Sound quality evaluation method based on CEEMD and CWD

Abstract

The invention provides a sound quality evaluation method based on CEEMD and CWD, and the evaluation method comprises the steps: carrying out the CEEMD of each collected noise signal sample, obtaining a plurality of orders of intrinsic mode functions (IMF), recognizing false components in the IMF through the entropy value of the sample, removing the false components, carrying out the CWD analysis of the reserved IMF components, and carrying out the CWD analysis of the residual IMF components. A brand new sound quality evaluation index concept, namely a sound quality evaluation index SQDF, is provided according to a CWD analysis result, the SQDF is used as a characteristic parameter of a corresponding noise signal sample, a relation between the characteristic parameter and a subjective evaluation value is established based on an extreme learning machine ELM, and the sound quality of noise can be evaluated by using the relation. The method can be used for sound quality evaluation of various kinds of noise, the subjective evaluation cost of the noise can be reduced, the influence caused by various adverse factors in the subjective evaluation process can be avoided, the sound quality of the product can be evaluated in the design stage of the product, and therefore the design quality is improved, and the research and development cost is reduced.

Description

technical field [0001] The present application belongs to the technical field of noise analysis and evaluation, and particularly relates to a sound quality evaluation method based on CEEMD (Complete Ensemble Empirical Mode Decomposition) and CWD (Choi-Williams Distribution). Background technique [0002] At present, with the development of the automobile industry, simple sound pressure level control has been unable to meet the customer's sound preference and the manufacturer's sound quality improvement and design needs. At present, loudness, sharpness, roughness, jitter and speech intelligibility are often used to evaluate sound quality. However, this traditional sound quality evaluation parameter is easily affected by the wideband and narrowband noise of signal characteristics, which leads to the use of traditional sound quality. There is no stable correlation between subjective feelings and psychological parameters in the subjective evaluation of quality evaluation methods...

AI Technical Summary

Problems solved by technology

[0002] At present, with the development of the automobile industry, simple sound pressure level control can no longer meet the needs of customers for sound preferences and manufacturers for sound quality improvement and design.

At present, loudness, sharpness, roughness, jitter and speech clarity are often used to evaluate sound quality. However, such traditional sound quality evaluation parameters are easily affected by broadband and narrowband noise of signal characteristics, which leads to the use of traditional sound quality When the quality evaluation method is used for subjective evaluation, there is no stable correlation between subjective feelings and psychological parameters

Images