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

An Evaluation Method of Objective Speech Quality Based on Output

A voice quality and objective technology, applied in voice analysis, instruments, etc., can solve problems such as high cost and difficulty in obtaining original voice

Active Publication Date: 2018-06-15
HUNAN INST OF METROLOGY & TEST
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In many fields, such as wireless mobile communication, space navigation, and modern military, etc., it is often required that the evaluation method has high flexibility, real-time and versatility, and it must also be able to evaluate the voice quality when the original input voice signal is not available. For evaluation, it is often difficult to obtain the corresponding original voice in the objective evaluation based on the input-output method, and the cost of voice storage and other aspects is greater, and there are certain disadvantages in these application scenarios

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
  • An Evaluation Method of Objective Speech Quality Based on Output
  • An Evaluation Method of Objective Speech Quality Based on Output
  • An Evaluation Method of Objective Speech Quality Based on Output

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0103] An evaluation method of objective voice quality based on output, see figure 1 , specifically includes: calculating the Mel frequency cepstral coefficient of the distorted voice after system transmission (the original voice is transmitted through the system to obtain the distorted voice; the process of calculating the Mel frequency cepstral coefficient is the MFCC parameter extraction process); The reference model of human auditory characteristics (first extract the MFCC parameters of the reference speech, and then obtain the GMM-HMM model); the consistency measurement calculation of the Mel frequency cepstral coefficient of the distorted speech and the reference model conforming to the human auditory characteristics (ie Consistency calculation); insert a sequence in the original speech, and calculate the bit error rate of the sequence extracted from the distorted speech after system transmission; establish the relationship between the subjective MOS score and the consist...

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 present invention provides an output-based objective voice quality evaluation method, comprising the following steps: calculating the Mel-frequency cepstrum coefficient of a distorted voice transmitted through a system; obtaining a reference model conforming to human auditory characteristics; performing consistency measurement and calculation on the Mel-frequency cepstrum coefficient of the distorted voice and the reference model conforming to human auditory characteristics; inserting a sequence in an original voice; calculating the bit error rate of the sequence extracted from distorted voice after the system transmission; according to the consistency measurement and the bit error rate, establishing the mapping relationship between the subjective MOS (mean opinion score) and the consistency measurement; obtaining an objective prediction model for the MOS of the voice to be evaluated; and objectively evaluating the voice quality through the objective prediction model. The method of the invention contains simple steps, is convenient to use, and can evaluate the voice quality objectively and effectively, without relying on the subjective evaluation.

Description

technical field [0001] The present invention relates to the technical field of speech signal processing, in particular, to an evaluation method of output-based objective speech quality. Background technique [0002] The objective evaluation of speech quality refers to the automatic judgment of speech quality by machine, which can be divided into two categories according to the need to use input speech: objective evaluation based on input-output mode and objective evaluation based on output mode. [0003] In many fields, such as wireless mobile communication, space navigation, and modern military, etc., it is often required that the evaluation method has high flexibility, real-time and versatility, and it must also be able to evaluate the voice quality when the original input voice signal is not available. For evaluation, it is often difficult to obtain the corresponding original voice in the objective evaluation based on the input-output method, and the cost of voice storage...

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): G10L25/24G10L25/60
CPCG10L25/24G10L25/60
Inventor 李庆先刘良江王晋威朱宪宇熊婕李彦博
Owner HUNAN INST OF METROLOGY & TEST
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