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

Front voice enhancement method for identifying speaker

A technology for speaker recognition and speech enhancement, applied in speech analysis, speech recognition, instruments, etc., can solve problems such as uneven distribution of noise energy and difficulty in recovering noisy speech signals, and achieve the effect of reducing residual noise

Inactive Publication Date: 2016-03-23
SHENZHEN YINJIAMI TECH CO LTD
View PDF6 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The noise signal in the actual environment is usually a highly non-stationary signal, and its noise energy is extremely unevenly distributed on the spectrum and changes rapidly with time. In this case, it is very difficult to estimate the noise and restore the corresponding speech signal

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
  • Front voice enhancement method for identifying speaker
  • Front voice enhancement method for identifying speaker
  • Front voice enhancement method for identifying speaker

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention will be further described below in conjunction with accompanying drawing:

[0050] Such as figure 1 As shown, the technical solution adopted by the present invention is as follows: a front-end speech enhancement method for speaker recognition, comprising the following steps:

[0051] The first step is to frame and window the input speech signal, and use FFT to calculate the short-term power spectrum for each frame and phase spectrum ;

[0052] The second step is to use the short-term power spectrum of the speech frame to estimate, and use the improved minimum control recursive average algorithm to estimate the power spectrum of the noise ;

[0053] The estimation algorithm obtains the final noise power spectrum result through two smoothings, two minimum value searches and calculating the probability of speech existence. The power spectrum smoothing formula and minimum value search method are as follows:

[0054]

[0055]

[0056] The ...

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 front voice enhancement method for identifying a speaker. The method, based on conventional spectral subtraction, through a non-stationary noise estimation algorithm, obtains preliminarily enhanced voice through multiband spectral subtraction to calculate an auditory masking threshold so as to adaptively adjust correlation parameters of the spectral subtraction and obtain a front processor applied to a speaker identification system. Compared to a conventional voice enhancement algorithm, the anti-noise performance of the speaker identification system can be improved, voice distortion can also be reduced to the maximum degree through parameter adjustment, voice features of the speaker is maintained, and the identification rate of the speaker identification system in a noisy environment can be effectively improved.

Description

technical field [0001] The invention relates to the field of speech signal processing, in particular to a front-end speech enhancement method for speaker recognition. Background technique [0002] Speaker recognition is a technology that identifies a specific person by the sound of their spoken voice. However, due to various noises in the real environment, the performance of the speaker recognition system will be significantly degraded. Adding a suitable speech enhancement module at the front end can effectively improve the recognition rate of the overall system, so as to improve the robustness of the system to noise. However, in practice, the speech enhancement algorithm cannot completely restore the pure speech signal, and the defect of the algorithm itself will bring residual noise and speech distortion. For speaker recognition technology, it uses the speaker's characteristic information contained in the speech to identify the speaker's identity, so it is very sensitive...

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): G10L15/02G10L15/20G10L17/14
CPCG10L15/02G10L15/20G10L17/14
Inventor 李昕朱杰
Owner SHENZHEN YINJIAMI 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
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