Nikhil Siddhartha
Nikhil currently works on Speech Research at Google. He has done MS in Computer Science with a focus in Natural Language Processing and Computer Vision from Stony Brook University, NY, US in 2019. He had done his B.Tech in Computer Science from International Institute of Information Technology Hyderabad, India in 2014. Before joining Google, Nikhil has also worked at Adobe. These are his Google Scholar and LinkedIn pages.
Research Areas
Authored Publications
Sort By
Large-scale ASR Domain Adaptation by Self- and Semi-supervised Learning
David Qiu
ICASSP (2022) (to appear)
Preview abstract
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.
View details