Conference paper

ŽMOLÍKOVÁ Kateřina, DELCROIX Marc, KINOSHITA Keisuke, HIGUCHI Takuya, OGAWA Atsunori and NAKATANI Tomohiro. Speaker-aware neural network based beamformer for speaker extraction in speech mixtures. In: Proceedings of Interspeech 2017. Stocholm: International Speech Communication Association, 2017, pp. 2655-2659. ISSN 1990-9772. Available from: http://www.isca-speech.org/archive/Interspeech_2017/pdfs/0667.PDF
Publication language:english
Original title:Speaker-aware neural network based beamformer for speaker extraction in speech mixtures
Title (cs):Směrovač paprsku založený na neuronové síti poučené o řečníkovi pro extrakci řečníka ze směsi řečových signálů
Pages:2655-2659
Proceedings:Proceedings of Interspeech 2017
Conference:Interspeech 2017
Place:Stocholm, SE
Year:2017
URL:http://www.isca-speech.org/archive/Interspeech_2017/pdfs/0667.PDF
Journal:Proceedings of Interspeech, Vol. 2017, No. 08, FR
ISSN:1990-9772
DOI:10.21437/Interspeech.2017-667
Publisher:International Speech Communication Association
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2017/zmolikova_interspeech2017_IS170667.pdf [PDF]
Keywords
speaker extraction, speaker-aware neural network, beamforming, mask estimation
Annotation
This article is about the speaker-aware neural network based beamformer for speaker extraction in speech mixtures.
Abstract
In this work, we address the problem of extracting one target speaker from a multichannel mixture of speech. We use a neural network to estimate masks to extract the target speaker and derive beamformer filters using these masks, in a similar way as the recently proposed approach for extraction of speech in presence of noise. To overcome the permutation ambiguity of neural network mask estimation, which arises in presence of multiple speakers, we propose to inform the neural network about the target speaker so that it learns to follow the speaker characteristics through the utterance. We investigate and compare different methods of passing the speaker information to the network such as making one layer of the network dependent on speaker characteristics. Experiments on mixture of two speakers demonstrate that the proposed scheme can track and extract a target speaker for both closed and open speaker set cases.
BibTeX:
@INPROCEEDINGS{
   author = {Kate{\v{r}}ina {\v{Z}}mol{\'{i}}kov{\'{a}} and Marc Delcroix
	and Keisuke Kinoshita and Takuya Higuchi and Atsunori Ogawa
	and Tomohiro Nakatani},
   title = {Speaker-aware neural network based beamformer for speaker
	extraction in speech mixtures},
   pages = {2655--2659},
   booktitle = {Proceedings of Interspeech 2017},
   journal = {Proceedings of Interspeech},
   volume = {2017},
   number = {08},
   year = {2017},
   location = {Stocholm, SE},
   publisher = {International Speech Communication Association},
   ISSN = {1990-9772},
   doi = {10.21437/Interspeech.2017-667},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11587}
}

Your IPv4 address: 54.80.87.62
Switch to IPv6 connection

DNSSEC [dnssec]