Voice recognition model acquisition method and device, electronic equipment and storage medium

A speech recognition model and acquisition method technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of poor speech recognition effect, no solution proposed, unsatisfactory speech separation effect and recognition accuracy, etc.

Pending Publication Date: 2022-01-04
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, the existing speaker recognition system and speech transcription system are not ideal for the situation where the speakers overlap, the speech separation effect and the recognition accuracy are not ideal, and the

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
  • Voice recognition model acquisition method and device, electronic equipment and storage medium
  • Voice recognition model acquisition method and device, electronic equipment and storage medium
  • Voice recognition model acquisition method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] According to an embodiment of the present disclosure, an embodiment of a method for acquiring a speech recognition model is provided. It should be noted that the steps shown in the flow chart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

[0032] figure 1 is a schematic flowchart of the steps of the method for acquiring a speech recognition model according to the first embodiment of the present disclosure, as shown in figure 1 As shown, the method includes the following steps:

[0033]Step S102, obtaining multiple sets of label data, wherein each set of data in the above multiple sets of label data includes: audio sample data of sample objects, and a set of sample objects obtained by performing feature vector extraction proc...

Embodiment 2

[0094] According to an embodiment of the present disclosure, an embodiment of an apparatus for implementing the above speech recognition method is also provided, Figure 4 is a schematic structural diagram of an acquisition device for a speech recognition model according to a second embodiment of the present disclosure, as shown in Figure 4 As shown, the acquisition device of the above-mentioned speech recognition model includes: an acquisition unit 40 and a training unit 42, wherein:

[0095] The acquisition unit 40 is configured to acquire multiple sets of label data, wherein each set of data in the above multiple sets of label data includes: audio sample data of sample objects, and a set of sample objects obtained by performing feature vector extraction processing on the above audio sample data, The above-mentioned audio sample data includes dialogue content of multiple above-mentioned sample objects;

[0096] The training unit 42 is configured to use multiple sets of lab...

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

No PUM Login to view more

Abstract

The invention provides a voice recognition model acquisition method and device, electronic equipment and a storage medium, and relates to the fields of natural voice understanding, voice technologies, intelligent customer service and voice transcription. According to the specific implementation scheme, obtaining multiple sets of label data, wherein each set of data in the multiple sets of label data comprises audio sample data of sample objects and a sample object set obtained by conducting feature vector extraction processing on the audio sample data, and the audio sample data comprise dialogue content of the multiple sample objects; and training a neural network model through machine learning by using the multiple groups of label data to obtain a speech recognition model. According to the invention, the technical problem of poor speech recognition effect of a speech recognition model in the prior art is solved.

Description

technical field [0001] The present disclosure relates to the field of artificial intelligence technology, in particular to the fields of natural speech understanding, speech technology, intelligent customer service, and speech transcription, and in particular to a method, device, electronic device, and storage medium for acquiring a speech recognition model. Background technique [0002] Most common speech recognition methods in the related art firstly separate the audio speakers, and then perform voice transcription on the separated audio to obtain the distinguished corresponding speaker's text. [0003] However, the existing speaker recognition system and speech transcription system are not ideal for the situation where the speakers overlap, the speech separation effect and the recognition accuracy are not ideal, and the number of speakers needs to be set in advance to determine the number of branches of the network. Speech recognition does not work well with varying numbe...

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
no application Login to view more
IPC IPC(8): G10L15/06G10L15/16G10L15/26G10L15/18G10L17/00G10L21/0272G10L21/0308
CPCG10L15/063G10L15/16G10L15/26G10L15/1822G10L17/00G10L21/0272G10L21/0308G10L2015/0631G10L2015/0635
Inventor 赵情恩
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap