An improved non-intrusive objective speech quality evaluation based on FGMM and FNN

Jing Wang; Ying Zhang; Yuling Song; Shenghui Zhao; Jingming Kuang

doi:10.1109/CISP.2010.5646757

An improved non-intrusive objective speech quality evaluation based on FGMM and FNN

Jing Wang^*, Ying Zhang, Yuling Song, Shenghui Zhao, Jingming Kuang

^*Corresponding author for this work

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

An improved non-intrusive objective speech quality evaluation method is proposed based on Fuzzy Gaussian Mixture Model (FGMM) and Fuzzy Neural Network (FNN). The degraded speech is separated into three classes (unvoiced, voiced and silence), then for each class the consistency measurement between Perceptual Linear Predictive (PLP) features of the degraded speech and the pre-trained FGMM reference model is calculated and mapped to an objective speech quality score using FNN mapping method. The proposed method performs better than the previous work using GMM and ITU-T P.563 under the test conditions used in this paper.

Original language	English
Title of host publication	Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010
Pages	3495-3499
Number of pages	5
DOIs	https://doi.org/10.1109/CISP.2010.5646757
Publication status	Published - 2010
Event	2010 3rd International Congress on Image and Signal Processing, CISP 2010 - Yantai, China Duration: 16 Oct 2010 → 18 Oct 2010

Publication series

Name	Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010
Volume	7

Conference

Conference	2010 3rd International Congress on Image and Signal Processing, CISP 2010
Country/Territory	China
City	Yantai
Period	16/10/10 → 18/10/10

Keywords

Fuzzy Gaussian mixture model (FGMM)
Fuzzy neural network (FNN)
Non-intrusive
Objective evaluation
Speech quality

Access to Document

10.1109/CISP.2010.5646757

Cite this

Wang, J., Zhang, Y., Song, Y., Zhao, S., & Kuang, J. (2010). An improved non-intrusive objective speech quality evaluation based on FGMM and FNN. In Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010 (pp. 3495-3499). Article 5646757 (Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010; Vol. 7). https://doi.org/10.1109/CISP.2010.5646757

@inproceedings{22584452bff64012af2b4dcf1a3f528d,

title = "An improved non-intrusive objective speech quality evaluation based on FGMM and FNN",

abstract = "An improved non-intrusive objective speech quality evaluation method is proposed based on Fuzzy Gaussian Mixture Model (FGMM) and Fuzzy Neural Network (FNN). The degraded speech is separated into three classes (unvoiced, voiced and silence), then for each class the consistency measurement between Perceptual Linear Predictive (PLP) features of the degraded speech and the pre-trained FGMM reference model is calculated and mapped to an objective speech quality score using FNN mapping method. The proposed method performs better than the previous work using GMM and ITU-T P.563 under the test conditions used in this paper.",

keywords = "Fuzzy Gaussian mixture model (FGMM), Fuzzy neural network (FNN), Non-intrusive, Objective evaluation, Speech quality",

author = "Jing Wang and Ying Zhang and Yuling Song and Shenghui Zhao and Jingming Kuang",

year = "2010",

doi = "10.1109/CISP.2010.5646757",

language = "English",

isbn = "9781424465149",

series = "Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010",

pages = "3495--3499",

booktitle = "Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010",

note = "2010 3rd International Congress on Image and Signal Processing, CISP 2010 ; Conference date: 16-10-2010 Through 18-10-2010",

}

Wang, J , Zhang, Y, Song, Y, Zhao, S & Kuang, J 2010, An improved non-intrusive objective speech quality evaluation based on FGMM and FNN. in Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010., 5646757, Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010, vol. 7, pp. 3495-3499, 2010 3rd International Congress on Image and Signal Processing, CISP 2010, Yantai, China, 16/10/10. https://doi.org/10.1109/CISP.2010.5646757

An improved non-intrusive objective speech quality evaluation based on FGMM and FNN. / Wang, Jing ; Zhang, Ying; Song, Yuling et al.
Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010. 2010. p. 3495-3499 5646757 (Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010; Vol. 7).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An improved non-intrusive objective speech quality evaluation based on FGMM and FNN

AU - Wang, Jing

AU - Zhang, Ying

AU - Song, Yuling

AU - Zhao, Shenghui

AU - Kuang, Jingming

PY - 2010

Y1 - 2010

N2 - An improved non-intrusive objective speech quality evaluation method is proposed based on Fuzzy Gaussian Mixture Model (FGMM) and Fuzzy Neural Network (FNN). The degraded speech is separated into three classes (unvoiced, voiced and silence), then for each class the consistency measurement between Perceptual Linear Predictive (PLP) features of the degraded speech and the pre-trained FGMM reference model is calculated and mapped to an objective speech quality score using FNN mapping method. The proposed method performs better than the previous work using GMM and ITU-T P.563 under the test conditions used in this paper.

AB - An improved non-intrusive objective speech quality evaluation method is proposed based on Fuzzy Gaussian Mixture Model (FGMM) and Fuzzy Neural Network (FNN). The degraded speech is separated into three classes (unvoiced, voiced and silence), then for each class the consistency measurement between Perceptual Linear Predictive (PLP) features of the degraded speech and the pre-trained FGMM reference model is calculated and mapped to an objective speech quality score using FNN mapping method. The proposed method performs better than the previous work using GMM and ITU-T P.563 under the test conditions used in this paper.

KW - Fuzzy Gaussian mixture model (FGMM)

KW - Fuzzy neural network (FNN)

KW - Non-intrusive

KW - Objective evaluation

KW - Speech quality

UR - http://www.scopus.com/inward/record.url?scp=78650575586&partnerID=8YFLogxK

U2 - 10.1109/CISP.2010.5646757

DO - 10.1109/CISP.2010.5646757

M3 - Conference contribution

AN - SCOPUS:78650575586

SN - 9781424465149

T3 - Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010

SP - 3495

EP - 3499

BT - Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010

T2 - 2010 3rd International Congress on Image and Signal Processing, CISP 2010

Y2 - 16 October 2010 through 18 October 2010

ER -

An improved non-intrusive objective speech quality evaluation based on FGMM and FNN

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this