Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement

Ming Ji; Daniel Crookes

doi:10.1109/ICASSP.2016.7472811

Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement

Ming Ji, Daniel Crookes

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

2 Citations (Scopus)

223 Downloads (Pure)

Abstract

It is shown that under certain conditions it is possible to obtain a good speech estimate from noise without requiring noise estimation. We study an implementation of the theory, namely wide matching, for speech enhancement. The new approach performs sentence-wide joint speech segment estimation subject to maximum recognizability to gain noise robustness. Experiments have been conducted to evaluate the new approach with variable noises and SNRs from -5 dB to noise free. It is shown that the new approach, without any estimation of the noise, significantly outperformed conventional methods in the low SNR conditions while retaining comparable performance in the high SNR conditions. It is further suggested that the wide matching and deep learning approaches can be combined towards a highly robust and accurate speech estimator.

Original language	English
Title of host publication	Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Number of pages	5
DOIs	https://doi.org/10.1109/ICASSP.2016.7472811
Publication status	Published - 19 May 2016
Event	The 41st IEEE International Conference on Acoustics, Speech and Signal Processing - Shanghai, China Duration: 20 Mar 2016 → 25 Mar 2016

Conference

Conference	The 41st IEEE International Conference on Acoustics, Speech and Signal Processing
Country	China
City	Shanghai
Period	20/03/2016 → 25/03/2016

Access to Document

10.1109/ICASSP.2016.7472811Licence: Unspecified

Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement
© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript, 299 KBLicence: Unspecified

Conference website

Fingerprint Dive into the research topics of 'Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement'. Together they form a unique fingerprint.

Ming Ji
- J.Ming@qub.ac.uk
- School of Electronics, Electrical Engineering and Computer Science - Professor
- Speech, Image and Vision Systems
Person: Academic

Cite this

@inproceedings{3aae22f165094dc4911b19833fa27653,

title = "Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement",

abstract = "It is shown that under certain conditions it is possible to obtain a good speech estimate from noise without requiring noise estimation. We study an implementation of the theory, namely wide matching, for speech enhancement. The new approach performs sentence-wide joint speech segment estimation subject to maximum recognizability to gain noise robustness. Experiments have been conducted to evaluate the new approach with variable noises and SNRs from -5 dB to noise free. It is shown that the new approach, without any estimation of the noise, significantly outperformed conventional methods in the low SNR conditions while retaining comparable performance in the high SNR conditions. It is further suggested that the wide matching and deep learning approaches can be combined towards a highly robust and accurate speech estimator.",

author = "Ming Ji and Daniel Crookes",

year = "2016",

month = may,

day = "19",

doi = "10.1109/ICASSP.2016.7472811",

language = "English",

booktitle = "Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

note = "The 41st IEEE International Conference on Acoustics, Speech and Signal Processing ; Conference date: 20-03-2016 Through 25-03-2016",

}

Ji, M & Crookes, D 2016, Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement. in Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing. Institute of Electrical and Electronics Engineers (IEEE), The 41st IEEE International Conference on Acoustics, Speech and Signal Processing, Shanghai, China, 20/03/2016. https://doi.org/10.1109/ICASSP.2016.7472811

TY - GEN

T1 - Wide Matching - An Approach to Improving Noise Robustness for Speech Enhancement

AU - Ji, Ming

AU - Crookes, Daniel

PY - 2016/5/19

Y1 - 2016/5/19

N2 - It is shown that under certain conditions it is possible to obtain a good speech estimate from noise without requiring noise estimation. We study an implementation of the theory, namely wide matching, for speech enhancement. The new approach performs sentence-wide joint speech segment estimation subject to maximum recognizability to gain noise robustness. Experiments have been conducted to evaluate the new approach with variable noises and SNRs from -5 dB to noise free. It is shown that the new approach, without any estimation of the noise, significantly outperformed conventional methods in the low SNR conditions while retaining comparable performance in the high SNR conditions. It is further suggested that the wide matching and deep learning approaches can be combined towards a highly robust and accurate speech estimator.

AB - It is shown that under certain conditions it is possible to obtain a good speech estimate from noise without requiring noise estimation. We study an implementation of the theory, namely wide matching, for speech enhancement. The new approach performs sentence-wide joint speech segment estimation subject to maximum recognizability to gain noise robustness. Experiments have been conducted to evaluate the new approach with variable noises and SNRs from -5 dB to noise free. It is shown that the new approach, without any estimation of the noise, significantly outperformed conventional methods in the low SNR conditions while retaining comparable performance in the high SNR conditions. It is further suggested that the wide matching and deep learning approaches can be combined towards a highly robust and accurate speech estimator.

UR - http://www.icassp2016.org/

U2 - 10.1109/ICASSP.2016.7472811

DO - 10.1109/ICASSP.2016.7472811

M3 - Conference contribution

BT - Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - The 41st IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 20 March 2016 through 25 March 2016

ER -