Identify Speakers in Cocktail Parties with End-to-End Attention

Zhu, Junzhe; Hasegawa‐Johnson, Mark; Sarı, Leda

doi:10.21437/interspeech.2020-2430

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Source separation is the task of decomposing a mixed signal into the original signals prior to the mixing procedure. This is an important task with many downstream applications, e.g., improve the accuracy of automatic speech recognition with multiple speakers [1] [2], or separating out singing voices and music [3]. Due to the recent progress in deep learning, supervised methods have received a lot of interest [4,5,6,7,8].…”

Section: Introductionmentioning

confidence: 99%

Multi-Decoder Dprnn: Source Separation for Variable Number of Speakers

Zhu

Yeh

Hasegawa‐Johnson

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We propose an end-to-end trainable approach to singlechannel speech separation with unknown number of speakers. Our approach extends the MulCat source separation backbone with additional output heads: a count-head to infer the number of speakers, and decoder-heads for reconstructing the original signals. Beyond the model, we also propose a metric on how to evaluate source separation with variable number of speakers. Specifically, we clear up the issue on how to evaluate the quality when the ground-truth has more or less speakers than the ones predicted by the model. We evaluate our approach on the WSJ0-mix datasets, with mixtures up to five speakers. We demonstrate that our approach outperforms state-of-the-art in counting the number of speakers and remains competitive in quality of reconstructed signals.

show abstract