Speech enhancement via combination of Wiener filter and blind source separation

Hongmei Hu; Jalil Taghia; Jinqiu Sang; Jalal Taghia; Nasser Mohammadiha; Masoumeh Azarpour; Rajyalakshmi Dokku; Shouyan Wang; Mark E. Lutman; Stefan Bleeck

doi:10.1007/978-3-642-25658-5_58

Speech enhancement via combination of Wiener filter and blind source separation

Hongmei Hu^*
, Jalil Taghia
, Jinqiu Sang
, Jalal Taghia
, Nasser Mohammadiha
, Masoumeh Azarpour
, Rajyalakshmi Dokku
, Shouyan Wang
, Mark E. Lutman
, Stefan Bleeck

^*Corresponding author for this work

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

Abstract

Automatic speech recognition (ASR) often fails in acoustically noisy environments. Aimed to improve speech recognition scores of an ASR in a real-life like acoustical environment, a speech pre-processing system is proposed in this paper, which consists of several stages: First, a convolutive blind source separation (BSS) is applied to the spectrogram of the signals that are pre-processed by binaural Wiener filtering (BWF). Secondly, the target speech is detected by an ASR system recognition rate based on a Hidden Markov Model (HMM). To evaluate the performance of the proposed algorithm, the signal-to-interference ratio (SIR), the improvement signal-to-noise ratio (ISNR) and the speech recognition rates of the output signals were calculated using the signal corpus of the CHiME database. The results show an improvement in SIR and ISNR, but no obvious improvement of speech recognition scores. Improvements for future research are suggested.

Original language	English
Title of host publication	Practical Applications of Intelligent Systems
Subtitle of host publication	Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011)
Editors	Yinglin Wang, Tianrui Li
Pages	485-494
Number of pages	10
DOIs	https://doi.org/10.1007/978-3-642-25658-5_58
State	Published - 2011
Externally published	Yes

Publication series

Name	Advances in Intelligent and Soft Computing
Volume	124
ISSN (Print)	1867-5662

Keywords

ASR
BSS
BWF

Access to Document

10.1007/978-3-642-25658-5_58

Cite this

Hu, H., Taghia, J., Sang, J., Taghia, J., Mohammadiha, N., Azarpour, M., Dokku, R., Wang, S., Lutman, M. E., & Bleeck, S. (2011). Speech enhancement via combination of Wiener filter and blind source separation. In Y. Wang, & T. Li (Eds.), Practical Applications of Intelligent Systems: Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011) (pp. 485-494). (Advances in Intelligent and Soft Computing; Vol. 124). https://doi.org/10.1007/978-3-642-25658-5_58

Hu, Hongmei ; Taghia, Jalil ; Sang, Jinqiu et al. / Speech enhancement via combination of Wiener filter and blind source separation. Practical Applications of Intelligent Systems: Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011). editor / Yinglin Wang ; Tianrui Li. 2011. pp. 485-494 (Advances in Intelligent and Soft Computing).

@inproceedings{4ed148992b244e2cad7d3ff845f982a7,

title = "Speech enhancement via combination of Wiener filter and blind source separation",

abstract = "Automatic speech recognition (ASR) often fails in acoustically noisy environments. Aimed to improve speech recognition scores of an ASR in a real-life like acoustical environment, a speech pre-processing system is proposed in this paper, which consists of several stages: First, a convolutive blind source separation (BSS) is applied to the spectrogram of the signals that are pre-processed by binaural Wiener filtering (BWF). Secondly, the target speech is detected by an ASR system recognition rate based on a Hidden Markov Model (HMM). To evaluate the performance of the proposed algorithm, the signal-to-interference ratio (SIR), the improvement signal-to-noise ratio (ISNR) and the speech recognition rates of the output signals were calculated using the signal corpus of the CHiME database. The results show an improvement in SIR and ISNR, but no obvious improvement of speech recognition scores. Improvements for future research are suggested.",

keywords = "ASR, BSS, BWF",

author = "Hongmei Hu and Jalil Taghia and Jinqiu Sang and Jalal Taghia and Nasser Mohammadiha and Masoumeh Azarpour and Rajyalakshmi Dokku and Shouyan Wang and Lutman, \{Mark E.\} and Stefan Bleeck",

year = "2011",

doi = "10.1007/978-3-642-25658-5\_58",

language = "英语",

isbn = "9783642256578",

series = "Advances in Intelligent and Soft Computing",

pages = "485--494",

editor = "Yinglin Wang and Tianrui Li",

booktitle = "Practical Applications of Intelligent Systems",

}

Hu, H, Taghia, J, Sang, J, Taghia, J, Mohammadiha, N, Azarpour, M, Dokku, R, Wang, S, Lutman, ME & Bleeck, S 2011, Speech enhancement via combination of Wiener filter and blind source separation. in Y Wang & T Li (eds), Practical Applications of Intelligent Systems: Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011). Advances in Intelligent and Soft Computing, vol. 124, pp. 485-494. https://doi.org/10.1007/978-3-642-25658-5_58

Speech enhancement via combination of Wiener filter and blind source separation. / Hu, Hongmei; Taghia, Jalil; Sang, Jinqiu et al.
Practical Applications of Intelligent Systems: Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011). ed. / Yinglin Wang; Tianrui Li. 2011. p. 485-494 (Advances in Intelligent and Soft Computing; Vol. 124).

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

TY - GEN

T1 - Speech enhancement via combination of Wiener filter and blind source separation

AU - Hu, Hongmei

AU - Taghia, Jalil

AU - Sang, Jinqiu

AU - Taghia, Jalal

AU - Mohammadiha, Nasser

AU - Azarpour, Masoumeh

AU - Dokku, Rajyalakshmi

AU - Wang, Shouyan

AU - Lutman, Mark E.

AU - Bleeck, Stefan

PY - 2011

Y1 - 2011

N2 - Automatic speech recognition (ASR) often fails in acoustically noisy environments. Aimed to improve speech recognition scores of an ASR in a real-life like acoustical environment, a speech pre-processing system is proposed in this paper, which consists of several stages: First, a convolutive blind source separation (BSS) is applied to the spectrogram of the signals that are pre-processed by binaural Wiener filtering (BWF). Secondly, the target speech is detected by an ASR system recognition rate based on a Hidden Markov Model (HMM). To evaluate the performance of the proposed algorithm, the signal-to-interference ratio (SIR), the improvement signal-to-noise ratio (ISNR) and the speech recognition rates of the output signals were calculated using the signal corpus of the CHiME database. The results show an improvement in SIR and ISNR, but no obvious improvement of speech recognition scores. Improvements for future research are suggested.

AB - Automatic speech recognition (ASR) often fails in acoustically noisy environments. Aimed to improve speech recognition scores of an ASR in a real-life like acoustical environment, a speech pre-processing system is proposed in this paper, which consists of several stages: First, a convolutive blind source separation (BSS) is applied to the spectrogram of the signals that are pre-processed by binaural Wiener filtering (BWF). Secondly, the target speech is detected by an ASR system recognition rate based on a Hidden Markov Model (HMM). To evaluate the performance of the proposed algorithm, the signal-to-interference ratio (SIR), the improvement signal-to-noise ratio (ISNR) and the speech recognition rates of the output signals were calculated using the signal corpus of the CHiME database. The results show an improvement in SIR and ISNR, but no obvious improvement of speech recognition scores. Improvements for future research are suggested.

KW - ASR

KW - BSS

KW - BWF

UR - https://www.scopus.com/pages/publications/84855232635

U2 - 10.1007/978-3-642-25658-5_58

DO - 10.1007/978-3-642-25658-5_58

M3 - 会议稿件

AN - SCOPUS:84855232635

SN - 9783642256578

T3 - Advances in Intelligent and Soft Computing

SP - 485

EP - 494

BT - Practical Applications of Intelligent Systems

A2 - Wang, Yinglin

A2 - Li, Tianrui

ER -

Hu H, Taghia J, Sang J, Taghia J, Mohammadiha N, Azarpour M et al. Speech enhancement via combination of Wiener filter and blind source separation. In Wang Y, Li T, editors, Practical Applications of Intelligent Systems: Proceedings of the Sixth International Conference on Intelligent Systems and Knowledge Engineering, Shanghai, China, Dec 2011 (ISKE2011). 2011. p. 485-494. (Advances in Intelligent and Soft Computing). doi: 10.1007/978-3-642-25658-5_58

Speech enhancement via combination of Wiener filter and blind source separation

Abstract

Publication series

Keywords

Access to Document

Other files and links

Fingerprint

Cite this