Model training method for unspecified person alone word, recognition system and recognition method

A non-specific person, recognition method technology, applied in the field of speech recognition, can solve the problems of low recognition rate and large amount of calculation, and achieve the effect of improving the recognition rate, reducing the amount of calculation, and reducing the amount of calculation.

Inactive Publication Date: 2008-01-30
大连三曦智能科技有限公司
View PDF0 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The purpose of the present invention is to provide a model training method, recognition system and recognition method for non-specific isolated words, improve the recognition rate, and effectively reduce the problems of large amount of calculation and low recognition rate in the existing training method

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
  • Model training method for unspecified person alone word, recognition system and recognition method
  • Model training method for unspecified person alone word, recognition system and recognition method
  • Model training method for unspecified person alone word, recognition system and recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] The present invention utilizes a plurality of training models obtained based on the DTW algorithm for non-specific isolated words to merge to obtain a central training model corresponding to the isolated word, so that each isolated word has only one central training model at last, so that in the model matching step, according to the input voice When the eigenvalues ​​of the data are matched with the training models of all isolated words, each isolated word only needs to be matched with one central training model, which greatly reduces the number of matching operations and reduces the amount of calculation during the matching operation.

[0051] As shown in Figure 3, the non-specific person isolated word model training method of the present invention specifically includes the following steps:

[0052] Step 31, using the DTW algorithm to obtain multiple training models of a non-specific isolated word;

[0053] Step 32, use the DTW algorithm to carry out two-two matching t...

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 a speaker independent isolated words model training method, an identifying system and an identifying method, wherein, the speaker independent isolated words model training method comprises getting a plurality of training modes procedures of the speaker independent isolated words using the dynamic time wrapping algorithm, and a centre training model getting procedure, combining a plurality of training models of the speaker independent isolated words into a centre training model of the speaker independent isolated words. The invention decreases the calculated amount when matching the operation by combining a plurality of training models of the speaker independent isolated words into a centre training mode. At the same time, the invention directly identifies the identifying result smaller than the rejection threshold as out-of-vocabulary words when in identification and process, and effectively improves the rejecting identification capability of out-of-vocabulary words. In addition, by limiting the searching area coverage and loosing the matching jumping-off, the invention increases the recognition rate and decreases the calculated amount.

Description

technical field [0001] The invention relates to speech recognition technology, especially the recognition of non-person-specific isolated words in the speech recognition technology. Background technique [0002] The existing non-person-specific isolated word recognition system is shown in Figure 1, including: [0003] The real-time voice receiving module is used to collect external sound signals in the form of single words; [0004] The voice feature value extraction module is used to extract representative feature value data from the collected original sound data; [0005] The identification module is used to correctly identify the eigenvalue data according to the model data, and output the identification result; [0006] The recognition result processing module is used to execute a defined processing program according to the recognition result. [0007] At present, the training method for the non-specific isolated word model includes backtracking the matching path based...

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): G10L15/06G10L15/08G10L15/28G10L15/00G10L15/12
Inventor 周金星
Owner 大连三曦智能科技有限公司
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