Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation

L. Drude, Takuya Higuchi, K. Kinoshita, T. Nakatani, R. Haeb-Umbach, in: ICASSP 2018, Calgary, Canada, 2018.

Conference Paper | English
Author
; ; ; ;
Abstract
Deep attractor networks (DANs) are a recently introduced method to blindly separate sources from spectral features of a monaural recording using bidirectional long short-term memory networks (BLSTMs). Due to the nature of BLSTMs, this is inherently not online-ready and resorting to operating on blocks yields a block permutation problem in that the index of each speaker may change between blocks. We here propose the joint modeling of spatial and spectral features to solve the block permutation problem and generalize DANs to multi-channel meeting recordings: The DAN acts as a spectral feature extractor for a subsequent model-based clustering approach. We first analyze different joint models in batch-processing scenarios and finally propose a block-online blind source separation algorithm. The efficacy of the proposed models is demonstrated on reverberant mixtures corrupted by real recordings of multi-channel background noise. We demonstrate that both the proposed batch-processing and the proposed block-online system outperform (a) a spatial-only model with a state-of-the-art frequency permutation solver and (b) a spectral-only model with an oracle block permutation solver in terms of signal to distortion ratio (SDR) gains.
Publishing Year
Proceedings Title
ICASSP 2018, Calgary, Canada
LibreCat-ID

Cite this

Drude L, Higuchi, Takuya , Kinoshita K, Nakatani T, Haeb-Umbach R. Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation. In: ICASSP 2018, Calgary, Canada. ; 2018.
Drude, L., Higuchi, Takuya , Kinoshita, K., Nakatani, T., & Haeb-Umbach, R. (2018). Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation. In ICASSP 2018, Calgary, Canada.
@inproceedings{Drude_Higuchi,_Kinoshita_Nakatani_Haeb-Umbach_2018, title={Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation}, booktitle={ICASSP 2018, Calgary, Canada}, author={Drude, Lukas and Higuchi, Takuya and Kinoshita, Keisuke and Nakatani, Tomohiro and Haeb-Umbach, Reinhold}, year={2018} }
Drude, Lukas, Takuya Higuchi, Keisuke Kinoshita, Tomohiro Nakatani, and Reinhold Haeb-Umbach. “Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation.” In ICASSP 2018, Calgary, Canada, 2018.
L. Drude, Takuya Higuchi, K. Kinoshita, T. Nakatani, and R. Haeb-Umbach, “Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation,” in ICASSP 2018, Calgary, Canada, 2018.
Drude, Lukas, et al. “Dual Frequency- and Block-Permutation Alignment for Deep Learning Based Block-Online Blind Source Separation.” ICASSP 2018, Calgary, Canada, 2018.

Link(s) to Main File(s)
Access Level
Restricted Closed Access
External material:
Supplementary Material
Description
Poster

Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar