Discovering Convolutive Speech Phones using Sparseness and Non-Negativity Constraints
O'Grady, Paul D. and Pearlmutter, Barak A. (2007) Discovering Convolutive Speech Phones using Sparseness and Non-Negativity Constraints. In: Proceedings of the Seventh International Conference on Independent Component Analysis, September 9-12, 2007, London, UK.
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 an extension to convolutive NMF that includes a sparseness constraint. In combination with a spectral magnitude transform of speech, this method extracts speech phones (and their associated sparse activation patterns), which we use in a supervised separation scheme for monophonic mixtures.
Repository Staff Only: item control page