Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Voice frequency selection method based on acoustic characteristics

A technology of acoustic features and frequencies, applied in speech analysis, speech recognition, instruments, etc., can solve problems such as masking human personality characteristics, and achieve the effect of simple and fast operation and improved reliability.

Inactive Publication Date: 2012-09-19
章钊
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The main purpose of the present invention is to provide a new voice frequency selection method, which is applied to the front-end acoustic feature extraction level, solves the problem of voice content covering up human personality characteristics, and improves the reliability of the text-related speaker recognition system when the registered voice content is leaked sex

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] 1. All program languages ​​are used. First, the traditional acoustic feature extraction involving the frequency domain is realized. The extraction process is carried out until the frequency domain is transformed. At this time, the time domain speech signal is transformed to the frequency domain for output.

[0014] 2. Then add the voice frequency selection method based on acoustic characteristics of the present invention, intercept the frequency part that can represent the personality information of people more, and discard the frequency part where there is no special distinguishing vowel between people, this time the frequency domain Speech signals only retain part of the sub-band information. Generally, the intercepted frequency range is between 1.5kHz and 6.5kHz.

[0015] 3. Finally, the traditional acoustic feature extraction involving the frequency domain is continued with the intercepted frequency domain speech signal.

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 extraction of acoustic characteristics is an important link in text-dependent speaker recognition, and in order to give consideration to the reliability of a text-dependent speaker recognition system in divulgence of registered voice contents, a frequency selection method based on acoustic characteristics provided by the invention can be introduced. The frequency selection method provides a good solution to the problem that personality characteristics of people are covered by the voice contents, is fast in realization, simple and convenient, and convenient in transplantation, and can be used in various extraction processes of acoustic characteristics involving in frequency domains.

Description

technical field [0001] The invention relates to the front-end feature extraction technology in text-related speaker recognition. Background technique [0002] Text-dependent speaker recognition emphasizes that the training speech is consistent with the corresponding correct test speech text content, and rejects the tester's login when the tester and the registrant are not the same person. This method is usually used for personal identification, and has broad application prospects in fields such as public security and justice, banks, and airports with high security requirements. [0003] The feature parameters used in text-related speaker recognition are all classical acoustic features, which contain not only speech content information, but also human personality characteristics. Speech content greatly affects the spatial distribution of acoustic features, which can mask human personality to some extent. When the voice content is eavesdropped, the impersonator's imitation w...

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): G10L15/02
Inventor 章钊
Owner 章钊
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
Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com