on the digital signals, extracting characteristic parameters of each frame of the digital signals and obtaining a characteristic vector sequence;
the coding module is used for converting the proper characteristic sequence into a characteristic code word sequence;
the code book module stores the probability value of the code word in the code book corresponding to each code word;
the decoding operation module is used for carrying out decoding operation on the characteristic code word sequence to obtain an identification result, and directly searching a cipher word with the maximum matching probability from the cipher book module for each code word in the characteristic code word sequence in the operation to obtain a decoding result;
the comparison module is used for detecting the accuracy of the decoding result;
the camera module is used for providing an image signal of a user;
and the intention judging module is used for judging whether the user has the intention of operating the controlled device or not according to the image signal shot by the camera module and the recognized voice.
Preferably, the codebook is a gaussian codebook.
Preferably, the encoding module converts the feature vector sequence into a feature codeword sequence according to the following steps:
s1: dividing the feature vector sequence into a plurality of subspaces, wherein each subspace corresponds to a codebook;
s2: calculating distance measurement between all the feature vectors in each subspace and each code word in a corresponding codebook, and taking the code word with the minimum distance measurement with the feature vector as the code word corresponding to the feature vector in the feature code word sequence;
s3: and combining the code words corresponding to all vectors in each subspace of the characteristic vector sequence according to the original vector sequence to obtain the corresponding characteristic code word sequence.
Preferably, the codebook module is generated by the following steps:
l1: calculating a mean value and a variance vector corresponding to each code word in the Gaussian codebook;
l2: calculating the logarithm probability value of each code word in the characteristic codebook and the logarithm probability value of each code word in the Gaussian codebook by using the mean value and the variance vector;
l3: and storing the probability values of all code words in the characteristic codebook matched with all code words in the Gaussian codebook to obtain the codebook module.
Preferably, the comparison module stores a plurality of commonly used specific sentence texts, and compares the result recognized by the speech recognition module with the specific sentence texts to judge the recognition accuracy of the speech recognition module.
Preferably, the camera module highlights the eye focus and lip movement of the user.
Preferably, the intention determination module determines, when it is determined that there is an operation intention, a degree of reliability indicating a degree of intention of the operation.
Preferably, the control device further includes a control state changing module that changes the control state of the controlled device to a direction that is less noticeable to the user than when the intention determination unit determines that there is no operation intention.
Preferably, the control state changing module changes the control state of the controlled device to a direction that is less noticeable to the user than when the reliability determined by the intention determining module is low.
Preferably, the control state changing module controls the controlled device to notify the user of the recognition failure when the recognition of the voice uttered by the user fails, and changes the notified state to a direction that is not recognized by the user when the reliability of the operation intention with respect to the utterance is low as compared with when the reliability is high.
Compared with the related art, the voice recognition system based on the artificial intelligence algorithm has the following beneficial effects:
the invention provides a voice recognition system based on artificial intelligence algorithm, which adds the steps of dynamically merging and splitting subsets according to the vector quantity in the subsets and the total distance measurement of the vectors in the subsets in the process of clustering the voice feature vector set to obtain a codebook, reduces the distance measurement sum of the vectors in the clustered set and the corresponding code words, improves the precision of the clustering algorithm, ensures the recognition performance of the voice system, and greatly reduces the storage capacity of the system; in addition, when it is determined that the user has no operation intention, the control state of the controlled device is changed to a direction that is not recognized by the user, as compared with the case where it is determined that the user has an operation intention, thereby increasing the comfort of use for the user.
Detailed Description
The present invention will be further described with reference to the following embodiments.
An artificial intelligence algorithm based speech recognition system comprising:
a user interface for displaying content;
the voice receiving module is used for receiving a voice signal;
a speech recognition module for recognizing the speech signal, the speech recognition module comprising:
the device comprises a signal conversion module, a feature extraction module, an encoding module, a codebook module and an operation decoding module;
the signal conversion module is used for converting the voice signal into a digital signal;
the characteristic extraction module is used for performing framing processing on the digital signals, extracting characteristic parameters of each frame of the digital signals and obtaining a characteristic vector sequence;
the coding module is used for converting the proper characteristic sequence into a characteristic code word sequence;
the code book module stores the probability value of the code word in the code book corresponding to each code word;
the decoding operation module is used for carrying out decoding operation on the characteristic code word sequence to obtain an identification result, and directly searching a cipher word with the maximum matching probability from the cipher book module for each code word in the characteristic code word sequence in the operation to obtain a decoding result;
the comparison module is used for detecting the accuracy of the decoding result;
the camera module is used for providing an image signal of a user;
and the intention judging module is used for judging whether the user has the intention of operating the controlled device or not according to the image signal shot by the camera module and the recognized voice.
The codebook is a gaussian codebook.
The encoding module converts the feature vector sequence into a feature codeword sequence according to the following steps:
s1: dividing the feature vector sequence into a plurality of subspaces, wherein each subspace corresponds to a codebook;
s2: calculating distance measurement between all the feature vectors in each subspace and each code word in a corresponding codebook, and taking the code word with the minimum distance measurement with the feature vector as the code word corresponding to the feature vector in the feature code word sequence;
s3: and combining the code words corresponding to all vectors in each subspace of the characteristic vector sequence according to the original vector sequence to obtain the corresponding characteristic code word sequence.
The codebook module is generated by the following steps:
l1: calculating a mean value and a variance vector corresponding to each code word in the Gaussian codebook;
l2: calculating the logarithm probability value of each code word in the characteristic codebook and the logarithm probability value of each code word in the Gaussian codebook by using the mean value and the variance vector;
l3: and storing the probability values of all code words in the characteristic codebook matched with all code words in the Gaussian codebook to obtain the codebook module.
The comparison module stores a plurality of common specific sentence texts, compares the result recognized by the voice recognition module with the specific sentence texts, and judges the recognition accuracy of the voice recognition module.
The camera module highlights the eye focus and lip movement of the user.
The intention judgment module judges the reliability indicating the degree of the intention of the operation when judging that the operation intention exists.
The control device further includes a control state changing module that changes the control state of the controlled device to a direction that is less noticeable to the user than when the intention determination unit determines that there is no operation intention.
The control state changing module changes the control state of the controlled device to a direction that is not recognized by the user when the reliability determined by the intention determining module is low, as compared with a case where the reliability is high.
The control state changing module controls the controlled device to notify the user of the recognition failure when the recognition of the voice uttered by the user fails, and changes a state of the notification to a direction that is not recognized by the user when the reliability of the operation intention of the uttering is low as compared with a case where the reliability is high.
Compared with the related art, the voice recognition system based on the artificial intelligence algorithm has the following beneficial effects:
the invention provides a voice recognition system based on artificial intelligence algorithm, which adds the steps of dynamically merging and splitting subsets according to the vector quantity in the subsets and the total distance measurement of the vectors in the subsets in the process of clustering the voice feature vector set to obtain a codebook, reduces the distance measurement sum of the vectors in the clustered set and the corresponding code words, improves the precision of the clustering algorithm, ensures the recognition performance of the voice system, and greatly reduces the storage capacity of the system; in addition, when it is determined that the user has no operation intention, the control state of the controlled device is changed to a direction that is not recognized by the user, as compared with the case where it is determined that the user has an operation intention, thereby increasing the comfort of use for the user.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by the present specification, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1.一种基于人工智能算法的语音识别系统，其特征在于，包括：1. a speech recognition system based on artificial intelligence algorithm, is characterized in that, comprises: 用户界面，所述用户界面用于显示内容；a user interface for displaying content; 语音接收模块，所述语音接收模块用于接收语音信号；a voice receiving module, the voice receiving module is used for receiving voice signals; 语音识别模块，所述语音识别模块用于将所述语音信号进行识别，所述语音识别模块包括：A speech recognition module, the speech recognition module is used for recognizing the speech signal, and the speech recognition module includes: 信号转换模块、特征提取模块、编码模块、密码本模块和运算解码模块；Signal conversion module, feature extraction module, encoding module, codebook module and operation decoding module; 所述信号转换模块用于将所述语音信号转换为数字信号；The signal conversion module is used for converting the voice signal into a digital signal; 所述特征提取模块用于对所述数字信号进行分帧处理，提取每一帧所述数字信号的特征参数，得到特征矢量序列；The feature extraction module is used to perform frame-by-frame processing on the digital signal, extract the feature parameters of the digital signal in each frame, and obtain a feature vector sequence; 编码模块用于将所述特征适量序列转换为特征码字序列；The encoding module is used to convert the feature quantity sequence into a feature codeword sequence; 密码本模块储存有每个码字对应的密码本中的密码字的概率值；The codebook module stores the probability value of the codeword in the codebook corresponding to each codeword; 解码运算模块用于对该特征码字序列进行解码运算得到识别结果，运算中对该特征码字序列中的各个码字，从密码本模块中直接查找与其具有最大匹配概率的密码字，得到解码结果；The decoding operation module is used to perform decoding operation on the characteristic codeword sequence to obtain the identification result. In the operation, each codeword in the characteristic codeword sequence is directly searched from the codebook module for the codeword with the maximum matching probability, and the decoding operation is obtained. result; 对比模块，所述对比模块用于检测解码结果的准确率；a comparison module, which is used to detect the accuracy of the decoding result; 摄像模块，所述摄像模块用于提供用户的图像信号；a camera module, the camera module is used to provide a user's image signal; 意图判断模块，所述意图判断模块针对所述摄像模块所拍摄的图像信号和所识别出的发声，判断所述用户有无操作所述被控制装置的意图。An intention judgment module, the intention judgment module judges whether the user has an intention to operate the controlled device according to the image signal captured by the camera module and the recognized sound. 2.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，所述密码本为高斯码本。2 . The artificial intelligence algorithm-based speech recognition system according to claim 1 , wherein the codebook is a Gaussian codebook. 3 . 3.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，所述编码模块根据将特征矢量序列转换为特征码字序列的步骤如下：3. the speech recognition system based on artificial intelligence algorithm according to claim 1, is characterized in that, described coding module is as follows according to the step that feature vector sequence is converted into feature code word sequence: S1：将所述特征矢量序列划分为多个子空间，每一个所述子空间对应于一个码本；S1: Divide the feature vector sequence into multiple subspaces, each of which corresponds to a codebook; S2：计算各子空间中所有特征矢量与相应码本中的各码字之间的距离度量，将与该特征矢量具有最小距离度量的码字作为所述特征码字序列中对应该特征矢量的码字；S2: Calculate the distance metric between all feature vectors in each subspace and each codeword in the corresponding codebook, and use the codeword with the smallest distance metric from the feature vector as the feature vector in the feature codeword sequence corresponding to the feature vector. numbers; S3：将所述特征矢量序列各个子空间所有矢量所对应的码字按原矢量顺序组合起来，即得到对应的特征码字序列。S3: Combine the codewords corresponding to all the vectors in each subspace of the feature vector sequence in the order of the original vectors, that is, to obtain the corresponding feature codeword sequence. 4.根据权利要求2所述的基于人工智能算法的语音识别系统，其特征在于，所述密码本模块通过以下步骤生成的：4. the speech recognition system based on artificial intelligence algorithm according to claim 2, is characterized in that, described codebook module is generated by the following steps: L1：计算高斯码本中各码字对应的均值和方差矢量；L1: Calculate the mean value and variance vector corresponding to each codeword in the Gaussian codebook; L2：利用上述均值和方差矢量，计算所述特征码本中各个码字与高斯码本中:各个码字相匹配的对数概率值；L2: Utilize above-mentioned mean value and variance vector, calculate each codeword in described characteristic codebook and in Gaussian codebook: the logarithmic probability value that each codeword matches; L3：将特征码本中的所有码字与高斯码本中的所有码字相匹配的概率值储存起来即可得到密码本模块。L3: The codebook module can be obtained by storing the probability values that all codewords in the feature codebook match all codewords in the Gaussian codebook. 5.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，所述对比模块中储存多条常用特定句子文本，所述对比模块将语音识别模块识别的结果与特定句子文本进行对比，判断语音识别模块识别的准确率。5. the speech recognition system based on artificial intelligence algorithm according to claim 1, is characterized in that, in described contrast module, store a plurality of commonly used specific sentence texts, described contrast module is by the result of speech recognition module recognition and specific sentence text Compare and judge the accuracy of speech recognition module recognition. 6.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，所述摄像模块重点标识用户的眼睛关注点和嘴唇移动。6 . The speech recognition system based on an artificial intelligence algorithm according to claim 1 , wherein the camera module focuses on identifying the user's eye focus and lip movement. 7 . 7.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，所述意图判断模块在判断为有操作意图的情况下，对表示该操作的意图达到何种程度的可靠度进行判断。7. The speech recognition system based on artificial intelligence algorithm according to claim 1, is characterized in that, under the situation that described intention judgment module is judged to have operation intention, to express the degree of reliability of the intention of this operation make a judgment. 8.根据权利要求1所述的基于人工智能算法的语音识别系统，其特征在于，还包括控制状态变更模块，所述控制状态变更模块在所述意图判断部中判断为无操作意图的情况下，与判断为有操作意图的情况相比，将所述被控制装置的控制的状态向不让所述用户意识到的方向进行变更。8. The artificial intelligence algorithm-based speech recognition system according to claim 1, further comprising a control state change module, wherein the control state change module is judged as no operation intention in the case of the intention judgment part , compared with the case where it is determined that there is an operation intention, the state of the control of the controlled device is changed in a direction that the user is not aware of. 9.根据权利要求8所述的基于人工智能算法的语音识别系统，其特征在于，所述控制状态变更模块在所述意图判断模块中判断出的可靠度较低的情况下，与可靠度较高的情况相比，将被控制装置的控制的状态向不让所述用户意识到的方向进行变更。9. The speech recognition system based on an artificial intelligence algorithm according to claim 8, wherein the control state change module, when the reliability judged in the intention judgment module is low, is compared with the reliability. Compared with the case of high, the state of the controlled device is changed in a direction that the user is not aware of. 10.根据权利要求8所述的基于人工智能算法的语音识别系统，其特征在于，所述控制状态变更模块在用户所发出的语音的识别失败的情况下，对所述被控制装置进行控制，使所述被控制装置向所述用户通知识别失败，并在关于发声的操作意图的可靠度较低的情况下，与可靠度较高的情况相比，将该通知的状态向不让所述用户意识到的方向进行变更。10. The speech recognition system based on an artificial intelligence algorithm according to claim 8, wherein the control state change module controls the controlled device when the recognition of the speech sent by the user fails, causing the controlled device to notify the user of a recognition failure, and in a case where the reliability of the operation intention to utter a sound is low, the state of the notification is changed to not allow the The user is aware of the direction to change.
CN202110268134.1A 2021-03-11 2021-03-11 Speech recognition system based on artificial intelligence algorithm Pending CN113035197A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110268134.1A CN113035197A (en) 2021-03-11 2021-03-11 Speech recognition system based on artificial intelligence algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110268134.1A CN113035197A (en) 2021-03-11 2021-03-11 Speech recognition system based on artificial intelligence algorithm

Publications (1)

Publication Number Publication Date
CN113035197A true CN113035197A (en) 2021-06-25

Family

ID=76469854

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110268134.1A Pending CN113035197A (en) 2021-03-11 2021-03-11 Speech recognition system based on artificial intelligence algorithm

Country Status (1)

Country Link
CN (1) CN113035197A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1455389A (en) * 2002-09-30 2003-11-12 中国科学院声学研究所 Voice identifying system and compression method of characteristic vector set for voice identifying system
CN106796786A (en) * 2014-09-30 2017-05-31 三菱电机株式会社 Speech recognition system
CN110266900A (en) * 2019-06-11 2019-09-20 平安科技（深圳）有限公司 Recognition methods, device and the customer service system that client is intended to
CN110300001A (en) * 2019-05-21 2019-10-01 深圳壹账通智能科技有限公司 Conference audio control method, system, equipment and computer readable storage medium
WO2021012495A1 (en) * 2019-07-23 2021-01-28 平安科技（深圳）有限公司 Method and device for verifying speech recognition result, computer apparatus, and medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1455389A (en) * 2002-09-30 2003-11-12 中国科学院声学研究所 Voice identifying system and compression method of characteristic vector set for voice identifying system
CN106796786A (en) * 2014-09-30 2017-05-31 三菱电机株式会社 Speech recognition system
CN110300001A (en) * 2019-05-21 2019-10-01 深圳壹账通智能科技有限公司 Conference audio control method, system, equipment and computer readable storage medium
CN110266900A (en) * 2019-06-11 2019-09-20 平安科技（深圳）有限公司 Recognition methods, device and the customer service system that client is intended to
WO2021012495A1 (en) * 2019-07-23 2021-01-28 平安科技（深圳）有限公司 Method and device for verifying speech recognition result, computer apparatus, and medium

Similar Documents

Publication Publication Date Title
KR102648306B1 (en) Speech recognition error correction method, related devices, and readable storage medium
CN111583909B (en) Voice recognition method, device, equipment and storage medium
CN102543071B (en) Voice recognition system and method used for mobile equipment
KR102371188B1 (en) Apparatus and method for speech recognition, and electronic device
CN110097870B (en) Speech processing method, device, device and storage medium
CN100403828C (en) A portable digital mobile communication device and its voice control method and system
CN110610707A (en) Speech keyword recognition method, device, electronic device and storage medium
WO2013102697A1 (en) Methods, apparatuses and computer program products for implementing automatic speech recognition and sentiment detection on a device
US10553206B2 (en) Voice keyword detection apparatus and voice keyword detection method
CN115104151A (en) Offline voice recognition method and device, electronic equipment and readable storage medium
CN112669842A (en) Man-machine conversation control method, device, computer equipment and storage medium
JPWO2020003413A1 (en) Information processing equipment, control methods, and programs
CN113823265A (en) A kind of speech recognition method, apparatus and computer equipment
CN116343765A (en) Method and system for automatic context-bound domain-specific speech recognition
CN119600997A (en) Mixed identification processing method, device, equipment and medium
CN118335084A (en) Speech recognition method and related device
CN116959418A (en) An audio processing method and device
CN113035197A (en) Speech recognition system based on artificial intelligence algorithm
JP2017116876A (en) Speaker recognition device, discriminant value generation method, and program
KR102392992B1 (en) User interfacing device and method for setting wake-up word activating speech recognition
CN116959442B (en) Chip for intelligent switch panel and method thereof
CN118982986A (en) A speech recognition method and related device, equipment and storage medium
CN117975950A (en) Speech recognition method, system, equipment and storage medium
CN116189666A (en) Speech recognition method, system and terminal
JP2015087540A (en) Voice recognition device, voice recognition system, and voice recognition program

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

Application publication date: 20210625