Output-based objective voice quality evaluation method

A technology for voice quality and objective evaluation, applied in voice analysis, instruments, etc., can solve problems such as large amount of transformation and calculation

Inactive Publication Date: 2015-02-18
HUNAN INST OF METROLOGY & TEST +1
View PDF5 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Obviously, when extracting the above two feature parameters, the calculation amount of transformation between linear frequency and bending frequency is relatively large

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
  • Output-based objective voice quality evaluation method
  • Output-based objective voice quality evaluation method
  • Output-based objective voice quality evaluation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0106] 1. NLPC cepstrum feature parameter extraction

[0107] In order to reduce the amount of calculation, the present invention uses Bark bilinear transformation to approach the Bark frequency scale, thereby stretching the low-frequency information of speech and suppressing high-frequency information, and obtaining a characteristic parameter called non-uniform linear predictive cepstrum. It can be used for speech robustness analysis and speech recognition, suppressing speaker-dependent features and preserving the linguistic quality of speech segments. Considering that the high-order parameters contain more feature information that depends on the speaker, the present invention uses the feature parameters of the 5th-order Non-uniform Perceptual Linear Prediction Cepstrum (NLPC, Non-uniform Linear Prediction Cepstrum) to analyze the speech quality .

[0108] 1.1. Preprocessing

[0109]Preprocessing is required before extracting NLPC parameters. Preprocessing includes pre-emph...

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 provides an output-based objective voice quality evaluation method. The method includes the following steps that firstly, inhomogeneous linear prediction cepstrum coefficients of clean voice are extracted and used for training a GMM-HMM model, and a reference model is built for the clean voice through training; then the consistency measure between the reference model and the inhomogeneous linear prediction cepstrum coefficient vector of the distortion voice can be obtained through the reference model and the inhomogeneous linear prediction cepstrum coefficient vector of the distortion voice; finally, the mapping relation between the subjective MOS and the consistency measure is built through a multivariate nonlinear regression model, an objective prediction model of the MOS can be obtained, and objective evaluation is conducted on the voice quality through the objective prediction model. According to the output-based objective voice quality evaluation method, the mapping relation between the subjective MOS and the objective measure is built, and a subjective MOS prediction model is obtained, so that the score is closer to the subjective quality.

Description

technical field [0001] The invention relates to the technical field of voice signal processing, in particular to an output-based objective voice quality assessment method. Background technique [0002] The objective evaluation of voice quality can be divided into intrusive and non-intrusive in terms of evaluation methods. Intrusive evaluation is usually also called double-ended or input-output-based evaluation. It judges the quality of the voice by the error between the input signal and the output signal of the voice system. The characteristic parameters are used to establish the evaluation model. Non-intrusive evaluation, also known as single-ended or output-based evaluation, performs quality evaluation based only on the output signal of the speech system. Based on the evaluation of input-output, original speech is necessary when evaluating speech quality. However, in practical applications, it is difficult or inconvenient to collect original speech materials in some occ...

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): G10L25/60
Inventor 李庆生刘良江卞昕柏文琦周鑫彭正梁徐昱
Owner HUNAN INST OF METROLOGY & TEST
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
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap