End-to-end speech recognition with reinforcement learning

Zilong Chen; Wenlin Zhang

doi:10.1117/12.2682509

20 June 2023 End-to-end speech recognition with reinforcement learning

Zilong Chen, Wenlin Zhang

Proceedings Volume 12715, Eighth International Conference on Electronic Technology and Information Science (ICETIS 2023); 127151K (2023) https://doi.org/10.1117/12.2682509
Event: Eighth International Conference on Electronic Technology and Information Science (ICETIS 2023), 2023, Dalian, China

Abstract

In recent years, end-to-end automatic speech recognition (ASR) based on deep neural networks become popular because of its simple pipeline and excellent performance. However, there exists a main mismatch between its training and testing that might lead to performance degradation: in the training stage, existing method use the maximum likelihood criterion which aims to maximize log-likelihood of the training data, while in the testing stage the performance is evaluated by word error rate (WER), not log-likelihood. In this paper, we propose an alternative method based on reinforcement learning to make the goals of training and testing more consistent. Viewing speech recognition as a sequence decision process, the encoder-decoder based neural network is used as the policy function. The encoder is a pre-trained speech representation model (Wav2vec2.0), which generates the environment state encoding. The decoder is trained using a policy gradient algorithm based on a mix reward function which reflects both the word error rate and language model score. Experimental results on the LibriSpeech corpus show that our proposed method achieves 4% relative improvements over the baseline with a language model in terms of WER.

Citation Download Citation

Zilong Chen and Wenlin Zhang "End-to-end speech recognition with reinforcement learning", Proc. SPIE 12715, Eighth International Conference on Electronic Technology and Information Science (ICETIS 2023), 127151K (20 June 2023); https://doi.org/10.1117/12.2682509

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available