Monaural Speech Separation using Source-Adapted Models
- 1 October 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 114-117
Abstract
We propose a model-based source separation system for use on single channel speech mixtures where the precise source characteristics are not known a priori. We do this by representing the space of source variation with a parametric signal model based on the eigenvoice technique for rapid speaker adaptation. We present an algorithm to infer the characteristics of the sources present in a mixture, allowing for significantly improved separation performance over that obtained using unadapted source models. The algorithm is evaluated on the task defined in the 2006 Speech Separation Challenge [1] and compared with separation using source-dependent models.Keywords
This publication has 6 references indexed in Scilit:
- Super-human multi-talker speech recognition: the IBM 2006 speech separation challenge system2006
- One microphone singing voice separation using source-adapted models2005
- Factorial models and refiltering for speech separation and denoising2003
- A phonetic vocoder2003
- Hidden Markov model decomposition of speech and noise2002
- Rapid speaker adaptation in eigenvoice spaceIEEE Transactions on Speech and Audio Processing, 2000