학술논문
Improving The Latency And Quality Of Cascaded Encoders
Document Type
Conference
Author
Sainath, Tara N.; He, Yanzhang; Narayanan, Arun; Botros, Rami; Wang, Weiran; Qiu, David; Chiu, Chung-Cheng; Prabhavalkar, Rohit; Gruenstein, Alexander; Gulati, Anmol; Li, Bo; Rybach, David; Guzman, Emmanuel; McGraw, Ian; Qin, James; Choromanski, Krzysztof; Liang, Qiao; David, Robert; Pang, Ruoming; Chang, Shuo-Yiin; Strohman, Trevor; Huang, W. Ronny; Han, Wei; Wu, Yonghui; Zhang, Yu
Source
ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Acoustics, Speech and Signal Processing (ICASSP), ICASSP 2022 - 2022 IEEE International Conference on. :8112-8116 May, 2022
Subject
Language
ISSN
2379-190X
Abstract
In this paper, we explore reducing computational latency of the 2-pass cascaded encoder model [1]. Specifically, we experiment with reducing the size of the causal 1st-pass and adding capacity to the non-causal 2nd-pass, such that the overall latency can be reduced without loss of quality. In addition, we explore using a confidence model for deciding to stop 2nd-pass recognition if we are confident in the 1st-pass hypothesis. Overall, we are able to reduce latency by a factor of 1.7X, compared to the baseline cascaded encoder from [1]. Secondly, with the added capacity in the non-causal 2nd-pass, we find that we can improve WER by up to 7% relative using wav2vec and minimum word-error-rate (MWER) training.