ISCA Archive Eurospeech 1997
ISCA Archive Eurospeech 1997

Voice activity detection using source separation techniques

Tomohiko Taniguchi, Shoji Kajita, Kazuya Takeda, Fumitada Itakura

Blind signal separation method based on minimiz- ing mutual information is applied to deal with multi- speaker problem in speech recognition. Recognition experiments performed under di erent acoustic environments, in a soundproof room and a reverberant room, clarify that 1) the method can improve recognition accuracy by about 20% where SNR condition is 0 dB, 2) the method is more effective when many speakers' speech exist than the simple overlapped situation, and that 3) the method does not work well under reverberant conditions.


doi: 10.21437/Eurospeech.1997-110

Cite as: Taniguchi, T., Kajita, S., Takeda, K., Itakura, F. (1997) Voice activity detection using source separation techniques. Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997), 1103-1106, doi: 10.21437/Eurospeech.1997-110

@inproceedings{taniguchi97_eurospeech,
  author={Tomohiko Taniguchi and Shoji Kajita and Kazuya Takeda and Fumitada Itakura},
  title={{Voice activity detection using source separation techniques}},
  year=1997,
  booktitle={Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997)},
  pages={1103--1106},
  doi={10.21437/Eurospeech.1997-110},
  issn={1018-4074}
}