GPU-ACCELERATED FORWARD-BACKWARD ALGORITHM WITH APPLICATION TO LATTICE-FREE MMI

被引：3

作者：

Ondel, Lucas ^{[1
]}

Lam-Yee-Mui, Lea-Marie ^{[1
,3
]}

Kocour, Martin ^{[2
]}

Corro, Caio Filippo ^{[1
]}

Burget, Lukas ^{[2
]}

机构：

[1] Univ Paris Saclay, LISN, CNRS, Orsay, France

[2] Brno Univ Technol, Fac Informat Technol, Brno, Czech Republic

[3] Vocapia Res, Orsay, France

来源：

2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) | 2022年

关键词：

Lattice-Free MMI; end-to-end ASR; Julia language; forward-backward;

D O I：

10.1109/ICASSP43922.2022.9746824

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

We propose to express the forward-backward algorithm in terms of operations between sparse matrices in a specific semiring. This new perspective naturally leads to a GPU-friendly algorithm which is easy to implement in Julia or any programming languages with native support of semiring algebra. We use this new implementation to train a TDNN with the LF-MMI objective function and we compare the training time of our system with PyChain-a recently introduced C++/CUDA implementation of the LF-MMI loss. Our implementation is about two times faster while not having to use any approximation such as the "leaky-HMM".

引用

页码：8417 / 8421

页数：5

共 50 条

[21] Fermilab multicore and GPU-accelerated clusters for lattice QCD
Fermi National Accelerator Laboratory, Batavia, IL, United States
J. Phys. Conf. Ser., PART 4
[22] Accelerated forward-backward algorithms for structured monotone inclusions
Mainge, Paul-Emile
Weng-Law, Andre
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2024, 88 (01) : 167 - 215
[23] Modified Inertial Forward-Backward Algorithm in Banach Spaces and Its Application
Song, Yanlai
Postolache, Mihai
MATHEMATICS, 2021, 9 (12)
[24] GPU-Accelerated RDP Algorithm for Data Segmentation
Cebrian, Pau
Moure, Juan Carlos
COMPUTATIONAL SCIENCE - ICCS 2020, PT I, 2020, 12137 : 234 - 247
[25] STABILITY OF OVER-RELAXATIONS FOR THE FORWARD-BACKWARD ALGORITHM, APPLICATION TO FISTA
Aujol, J. -F.
Dossal, Ch
SIAM JOURNAL ON OPTIMIZATION, 2015, 25 (04) : 2408 - 2433
[26] Application of the fast multipole method to the generalized forward-backward iterative algorithm
Pino, MR
Obelleiro, F
Landesa, L
Burkholder, RJ
MICROWAVE AND OPTICAL TECHNOLOGY LETTERS, 2000, 26 (02) : 78 - 83
[27] Fermilab multicore and GPU-accelerated clusters for lattice QCD
Holmgren, D.
Seenu, N.
Simone, J.
Singh, A.
INTERNATIONAL CONFERENCE ON COMPUTING IN HIGH ENERGY AND NUCLEAR PHYSICS 2012 (CHEP2012), PTS 1-6, 2012, 396
[28] GPU-accelerated algorithm for asteroid shape modeling
Engels, M.
Hudson, S.
Magri, C.
ASTRONOMY AND COMPUTING, 2019, 28
[29] Integrating Lattice-Free MMI Into End-to-End Speech Recognition
Tian, Jinchuan
Yu, Jianwei
Weng, Chao
Zou, Yuexian
Yu, Dong
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2023, 31 (25-38) : 25 - 38
[30] End-to-end speech recognition using lattice-free MMI
Hadian, Hossein
Sameti, Hossein
Povey, Daniel
Khudanpur, Sanjeev
19TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2018), VOLS 1-6: SPEECH RESEARCH FOR EMERGING MARKETS IN MULTILINGUAL SOCIETIES, 2018, : 12 - 16

← 1 2 3 4 5 →