Federated Acoustic Model Optimization for Automatic Speech Recognition

被引：4

作者：

Tan, Conghui ^{[1
]}

Jiang, Di ^{[1
]}

Mo, Huaxiao ^{[1
]}

Peng, Jinhua ^{[1
]}

Tong, Yongxin ^{[2
,3
]}

Zhao, Weiwei ^{[1
]}

Chen, Chaotao ^{[1
]}

Lian, Rongzhong ^{[1
]}

Song, Yuanfeng ^{[1
]}

Xu, Qian ^{[1
]}

机构：

[1] WeBank Co Ltd, AI Grp, Shenzhen, Peoples R China

[2] Beihang Univ, SKLSDE Lab, BDBC, Beijing, Peoples R China

[3] Beihang Univ, IRI, BDBC, Beijing, Peoples R China

来源：

DATABASE SYSTEMS FOR ADVANCED APPLICATIONS (DASFAA 2020), PT III | 2020年 / 12114卷

关键词：

Automatic Speech Recognition; Federated learning;

D O I：

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

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.

引用

页码：771 / 774

页数：4

共 50 条

[1] FEDERATED ACOUSTIC MODELING FOR AUTOMATIC SPEECH RECOGNITION
Cui, Xiaodong
Lu, Songtao
Kingsbury, Brian
[J]. 2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 6748 - 6752
[2] PRIVACY ATTACKS FOR AUTOMATIC SPEECH RECOGNITION ACOUSTIC MODELS IN A FEDERATED LEARNING FRAMEWORK
Tomashenko, Natalia
Mdhaffar, Salima
Tommasi, Marc
Esteve, Yannick
Bonastre, Jean-Francois
[J]. 2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2022, : 6972 - 6976
[3] Crosslingual acoustic model development for automatic speech recognition
Diehl, Frank
Moreno, Asuncion
Monte, Enric
[J]. 2007 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING, VOLS 1 AND 2, 2007, : 425 - 430
[4] Acoustic Model Optimization Based On Evolutionary Stochastic Gradient Descent with Anchors for Automatic Speech Recognition
Cui, Xiaodong
Picheny, Michael
[J]. INTERSPEECH 2019, 2019, : 1581 - 1585
[5] A De Novo Divide-and-Merge Paradigm for Acoustic Model Optimization in Automatic Speech Recognition
Tan, Conghui
Jiang, Di
Peng, Jinhua
Wu, Xueyang
Xu, Qian
Yang, Qiang
[J]. PROCEEDINGS OF THE TWENTY-NINTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2020, : 3709 - 3715
[6] A hybrid HMM/BN acoustic model for automatic speech recognition
Markov, K
Nakamura, S
[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2003, E86D (03): : 438 - 445
[7] MARKOV MODEL ACOUSTIC PHONETIC COMPONENT FOR AUTOMATIC SPEECH RECOGNITION
TAPPERT, CC
[J]. INTERNATIONAL JOURNAL OF MAN-MACHINE STUDIES, 1977, 9 (03): : 363 - 373
[8] Acoustic Analysis for Automatic Speech Recognition
O'Shaughnessy, Douglas
[J]. PROCEEDINGS OF THE IEEE, 2013, 101 (05) : 1038 - 1053
[9] Joint Training of Speech Separation, Filterbank and Acoustic Model for Robust Automatic Speech Recognition
Wang, Zhong-Qiu
Wang, DeLiang
[J]. 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5, 2015, : 2839 - 2843
[10] Using Privacy-Transformed Speech in the Automatic Speech Recognition Acoustic Model Training
Salimbajevs, Askars
[J]. HUMAN LANGUAGE TECHNOLOGIES - THE BALTIC PERSPECTIVE (HLT 2020), 2020, 328 : 47 - 54

← 1 2 3 4 5 →