This paper addresses the method of multichannel signal separation with its application to cocktail party speech recognition. First, we present a fundamental principle for multi-channel signal separation which describes what spatial independence criterion results in. Second, for practical implementation of the signal separation filter, we consider a dynamic recurrent network and develop a new simple learning algorithm. The performance of the proposed method is evaluated in terms of word recognition error rate (WER). Experimental results show that our proposed method dramatically improves the word recognition performance in the case of two simultaneous speeches.
Cite as: Choi, S., Hong, H., Glotin, H., Berthommier, F. (2000) Multichannel signal separation for cocktail party speech recognition: a dynamic recurrent network. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 3, 83-86, doi: 10.21437/ICSLP.2000-483
@inproceedings{choi00d_icslp,
author={Seungjin Choi and Heonseok Hong and Hervé Glotin and Frédéric Berthommier},
title={{Multichannel signal separation for cocktail party speech recognition: a dynamic recurrent network}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 3, 83-86},
doi={10.21437/ICSLP.2000-483}
}