Shoko Araki
Nippon Telegraph and Telephone
276 Papers
1.7K Citations
Shoko Araki is an academic researcher from Nippon Telegraph and Telephone. The author has contributed to research in topics: Blind signal separation & Source separation. The author has an hindex of 42, co-authored 238 publications. Previous affiliations of Shoko Araki include Hokkaido University & University of Tsukuba.
Chat about Author
Papers
Statistical speech activity detection based on spatial power distribution for analyses of poster presentations.
Kentaro Ishizuka,Shoko Araki,Tatsuya Kawahara +2 more
- 22 Sep 2008
TL;DR: The proposed microphone array based statistical speech activity detection method can exploit the enhanced signals obtained from timefrequency masking, and work even in the presence of environmental noise by utilizing the a priori signal-to-noise ratios of the spatial power distributions.
3
•Posted Content
PILOT: Introducing Transformers for Probabilistic Sound Event Localization
Christopher Schymura,Benedikt Bönninghoff,Tsubasa Ochiai,Marc Delcroix,Keisuke Kinoshita,Tomohiro Nakatani,Shoko Araki,Dorothea Kolossa +7 more
TL;DR: In this paper, a transformer-based sound event localization framework is proposed, where temporal dependencies in the received multi-channel audio signals are captured via self-attention mechanisms and estimated sound event positions are represented as multivariate Gaussian variables, yielding an additional notion of uncertainty.
3
Reverberation-robust underdetermined source separation with non-negative tensor double deconvolution
Naoki Murata,Hirokazu Kameoka,Keisuke Kinoshita,Shoko Araki,Tomohiro Nakatani,Shoichi Koyama,Hiroshi Saruwatari +6 more
- 01 Aug 2016
TL;DR: This paper proposes an extension of the multichannel nonnegative matrix factorization (NMF) approach to deal with the problem of underdetermined source separation in time-variant reverberant environments by models the mixing system as a non-negative convolutive mixture based on the concept of a “semi-time- Variant system” to handle the reverberation in a room as well.
3
The 2011 signal separation evaluation campaign (SiSEC2011): - biomedical data analysis -
Guido Nolte,Dominik Lutter,Andreas Ziehe,Francesco Nesta,Emmanuel Vincent,Zbyněk Koldovský,Alexis Benichoux,Shoko Araki +7 more
- 12 Mar 2012
TL;DR: This paper summarizes the bio part of the 2011 community based Signal Separation Evaluation Campaign (SiSEC2011), which provided an overview of the biomedical datasets, tasks and criteria, and the achieved results.
•Posted Content
Listen to What You Want: Neural Network-based Universal Sound Selector
TL;DR: In this paper, a universal sound selection neural network is proposed to directly select AE sounds from a mixture given user-specified target AE classes, independently of the number of sources in the mixture.
3