Improvement of PESQ based on UVS classification and syllable stability detection

Shilei Huang; Jing Wang; Yi Liu

Improvement of PESQ based on UVS classification and syllable stability detection

Shilei Huang^*
, Jing Wang
, Yi Liu

^*此作品的通讯作者

信息与电子学院

Peking University
PKU-HKUST Shenzhen-Hong Kong Institution
Tsinghua University

科研成果: 会议稿件 › 论文 › 同行评审

摘要

Hearing perception is different to speech segment, especially to unvoiced, voiced and silence (UVS) and it is also affected by the stability of frame distortions in each syllable. A good classification of UVS is an important role in speech quality assessment. Therefore, the Perceptual Evaluation of Speech Quality (PESQ) algorithm described in ITU-T P.862 is improved to get a better performance based on UVS classification and syllable stability detection. Experimental results show that the UVS based approach for PESQ is able to improve correlation results for narrowband speech and is potential for other usages.

源语言	英语
页	474-477
页数	4
出版状态	已出版 - 2010
活动	2nd Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2010 - Biopolis, 新加坡期限: 14 12月 2010 → 17 12月 2010

会议

会议	2nd Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2010
国家/地区	新加坡
市	Biopolis
时期	14/12/10 → 17/12/10

其它文件与链接

链接到 Scopus 的出版物

引用此

@conference{e11ccbc9de3f48ef80c777eb5fc8b45d,

title = "Improvement of PESQ based on UVS classification and syllable stability detection",

abstract = "Hearing perception is different to speech segment, especially to unvoiced, voiced and silence (UVS) and it is also affected by the stability of frame distortions in each syllable. A good classification of UVS is an important role in speech quality assessment. Therefore, the Perceptual Evaluation of Speech Quality (PESQ) algorithm described in ITU-T P.862 is improved to get a better performance based on UVS classification and syllable stability detection. Experimental results show that the UVS based approach for PESQ is able to improve correlation results for narrowband speech and is potential for other usages.",

author = "Shilei Huang and Jing Wang and Yi Liu",

year = "2010",

language = "English",

pages = "474--477",

note = "2nd Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2010 ; Conference date: 14-12-2010 Through 17-12-2010",

}

TY - CONF

T1 - Improvement of PESQ based on UVS classification and syllable stability detection

AU - Huang, Shilei

AU - Wang, Jing

AU - Liu, Yi

PY - 2010

Y1 - 2010

N2 - Hearing perception is different to speech segment, especially to unvoiced, voiced and silence (UVS) and it is also affected by the stability of frame distortions in each syllable. A good classification of UVS is an important role in speech quality assessment. Therefore, the Perceptual Evaluation of Speech Quality (PESQ) algorithm described in ITU-T P.862 is improved to get a better performance based on UVS classification and syllable stability detection. Experimental results show that the UVS based approach for PESQ is able to improve correlation results for narrowband speech and is potential for other usages.

AB - Hearing perception is different to speech segment, especially to unvoiced, voiced and silence (UVS) and it is also affected by the stability of frame distortions in each syllable. A good classification of UVS is an important role in speech quality assessment. Therefore, the Perceptual Evaluation of Speech Quality (PESQ) algorithm described in ITU-T P.862 is improved to get a better performance based on UVS classification and syllable stability detection. Experimental results show that the UVS based approach for PESQ is able to improve correlation results for narrowband speech and is potential for other usages.

UR - https://www.scopus.com/pages/publications/79958113707

M3 - Paper

AN - SCOPUS:79958113707

SP - 474

EP - 477

T2 - 2nd Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2010

Y2 - 14 December 2010 through 17 December 2010

ER -