End-to-end model training method and device, computer equipment and storage medium
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A model training and model technology, applied in speech analysis, speech recognition, instruments, etc., can solve problems such as dependence on training data, robustness, and low flexibility
Pending Publication Date: 2022-08-09
PING AN TECH (SHENZHEN) CO LTD
View PDF0 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0005] The purpose of the embodiment of the present application is to propose an end-to-end model training method, device, computer equipment and storage medium applied to speech recog
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 more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0044] continue to refer to figure 2 , which shows an implementation flow chart of the end-to-end model training method applied to speech recognition provided by Embodiment 1 of the present application. For convenience of description, only the parts related to the present application are shown.
[0045] The above-mentioned end-to-end model training method applied to speech recognition includes: step S201 , step S202 , step S203 , step S204 and step S205 .
[0046] In step S201, model training data is acquired, wherein the model training data includes an audio training set, and the audio training set includes training audio data and audio annotation text.
[0047] In this embodiment of the present application, the end-to-end model applied to speech recognition may use the standard CTC-Attention model, such as image 3 shown.
[0048] In the embodiment of the present application, the training audio data refers to the audio used for training the acoustic model in the end-to-en...
Embodiment 2
[0112] further reference Figure 9 , as a response to the above figure 2 The implementation of the shown method, the present application provides an embodiment of an end-to-end model training device applied to speech recognition, the device embodiment is the same as figure 2 Corresponding to the method embodiments shown, the apparatus can be specifically applied to various electronic devices.
[0113] like Figure 9 As shown, the end-to-end model training apparatus 200 applied to speech recognition in this embodiment includes: a training data acquisition module 210 , an audio recognition module 220 , a text fusion module 230 , a language translation module 240 and a joint training module 250 . in:
[0114] A training data acquisition module 210, configured to acquire model training data, wherein the model training data includes an audio training set, and the audio training set includes training audio data and audio annotation text;
[0115] The audio recognition module 2...
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 embodiment of the invention belongs to the technical field of voice recognition in artificial intelligence, and relates to an end-to-end model training method and device applied to voice recognition, computer equipment and a storage medium. According to the invention, the output of the acoustic model is used as the extended text of the audio training data, and the extended text and the audio annotation text are used as the language model input together to train the speech recognition model, so that the defect that the annotation text content in a traditional speech training set is too limited is effectively overcome; the language model of the speech recognition model learns richer and more comprehensive information, so that the recognition accuracy of the speech recognition model is effectively improved, meanwhile, the coupling degree of acoustic information and language information in the end-to-end model is reduced to a certain extent, and the recognition accuracy of the whole model in different scenes is improved. Particularly, the robustness of recognizing voices in different fields is improved, the problem that the accuracy is greatly reduced when an application scene is replaced is avoided, and the flexibility of the model in actual use and deployment is also improved.
Description
technical field [0001] The present application relates to the technical field of speech processing in artificial intelligence, and in particular, to an end-to-end model training method, apparatus, computer equipment and storage medium applied to speech recognition. Background technique [0002] With the rapid development of the field of artificial intelligence, speech recognition has become an increasingly important application technology in the field of artificial intelligence. Speech recognition technology has developed from the earlier hmm-gmm model and acoustic model to the currently commonly used end-to-end model. [0003] At present, the latest end-to-end model is an end-to-end model based on CTC-Attention. The end-to-end model based on CTC-Attention adopts the method of multi-task learning to jointly optimize speech and text, which combines the advantages of both. , compared with the traditional model, a great improvement has been achieved. [0004] However, the app...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more