Speech recognition method and device, electronic equipment and storage medium
A speech recognition and speech technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of large amount of calculation and low recognition rate of the system, and achieve the effect of ensuring reliability and accuracy and reducing the amount of calculation
Pending Publication Date: 2022-03-11
合肥讯飞数码科技有限公司
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] The present invention provides a voice recognition method, device, electronic equipment and storage medium, which are used to solve the large amount of system computation and low recognition rate caused by over-reliance on language recognition and unused mixed language semantics in multilingual recognition in the prior art Defects
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0045] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention , but not all examples. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.
[0046] At present, when performing intelligent speech recognition on mixed-language speech, it is necessary to rely on the results of the language recognizer to recognize the mixed-language speech, but there may be multiple languages in the mixed-language speech, and multiple language recognizers are required to recognize the mixed Recognition of voices in different languages will lead to...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides a voice recognition method and device, electronic equipment and a storage medium. The method comprises the steps of determining to-be-recognized voice; inputting the voice to be recognized into a mixed voice recognition model to obtain a recognition result output by the mixed voice recognition model; the mixed speech recognition model is obtained by training sample mixed language speech and mixed language text labels on the basis of a mixed semantic model used for extracting semantic features of mixed languages, and the mixed semantic model is obtained by training sample mixed language texts. According to the method and device, the electronic equipment and the storage medium provided by the invention, the voice containing the mixed language can be recognized through the mixed voice recognition model trained and completed on the basis of the mixed semantic model for extracting the semantic features of the mixed language, and the purpose of recognizing the voice containing the mixed language without a language recognizer is achieved. And the mixed language voice is accurately recognized while the system operation amount is reduced.
Description
technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a voice recognition method, device, electronic equipment and storage medium. Background technique [0002] At present, based on mixed multilingual speech recognition, there are mainly the following methods: [0003] 1) The language category of the input data is judged by the language recognizer first, and then the audio features are sent to the speech recognizer of the corresponding language to obtain the speech recognition result. [0004] 2) Divide words into complete word boundary fragments first according to the word segment, language recognition judges the language category of each fragment, and then reintegrate adjacent fragments of the same language and send them to the corresponding language recognizer for recognition, The final result is obtained by splicing the recognition results. [0005] 3) Add non-main language words to the main language sp...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/00G10L15/06G10L15/22G10L15/26
CPCG10L15/005G10L15/063G10L15/22G10L15/26G10L2015/0633
Inventor 胡鹏方磊何雪周振昆方四安
Owner 合肥讯飞数码科技有限公司