Blind speech separation using a joint model of speech production

Authors

Daniel Smith, University of WollongongFollow
Jason Lukasiak, University of WollongongFollow
Ian Burnett, University of WollongongFollow

RIS ID

12936

Publication Details

This article was originally published as: Smith, D, Lukasiak, J & Burnett, I, Blind speech separation using a joint model of speech production, IEEE Signal Processing Letters, November 2005, 12(11), 784-787. Copyright IEEE 2005.

Abstract

We propose a new blind signal separation (BSS)technique, developed specifically for speech, that exploits a priori knowledge of speech production mechanisms. In our approach, the autoregressive (AR) structure and fundamental frequency ( 0) production mechanisms of speech are jointly modeled. We compare the separation performance of our joint AR-F0 algorithm to existing BSS algorithms that model either speech’s AR structure [1] or 0 [2] individually. Experimental results indicate that the joint algorithm demonstrates superior separation performance to both the individual AR algorithm (up to 77% improvement) and F0 (up to 50% improvement) algorithms. This suggests that speech separation performance is improved by employing a BSS model with a more realistic description of the speech production process.