Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Quasi-clean Speech Construction Method for Objective Evaluation of Speech Quality

A clean voice, objective evaluation technology, applied in voice analysis, instruments, etc., can solve the problems of increasing computational complexity, unfavorable real-time estimation, and reducing algorithm efficiency, etc., to eliminate noise parts, improve accuracy, and accurately estimate noise power spectrum Effect

Active Publication Date: 2018-04-13
SOUTH CHINA UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the minimum control recursive averaging algorithm uniformly estimates the distorted speech when estimating and updating the noise spectrum, and does not distinguish the speech segment from the non-speech segment of the distorted speech, so there is a certain gap between the estimation result and the actual noise power spectrum. , and the unified estimation of the distorted speech noise spectrum increases the computational complexity, reduces the efficiency of the algorithm, and is not conducive to real-time estimation

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
  • A Quasi-clean Speech Construction Method for Objective Evaluation of Speech Quality
  • A Quasi-clean Speech Construction Method for Objective Evaluation of Speech Quality
  • A Quasi-clean Speech Construction Method for Objective Evaluation of Speech Quality

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0039] A quasi-clean speech construction method for objective evaluation of speech quality, comprising the steps of:

[0040] 1. Carry out framing and windowing on the distorted speech (frame length 30ms, frame shift 15ms, plus Hamming window), calculate the short-term energy and zero-crossing rate of each frame respectively; then calculate the average energy of the distorted speech, the energy upper threshold, Energy lower threshold, average zero-crossing number, and zero-crossing number threshold. The upper energy threshold is 0.05 times the average energy; the lower energy threshold is 0.25 times the upper energy threshold; the zero-crossing threshold is 0.3 times the average zero-crossing number.

[0041] 2. Adopt the double-threshold method based on energy and zero-crossing rate to determine the start frame and the end frame of the distorted voice segment; use the above-mentioned determined distorted voice segment as the input data of the sohn voice activity detection alg...

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 quasi-clean speech construction method for objective evaluation of speech quality. The method adopts an improved minimum value control recursive average algorithm and multi-band spectrum subtraction to obtain quasi-clean speech of distorted speech, mainly comprising: (1 ) distinguish distorted voice non-speech segment and speech segment; (2) estimate the noise power spectrum of non-speech segment and speech segment respectively according to the division of non-speech segment and speech segment; (3) estimate the noise spectrum according to non-speech segment and speech segment , to calculate the quasi-clean speech power spectrum of the distorted speech. It has the advantages of using quasi-clean speech and distorted speech as the input speech of the PESQ algorithm, and obtaining the objective evaluation score of the distorted speech.

Description

technical field [0001] The invention relates to an objective speech quality evaluation technology, in particular to a quasi-clean speech construction method for the speech quality objective evaluation, and the speech construction method belongs to the field of non-intrusive speech quality objective evaluation. Background technique [0002] Voice quality is one of the important criteria for evaluating the quality of a voice communication system. Voice quality evaluation is generally divided into subjective evaluation methods and objective evaluation methods. The subjective evaluation method relies on the opinions of the listeners to make a judgment on the voice quality, which directly reflects the user's point of view on the quality of the system. Among them, the MOS (Mean Opinion Score) proposed by ITU-T Recommendation P.830 is a widely used subjective evaluation method. However, the subjective evaluation method has poor repeatability, is difficult to organize and implemen...

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 Patents(China)
IPC IPC(8): G10L25/60G10L21/0216
Inventor 贺前华周伟力李洪韬
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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