Transfer Learning Approaches for Streaming End-to-End Speech Recognition System

Vikas Joshi, Rui Zhao, Rupesh R. Mehta, Kshitiz Kumar, Jinyu Li


Transfer learning (TL) is widely used in conventional hybrid automatic speech recognition (ASR) system, to transfer the knowledge from source to target language. TL can be applied to end-to-end (E2E) ASR system such as recurrent neural network transducer (RNN-T) models, by initializing the encoder and/or prediction network of the target language with the pre-trained models from source language. In the hybrid ASR system, transfer learning is typically done by initializing the target language acoustic model (AM) with source language AM. Several transfer learning strategies exist in the case of the RNN-T framework, depending upon the choice of the initialization model for encoder and prediction networks. This paper presents a comparative study of four different TL methods for RNN-T framework. We show 10%–17% relative word error rate reduction with different TL methods over randomly initialized RNN-T model. We also study the impact of TL with varying amount of training data ranging from 50 hours to 1000 hours and show the efficacy of TL for languages with a very small amount of training data.


 DOI: 10.21437/Interspeech.2020-2345

Cite as: Joshi, V., Zhao, R., Mehta, R.R., Kumar, K., Li, J. (2020) Transfer Learning Approaches for Streaming End-to-End Speech Recognition System. Proc. Interspeech 2020, 2152-2156, DOI: 10.21437/Interspeech.2020-2345.


@inproceedings{Joshi2020,
  author={Vikas Joshi and Rui Zhao and Rupesh R. Mehta and Kshitiz Kumar and Jinyu Li},
  title={{Transfer Learning Approaches for Streaming End-to-End Speech Recognition System}},
  year=2020,
  booktitle={Proc. Interspeech 2020},
  pages={2152--2156},
  doi={10.21437/Interspeech.2020-2345},
  url={http://dx.doi.org/10.21437/Interspeech.2020-2345}
}