Federated Acoustic Model Optimization for Automatic Speech Recognition

Conghui Tan; Di Jiang; Huaxiao Mo; Jinhua Peng; Yongxin Tong; Weiwei Zhao; Chaotao Chen; Rongzhong Lian; Yuanfeng Song; Qian Xu

doi:10.1007/978-3-030-59419-0_54

Federated Acoustic Model Optimization for Automatic Speech Recognition

Conghui Tan^*
, Di Jiang
, Huaxiao Mo
, Jinhua Peng
, Yongxin Tong
, Weiwei Zhao
, Chaotao Chen
, Rongzhong Lian
, Yuanfeng Song
, Qian Xu

^*Corresponding author for this work

School of Computer Science and Engineering

Ai Group

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

6 Scopus citations

Abstract

Traditional Automatic Speech Recognition (ASR) systems are usually trained with speech records centralized on the ASR vendor’s machines. However, with data regulations such as General Data Protection Regulation (GDPR) coming into force, sensitive data such as speech records are not allowed to be utilized in such a centralized approach anymore. In this demonstration, we propose and show the method of federated acoustic model optimization in order to solve this problem. This demonstration does not only vividly show the underlying working mechanisms of the proposed method but also provides an interface for the user to customize its hyperparameters. With this demonstration, the audience can experience the effect of federated learning in an interactive fashion and we wish this demonstration would inspire more research on GDPR-compliant ASR technologies.

Original language	English
Title of host publication	Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings
Editors	Yunmook Nah, Bin Cui, Sang-Won Lee, Jeffrey Xu Yu, Yang-Sae Moon, Steven Euijong Whang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	771-774
Number of pages	4
ISBN (Print)	9783030594183
DOIs	https://doi.org/10.1007/978-3-030-59419-0_54
State	Published - 2020
Event	25th International Conference on Database Systems for Advanced Applications, DASFAA 2020 - Jeju, Korea, Republic of Duration: 24 Sep 2020 → 27 Sep 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12114 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Database Systems for Advanced Applications, DASFAA 2020
Country/Territory	Korea, Republic of
City	Jeju
Period	24/09/20 → 27/09/20

Keywords

Automatic Speech Recognition
Federated learning

Access to Document

10.1007/978-3-030-59419-0_54

Cite this

Tan, C., Jiang, D., Mo, H., Peng, J., Tong, Y., Zhao, W., Chen, C., Lian, R., Song, Y., & Xu, Q. (2020). Federated Acoustic Model Optimization for Automatic Speech Recognition. In Y. Nah, B. Cui, S.-W. Lee, J. X. Yu, Y.-S. Moon, & S. E. Whang (Eds.), Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings (pp. 771-774). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12114 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-59419-0_54

Tan, Conghui ; Jiang, Di ; Mo, Huaxiao et al. / Federated Acoustic Model Optimization for Automatic Speech Recognition. Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. editor / Yunmook Nah ; Bin Cui ; Sang-Won Lee ; Jeffrey Xu Yu ; Yang-Sae Moon ; Steven Euijong Whang. Springer Science and Business Media Deutschland GmbH, 2020. pp. 771-774 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{bfeb7140896548fa946e1b73c2b5b2c3,

title = "Federated Acoustic Model Optimization for Automatic Speech Recognition",

abstract = "Traditional Automatic Speech Recognition (ASR) systems are usually trained with speech records centralized on the ASR vendor{\textquoteright}s machines. However, with data regulations such as General Data Protection Regulation (GDPR) coming into force, sensitive data such as speech records are not allowed to be utilized in such a centralized approach anymore. In this demonstration, we propose and show the method of federated acoustic model optimization in order to solve this problem. This demonstration does not only vividly show the underlying working mechanisms of the proposed method but also provides an interface for the user to customize its hyperparameters. With this demonstration, the audience can experience the effect of federated learning in an interactive fashion and we wish this demonstration would inspire more research on GDPR-compliant ASR technologies.",

keywords = "Automatic Speech Recognition, Federated learning",

author = "Conghui Tan and Di Jiang and Huaxiao Mo and Jinhua Peng and Yongxin Tong and Weiwei Zhao and Chaotao Chen and Rongzhong Lian and Yuanfeng Song and Qian Xu",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020 ; Conference date: 24-09-2020 Through 27-09-2020",

year = "2020",

doi = "10.1007/978-3-030-59419-0\_54",

language = "英语",

isbn = "9783030594183",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "771--774",

editor = "Yunmook Nah and Bin Cui and Sang-Won Lee and Yu, \{Jeffrey Xu\} and Yang-Sae Moon and Whang, \{Steven Euijong\}",

booktitle = "Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings",

address = "德国",

}

Tan, C, Jiang, D, Mo, H, Peng, J, Tong, Y, Zhao, W, Chen, C, Lian, R, Song, Y & Xu, Q 2020, Federated Acoustic Model Optimization for Automatic Speech Recognition. in Y Nah, B Cui, S-W Lee, JX Yu, Y-S Moon & SE Whang (eds), Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12114 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 771-774, 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020, Jeju, Korea, Republic of, 24/09/20. https://doi.org/10.1007/978-3-030-59419-0_54

Federated Acoustic Model Optimization for Automatic Speech Recognition. / Tan, Conghui; Jiang, Di; Mo, Huaxiao et al.
Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. ed. / Yunmook Nah; Bin Cui; Sang-Won Lee; Jeffrey Xu Yu; Yang-Sae Moon; Steven Euijong Whang. Springer Science and Business Media Deutschland GmbH, 2020. p. 771-774 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12114 LNCS).

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

TY - GEN

T1 - Federated Acoustic Model Optimization for Automatic Speech Recognition

AU - Tan, Conghui

AU - Jiang, Di

AU - Mo, Huaxiao

AU - Peng, Jinhua

AU - Tong, Yongxin

AU - Zhao, Weiwei

AU - Chen, Chaotao

AU - Lian, Rongzhong

AU - Song, Yuanfeng

AU - Xu, Qian

PY - 2020

Y1 - 2020

N2 - Traditional Automatic Speech Recognition (ASR) systems are usually trained with speech records centralized on the ASR vendor’s machines. However, with data regulations such as General Data Protection Regulation (GDPR) coming into force, sensitive data such as speech records are not allowed to be utilized in such a centralized approach anymore. In this demonstration, we propose and show the method of federated acoustic model optimization in order to solve this problem. This demonstration does not only vividly show the underlying working mechanisms of the proposed method but also provides an interface for the user to customize its hyperparameters. With this demonstration, the audience can experience the effect of federated learning in an interactive fashion and we wish this demonstration would inspire more research on GDPR-compliant ASR technologies.

AB - Traditional Automatic Speech Recognition (ASR) systems are usually trained with speech records centralized on the ASR vendor’s machines. However, with data regulations such as General Data Protection Regulation (GDPR) coming into force, sensitive data such as speech records are not allowed to be utilized in such a centralized approach anymore. In this demonstration, we propose and show the method of federated acoustic model optimization in order to solve this problem. This demonstration does not only vividly show the underlying working mechanisms of the proposed method but also provides an interface for the user to customize its hyperparameters. With this demonstration, the audience can experience the effect of federated learning in an interactive fashion and we wish this demonstration would inspire more research on GDPR-compliant ASR technologies.

KW - Automatic Speech Recognition

KW - Federated learning

UR - https://www.scopus.com/pages/publications/85092113835

U2 - 10.1007/978-3-030-59419-0_54

DO - 10.1007/978-3-030-59419-0_54

M3 - 会议稿件

AN - SCOPUS:85092113835

SN - 9783030594183

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 771

EP - 774

BT - Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings

A2 - Nah, Yunmook

A2 - Cui, Bin

A2 - Lee, Sang-Won

A2 - Yu, Jeffrey Xu

A2 - Moon, Yang-Sae

A2 - Whang, Steven Euijong

PB - Springer Science and Business Media Deutschland GmbH

T2 - 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020

Y2 - 24 September 2020 through 27 September 2020

ER -

Tan C, Jiang D, Mo H, Peng J, Tong Y, Zhao W et al. Federated Acoustic Model Optimization for Automatic Speech Recognition. In Nah Y, Cui B, Lee SW, Yu JX, Moon YS, Whang SE, editors, Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 771-774. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-59419-0_54

Federated Acoustic Model Optimization for Automatic Speech Recognition

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this