Unlock instant, AI-driven research and patent intelligence for your innovation.

Text semi-relevant voiceprint recognition method and system

A voiceprint recognition and text technology, which is applied in the field of text-related voiceprint recognition, can solve the problems of low voiceprint recognition accuracy, and the voiceprint recognition method does not take into account the digital recognition effect, so as to improve the voiceprint recognition accuracy and ensure Safe, anti-recording effect

Active Publication Date: 2019-03-15
XIAMEN KUAISHANGTONG INFORMATION TECH CO LTD
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The above voiceprint recognition method does not take into account the recognition effect of each number, and the voiceprint recognition accuracy is low

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
  • Text semi-relevant voiceprint recognition method and system
  • Text semi-relevant voiceprint recognition method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0076] The embodiment of the present invention provides a text semi-correlated voiceprint recognition method, as shown in the attached figure 2 As shown, the method includes the following:

[0077] Step 1, in the registration process, obtain the voice of the registered user reading several different character strings, segment the voice according to the content, and perform feature extraction on the segmented voice segment; the segmentation is to divide the voice content Each number of each number is cut into a speech segment; in the embodiment of the present invention, preferably, the character string is a number string; it should be noted that the character string of the present invention is not limited to a number string, it can be a character string or a number , letters, special symbols, Chinese characters, etc.

[0078] Step 2, randomly generating a section of verification number string, said section of verification number string containing n numbers, and obtaining a se...

Embodiment 3

[0097] In addition, the embodiment of the present invention also provides a text semi-correlated voiceprint recognition system, as shown in the attached figure 2 As shown, the system includes a registration module 1 , a verification module 2 , a verification feature module 4 , an improvement module 5 , a registration feature module 3 , and a judgment module 6 .

[0098] The registration module is used to provide several different character strings for the registered user to read, obtain the voice of the registered user to read several different character strings, and perform feature extraction for the voice segment after the voice is segmented; the segmentation is Cut each character of the speech content into a speech segment;

[0099] The verification module 2 is used to randomly generate a verification character string, and the verification character string includes n characters;

[0100] The verification feature module 4 is used to obtain a section of verification charact...

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 text semi-relevant voiceprint recognition method and system. The method comprises the steps that voices of different character strings which are read by a registered user areobtained in the registration process, the voices are segmented according to the content, and feature extraction is carried out on the segmented voice segments; a verification character string segmentis generated randomly, a verification feature vector V is obtained, the verification feature vector V is weighted to obtain an improved verification feature vector V', a registration feature vector Uis obtained according to characters corresponding to the verification character string segment, and the similarity llr of the improved verification feature vector V' and the registration feature vector U is calculated to judge whether a verification user is consistent with the registered user or not. The voiceprint recognition method and system are simple, reliable and safe, meanwhile, differentrecognition effects of all numbers are fully considered, and the voiceprint recognition precision is greatly improved.

Description

technical field [0001] The invention relates to the technical field of voiceprint recognition, in particular to a text semi-correlated voiceprint recognition method related to a string of random dynamic numbers. Background technique [0002] There are mainly two voiceprint recognition methods in the existing voiceprint recognition, namely fixed text (text-dependent) voiceprint recognition and text-independent (text-independent) voiceprint recognition. In practical applications, text-independent voiceprint recognition requires Only a very long speech can achieve the ideal recognition effect, while fixed text voiceprint recognition only needs a short sentence to achieve a good recognition effect. However, in practical applications, the fixed text voiceprint recognition method is difficult to prevent recording and other situations due to the fixed text content, and the fixed text voiceprint recognition method has a large security problem. [0003] In the practical application ...

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): G10L17/08G10L17/02G10L17/24
CPCG10L17/02G10L17/08G10L17/24
Inventor 洪国强肖龙源蔡振华李稀敏刘晓葳谭玉坤
Owner XIAMEN KUAISHANGTONG INFORMATION TECH CO LTD