BIROn - Birkbeck Institutional Research Online

    Bootstrap averaging for model-based source separation in reverberant conditions

    Chandna, Swati and Wang, W. (2018) Bootstrap averaging for model-based source separation in reverberant conditions. IEEE/ACM Transactions on Audio, Speech and Language Processing 26 (4), pp. 806-819. ISSN 2329-9290.

    [img]
    Preview
    Text
    ChandnaWang_TASLP5.pdf - Author's Accepted Manuscript

    Download (385kB) | Preview

    Abstract

    Recently proposed model-based methods use timefrequency (T-F) masking for source separation, where the T-F masks are derived from various cues described by a frequency domain Gaussian Mixture Model (GMM). These methods work well for separating mixtures recorded in low-to-medium level of reverberation, however, their performance degrades as the level of reverberation is increased. We note that the relatively poor performance of these methods under reverberant conditions can be attributed to the high variance of the frequency-dependent GMM parameter estimates. To address this limitation, a novel bootstrap-based approach is proposed to improve the accuracy of expectation maximization (EM) estimates of a frequencydependent GMM based on an a priori chosen initialization scheme. It is shown how the proposed technique allows us to construct time-frequency masks which lead to improved model-based source separation for reverberant speech mixtures. Experiments and analysis are performed on speech mixtures formed using real room-recorded impulse responses.

    Metadata

    Item Type: Article
    Additional Information: (c) 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, 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 components of this work in other works.
    School: Birkbeck Schools and Departments > School of Business, Economics & Informatics > Economics, Mathematics and Statistics
    Divisions > Birkbeck Schools and Departments > School of Business, Economics & Informatics > Economics, Mathematics and Statistics
    Depositing User: Swati Chandna
    Date Deposited: 12 Jan 2018 08:46
    Last Modified: 27 Jun 2020 20:52
    URI: http://eprints.bbk.ac.uk/id/eprint/20811

    Statistics

    Downloads
    Activity Overview
    167Downloads
    128Hits

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item