Sorry, you need to enable JavaScript to visit this website.

Large-scale ASR Domain Adaptation using Self- and Semi-supervised Learning

Citation Author(s):
Dongseong Hwang, Ananya Misra, Zhouyuan Huo, Nikhil Siddhartha, Shefali Garg, David Qiu, Khe Chai Sim, Trevor Strohman, Françoise Beaufays, Yanzhang He
Submitted by:
Dongseong Hwang
Last updated:
4 May 2022 - 9:06pm
Document Type:
Presentation Slides
Document Year:
2022
Event:
Presenters:
Dongseong Hwang
Paper Code:
SPE-22.1
 

Self- and semi-supervised learning methods have been actively investigated to reduce labeled training data or enhance the model performance. However, the approach mostly focus on in-domain performance for public datasets. In this study, we utilize the combination of self- and semi-supervised learning methods to solve unseen domain adaptation problem in a large-scale production setting for online ASR model. This approach demonstrates that using the source domain data with a small fraction of the target domain data (3%) can recover the performance gap compared to a full data baseline: relative 13.5% WER improvement for target domain data.

up
0 users have voted: