A method for objective assessment of speech quality based on auditory perception characteristics

A voice quality, objective evaluation technology, applied in voice analysis, instruments, etc., can solve problems such as high computational complexity, disadvantageous real-time evaluation of voice quality, etc., to achieve the effect of improving correlation, facilitating performance analysis and avoiding energy leakage

What is AI technical title?
AI technical title is built by PatSnap AI team. It summarizes the technical point description of the patent document.
A voice quality, objective evaluation technology, applied in voice analysis, instruments, etc., can solve problems such as high computational complexity, disadvantageous real-time evaluation of voice quality, etc., to achieve the effect of improving correlation, facilitating performance analysis and avoiding energy leakage

CN104485114BActive Publication Date: 2018-03-06HUNAN INST OF METROLOGY & TEST +1

4 Cites 2 Cited by

Image

Smart Image Click on the blue labels to locate them in the text.

Viewing Examples

Smart Image

Examples

Experimental program

Comparison scheme

Effect test

Embodiment Construction

[0044] 1. Gammatone filter

[0045] The Gammatone filter is a standard cochlear auditory filter, and the time-domain impulse response of the filter is:

[0046] g(t)=B n t n-1 e -2πBt cos(2πf 0 t+φ)u(t) (1)

[0047] Among them: u(t)=0 when t0; parameter B=b 1 ERB(f 0 ), ERB (f 0 ) is the equivalent rectangular bandwidth of the Gammatone filter (equivalent rectangular bandwidth: for the same white noise input, the width of the rectangular filter with the same energy as the specified filter, referred to as ERB), which is the same as the Gammatone filter center frequency f 0 The relation is ERB(f 0 )=24.7+0.108f 0 , parameter b 1 = 1.019 is a parameter introduced to make the function more consistent with physiological data; n is the order of the filter, and research shows that the Gammatone filter with n=4 can well simulate the filtering characteristics of the basilar membrane; the parameter φ is the initial phase of the filter.

[0048] The frequency response charac...

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

Login to View More

Abstract

A method for objective evaluation of speech quality based on auditory perception characteristics, characterized in that: the method is filtered by adding a Gammatone filter bank to a Bark spectrum module in spectrum mapping, and the concrete steps are: 1) by POLQA processing reference signals and degradation signal, then the reference signal and the degraded signal enter the core model; 2) the spectrum mapping in the core model is that the Barker spectrum module adds the Gammatone filter bank for filtering, and then performs auditory transformation to make the extracted auditory spectrum closer to people. 3) After the auditory transformation, the interference analysis is performed to analyze the distortion of the degraded signal relative to the reference signal, and the objective evaluation MOS score is obtained. Compared with other methods, the present invention effectively improves the correlation between the objective evaluation result and the subjective evaluation result.

Description

technical field [0001] The invention relates to the technical field of speech signal processing, in particular to a method for objectively evaluating speech quality based on auditory perception characteristics. Background technique [0002] Voice quality evaluation can be divided into two categories from the evaluation subject: subjective evaluation and objective evaluation. [0003] Subjective evaluation uses people as the main body to evaluate the quality of speech. Although this method is relatively complicated, since people are the final recipients of speech, this evaluation is a true reflection of speech quality. The Mean Opinion Score (MOS) proposed by the ITU organization in 1996 is a widely used subjective evaluation method, which uses the average opinion score of testers to intuitively reflect people's perception of voice quality. The advantage of subjective evaluation is that it conforms to people's feelings about voice quality, but the disadvantages are that it i...

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

06 Mar 2018

Publication

CN104485114B

IPC: G10L25/60

Inventors: 李庆先; 刘良江