Sparse overcomplete decomposition for single channel speaker separation

Madhusudana V.S. Shashanka, Bhiksha Raj, Paris Smaragdis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present an algorithm for separating multiple speakers from a mixed single channel recording. The algorithm is based on a model proposed by Raj and Smaragdis [6]. The idea is to extract certain characteristic spectro-temporal basis functions from training data for individual speakers and decompose the mixed signals as linear combinations of these learned bases. In other words, their model extracts a compact code of basis functions that can explain the space spanned by spectral vectors of a speaker. In our model, we generate a sparse-distributed code where we have more basis funcdons than the dimensionality of the space. We propose a probabilistic framework to achieve sparsity. Experiments show that the resulting sparse code better captures the structure in data and hence leads to better separation.

Original languageEnglish (US)
Title of host publication2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07
PagesII641-II644
DOIs
StatePublished - Aug 6 2007
Externally publishedYes
Event2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07 - Honolulu, HI, United States
Duration: Apr 15 2007Apr 20 2007

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2
ISSN (Print)1520-6149

Other

Other2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07
Country/TerritoryUnited States
CityHonolulu, HI
Period4/15/074/20/07

Keywords

  • MAP estimation
  • Minimum entropy methods
  • Separation
  • Speech enhancement

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Sparse overcomplete decomposition for single channel speaker separation'. Together they form a unique fingerprint.

Cite this