Single Channel Speech Separation using Enhanced Learning on Embedding Features

Ha Minh Tan; Jia Ching Wang

doi:10.1109/GCCE53005.2021.9621886

Single Channel Speech Separation using Enhanced Learning on Embedding Features

Ha Minh Tan, Jia Ching Wang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

9 Scopus citations

Abstract

Speech separation has been utilized in many important applications such as automatic speech, mobile phones, hearing aids, and human-machine interactions. In particular, deep neural networks have been considered as a great potential for speech and music separation in recent years. In this paper, we propose a discriminative learning model to solve the single-channel speech separation. Firstly, deep clustering (DC) trains the embedding features. And then these features are utilized as the input for the deep neural network to directly isolate the component sources. The separation performances of the proposed model obtain 10.06 dB SDR, 16.50 dB SIR, 11.48 dB SAR, 9.06 dB SI-SNRi, 88% STOI, and 2.03 PESQ on the TSP dataset.

Original language	English
Title of host publication	2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	430-431
Number of pages	2
ISBN (Electronic)	9781665436762
DOIs	https://doi.org/10.1109/GCCE53005.2021.9621886
State	Published - 2021
Event	10th IEEE Global Conference on Consumer Electronics, GCCE 2021 - Kyoto, Japan Duration: 12 Oct 2021 → 15 Oct 2021

Publication series

Name	2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

Conference

Conference	10th IEEE Global Conference on Consumer Electronics, GCCE 2021
Country/Territory	Japan
City	Kyoto
Period	12/10/21 → 15/10/21

Keywords

Supervised speech separation
deep clustering
monaural source separation
speaker separation

Access to Document

10.1109/GCCE53005.2021.9621886

Cite this

Tan, H. M., & Wang, J. C. (2021). Single Channel Speech Separation using Enhanced Learning on Embedding Features. In 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021 (pp. 430-431). (2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GCCE53005.2021.9621886

@inproceedings{d64456f249e941b59857d81d8aee0b1b,

title = "Single Channel Speech Separation using Enhanced Learning on Embedding Features",

abstract = "Speech separation has been utilized in many important applications such as automatic speech, mobile phones, hearing aids, and human-machine interactions. In particular, deep neural networks have been considered as a great potential for speech and music separation in recent years. In this paper, we propose a discriminative learning model to solve the single-channel speech separation. Firstly, deep clustering (DC) trains the embedding features. And then these features are utilized as the input for the deep neural network to directly isolate the component sources. The separation performances of the proposed model obtain 10.06 dB SDR, 16.50 dB SIR, 11.48 dB SAR, 9.06 dB SI-SNRi, 88% STOI, and 2.03 PESQ on the TSP dataset.",

keywords = "Supervised speech separation, deep clustering, monaural source separation, speaker separation",

author = "Tan, {Ha Minh} and Wang, {Jia Ching}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 10th IEEE Global Conference on Consumer Electronics, GCCE 2021 ; Conference date: 12-10-2021 Through 15-10-2021",

year = "2021",

doi = "10.1109/GCCE53005.2021.9621886",

language = "???core.languages.en_GB???",

series = "2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "430--431",

booktitle = "2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021",

}

Tan, HM & Wang, JC 2021, Single Channel Speech Separation using Enhanced Learning on Embedding Features. in 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021. 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021, Institute of Electrical and Electronics Engineers Inc., pp. 430-431, 10th IEEE Global Conference on Consumer Electronics, GCCE 2021, Kyoto, Japan, 12/10/21. https://doi.org/10.1109/GCCE53005.2021.9621886

Single Channel Speech Separation using Enhanced Learning on Embedding Features. / Tan, Ha Minh; Wang, Jia Ching.
2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 430-431 (2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Single Channel Speech Separation using Enhanced Learning on Embedding Features

AU - Tan, Ha Minh

AU - Wang, Jia Ching

PY - 2021

Y1 - 2021

N2 - Speech separation has been utilized in many important applications such as automatic speech, mobile phones, hearing aids, and human-machine interactions. In particular, deep neural networks have been considered as a great potential for speech and music separation in recent years. In this paper, we propose a discriminative learning model to solve the single-channel speech separation. Firstly, deep clustering (DC) trains the embedding features. And then these features are utilized as the input for the deep neural network to directly isolate the component sources. The separation performances of the proposed model obtain 10.06 dB SDR, 16.50 dB SIR, 11.48 dB SAR, 9.06 dB SI-SNRi, 88% STOI, and 2.03 PESQ on the TSP dataset.

AB - Speech separation has been utilized in many important applications such as automatic speech, mobile phones, hearing aids, and human-machine interactions. In particular, deep neural networks have been considered as a great potential for speech and music separation in recent years. In this paper, we propose a discriminative learning model to solve the single-channel speech separation. Firstly, deep clustering (DC) trains the embedding features. And then these features are utilized as the input for the deep neural network to directly isolate the component sources. The separation performances of the proposed model obtain 10.06 dB SDR, 16.50 dB SIR, 11.48 dB SAR, 9.06 dB SI-SNRi, 88% STOI, and 2.03 PESQ on the TSP dataset.

KW - Supervised speech separation

KW - deep clustering

KW - monaural source separation

KW - speaker separation

UR - http://www.scopus.com/inward/record.url?scp=85123497261&partnerID=8YFLogxK

U2 - 10.1109/GCCE53005.2021.9621886

DO - 10.1109/GCCE53005.2021.9621886

M3 - 會議論文篇章

AN - SCOPUS:85123497261

T3 - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

SP - 430

EP - 431

BT - 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th IEEE Global Conference on Consumer Electronics, GCCE 2021

Y2 - 12 October 2021 through 15 October 2021

ER -

Single Channel Speech Separation using Enhanced Learning on Embedding Features

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this