| Paper: | MLSP-P3.10 | ||
| Session: | Speech and Audio Processing | ||
| Time: | Wednesday, May 19, 15:30 - 17:30 | ||
| Presentation: | Poster | ||
| Topic: | Machine Learning for Signal Processing: Blind Signal Separation and Independent Component Analysis | ||
| Title: | BAYESIAN SEPARATION OF AUDIO-VISUAL SPEECH SOURCES | ||
| Authors: | Shyamsundar Rajaram; University of Illinois at Urbana-Champaign | ||
| Ara Nefian; Intel Corporation | |||
| Thomas S. Huang; University of Illinois at Urbana-Champaign | |||
| Abstract: | In this paper we investigate the use of audio and visual ratherthan only audio features for the task of speech separation inacoustically noisy environments. The success of existingindependent component analysis (ICA) systems for the separation of a large variety of signals, including speech, is often limited by the ability of this technique to handle noise. In this paper, we introduce a Bayesian model for the mixing process that describes both the bimodality and the time dependency of speech sources. Our experimental results show that the online demixing process presented here outperforms both the ICA and the audio-only Bayesian model at all levels of noise. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops