Acoustic model building method and device for multi-language voice identification

An acoustic model and speech recognition technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of speech data differentiation and low recognition accuracy, and achieve the effect of improving accuracy

Active Publication Date: 2013-11-20
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF4 Cites 67 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the existing technology, neither the mixed Gaussian model algorithm nor the deep neural network algorithm can effectively distinguish the speech data of different languages, so the final recognition accuracy is not high

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
  • Acoustic model building method and device for multi-language voice identification
  • Acoustic model building method and device for multi-language voice identification
  • Acoustic model building method and device for multi-language voice identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and through specific embodiments.

[0025] Figure 1 to Figure 8 A first embodiment of the invention is shown.

[0026] figure 1 It is a flow chart of the acoustic model building method for multilingual speech recognition provided by the first embodiment of the present invention. see figure 1 , the method for establishing an acoustic model for multilingual speech recognition includes: step S110, using all speech feature data to train a deep neural network; step S120, using speech feature data of different languages ​​to train the multiple output layers corresponding to different languages ; and step S130, merging the multiple output layers corresponding to different languages ​​into a total output layer.

[0027] In step S110, a deep neural network is trained using all speech feature data.

[0028] figure 2 The structure of the deep neura...

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 discloses an acoustic model building method and device for multi-language voice identification. The method comprises the following steps: training a deep neural network by using all voice feature data; training a plurality of output layers which correspond to different languages respectively by using the voice feature data of different languages; combining the output layers which correspond to different languages into a main output layer. According to the acoustic model building method for multi-language voice identification disclosed by the invention, the acoustic model of multi-language voice identification is built by using a depth neural network, so that the identification of multi-language voice is realized, and the accuracy of the voice identification is increased.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to a method and device for establishing an acoustic model for multilingual speech recognition. Background technique [0002] The existing multilingual speech recognition acoustic model building algorithms can be roughly divided into two types, namely the Gaussian mixture model (GMM) algorithm and the deep neural network (DNN) algorithm. If the acoustic model is established using the Gaussian mixture model algorithm, multiple Gaussian probability density functions need to be established to obtain a Gaussian probability density function set. and obtained, the speech features are judged according to the acoustic score of each speech feature. If a deep neural network algorithm is used to build an acoustic model, the deep neural network needs to be trained with sample data including speech features, and the speech features are judged according to the output probability of the...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/06G10L15/18
Inventor 苏丹尹钊
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) 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