MURAL - Maynooth University Research Archive Library



    Discovering Convolutive Speech Phones Using Sparseness and Non-negativity


    O'Grady, Paul D. and Pearlmutter, Barak A. (2007) Discovering Convolutive Speech Phones Using Sparseness and Non-negativity. In: ICA 2007: Independent Component Analysis and Signal Separation. Lecture Notes in Computer Science book series (LNCS) (4666). Springer, pp. 520-527. ISBN 9783540744948

    [img]
    Preview
    Download (467kB) | Preview


    Share your research

    Twitter Facebook LinkedIn GooglePlus Email more...



    Add this article to your Mendeley library


    Abstract

    Discovering a representation that allows auditory data to be parsimoniously represented is useful for many machine learning and signal processing tasks. Such a representation can be constructed by Non-negative Matrix Factorisation (NMF), which is a method for finding parts-based representations of non-negative data. Here, we present a convolutive NMF algorithm that includes a sparseness constraint on the activations and has multiplicative updates. In combination with a spectral magnitude transform of speech, this method extracts speech phones that exhibit sparse activation patterns, which we use in a supervised separation scheme for monophonic mixtures.

    Item Type: Book Section
    Additional Information: Cite as: O’Grady P.D., Pearlmutter B.A. (2007) Discovering Convolutive Speech Phones Using Sparseness and Non-negativity. In: Davies M.E., James C.J., Abdallah S.A., Plumbley M.D. (eds) Independent Component Analysis and Signal Separation. ICA 2007. Lecture Notes in Computer Science, vol 4666. Springer, Berlin, Heidelberg
    Keywords: Separation Performance; Positive Matrix Factorization; Sparseness Constraint; Female Speaker; Reconstruction Objective;
    Academic Unit: Faculty of Science and Engineering > Research Institutes > Hamilton Institute
    Item ID: 10244
    Identification Number: https://doi.org/10.1007/978-3-540-74494-8_65
    Depositing User: Barak Pearlmutter
    Date Deposited: 27 Nov 2018 16:54
    Publisher: Springer
    Refereed: Yes
    URI:

    Repository Staff Only(login required)

    View Item Item control page

    Downloads

    Downloads per month over past year