Loughborough University
Browse
Salman_ICASSP2013.pdf (143.29 kB)

A new cascaded spectral subtraction approach for binaural speech dereverberation and its application in source separation

Download (143.29 kB)
conference contribution
posted on 2013-04-25, 07:58 authored by Muhammad Salman Khan, Syed M.R. Naqvi, Jonathon Chambers
In this work we propose a new binaural spectral subtraction method for the suppression of late reverberation. The pro- posed approach is a cascade of three stages. The first two stages exploit distinct observations to model and suppress the late reverberation by deriving a gain function. The musical noise artifacts generated due to the processing at each stage are compensated by smoothing the spectral magnitudes of the weighting gains. The third stage linearly combines the gains obtained from the first two stages and further enhances the binaural signals. The binaural gains, obtained by indepen- dently processing the left and right channel signals are com- bined using a new method. Experiments on real data are per- formed in two contexts: dereverberation-only and joint dere- verberation and source separation. Objective results verify the suitability of the proposed cascaded approach in both the contexts.

Funding

Thanks to UET, Peshawar and the Higher Education Commission (HEC) of Pakistan for funding M. S. Khan, and the Engineering and Physical Sciences Research Council (EPSRC) of the UK (Grant number EP/H049665/1).

History

School

  • Mechanical, Electrical and Manufacturing Engineering

Citation

KHAN, M.S., NAQVI, S.M. and CHAMBERS, J., 2013. A new cascaded spectral subtraction approach for binaural speech dereverberation and its application in source separation. IN: IEEE 2013 International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2013), 26-31 May 2013 Vancouver, British Columbia, Canada, pp.6566 - 6570.

Publisher

© IEEE

Version

  • AM (Accepted Manuscript)

Publication date

2013

Notes

This is a conference paper [© 2013 IEEE]. It was published in the IEEE 2013 International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2013). Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works, must be obtained from the IEEE.

Language

  • en

Usage metrics

    Loughborough Publications

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC