Conference paper

HIGUCHI Takuya, KINOSHITA Keisuke, DELCROIX Marc, ŽMOLÍKOVÁ Kateřina and NAKATANI Tomohiro. Deep clustering-based beamforming for separation with unknown number of sources. In: Proceedings of Interspeech 2017. Stockholm: International Speech Communication Association, 2017, pp. 1183-1187. ISSN 1990-9772. Available from: http://www.isca-speech.org/archive/Interspeech_2017/pdfs/0721.PDF
Publication language:english
Original title:Deep clustering-based beamforming for separation with unknown number of sources
Title (cs):Směrování paprsku založené na hlubokém shlukování s neznámým počtem zdrojů
Pages:1183-1187
Proceedings:Proceedings of Interspeech 2017
Conference:Interspeech 2017
Place:Stockholm, SE
Year:2017
URL:http://www.isca-speech.org/archive/Interspeech_2017/pdfs/0721.PDF
Journal:Proceedings of Interspeech, Vol. 2017, No. 08, FR
ISSN:1990-9772
Publisher:International Speech Communication Association
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2017/higuchi_interspeech2017_IS170721.pdf [PDF]
Keywords
source separation, source counting, timefrequency masking, beamforming
Annotation
This article is about the deep clustering-based beamforming for separation with unknown number of sources.
Abstract
This paper extends a deep clustering algorithm for use with
time-frequency masking-based beamforming and perform separation
with an unknown number of sources. Deep clustering
is a recently proposed single-channel source separation algorithm,
which projects inputs into the embedding space and
performs clustering in the embedding domain. In deep clustering,
bi-directional long short-term memory (BLSTM) recurrent
neural networks are trained to make embedding vectors
orthogonal for different speakers and concurrent for the same
speaker. Then, by clustering the embedding vectors at test time,
we can estimate time-frequency masks for separation. In this
paper, we extend the deep clustering algorithm to a multiple
microphone setup and incorporate deep clustering-based timefrequency
mask estimation into masking-based beamforming,
which has been shown to be more effective than masking for
automatic speech recognition. Moreover, we perform source
counting by computing the rank of the covariance matrix of the
embedding vectors. With our proposed approach, we can perform
masking-based beamforming in a multiple-speaker case
without knowing the number of speakers. Experimental results
show that our proposed deep clustering-based beamformer
achieves comparable source separation performance to that obtained
with a complex Gaussian mixture model-based beamformer,
which requires the number of sources in advance for
mask estimation.
BibTeX:
@INPROCEEDINGS{
   author = {Takuya Higuchi and Keisuke Kinoshita and Marc Delcroix and
	Kate{\v{r}}ina {\v{Z}}mol{\'{i}}kov{\'{a}} and Tomohiro
	Nakatani},
   title = {Deep clustering-based beamforming for separation with
	unknown number of sources},
   pages = {1183--1187},
   booktitle = {Proceedings of Interspeech 2017},
   journal = {Proceedings of Interspeech},
   volume = {2017},
   number = {08},
   year = {2017},
   location = {Stockholm, SE},
   publisher = {International Speech Communication Association},
   ISSN = {1990-9772},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11586}
}

Your IPv4 address: 54.224.44.168
Switch to IPv6 connection

DNSSEC [dnssec]