Skip to content
CaMLSysUniversity of Cambridge
Home People Publications Projects Contact

← All publications

Distilling Knowledge from Ensembles of Acoustic Models for Joint CTC-Attention End-to-End Speech Recognition

  • Yan GaoUniversity of Cambridge
  • Titouan ParcolletAvignon University
  • Nicholas D. LaneUniversity of Cambridge

IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2021

Abstract

Knowledge distillation has been widely used to compress existing deep learning models while preserving the performance on a wide range of applications. In the specific context of Automatic Speech Recognition (ASR), distillation from ensembles of acoustic models has recently shown promising results in increasing recognition performance. In this paper, we propose an extension of multi-teacher distillation methods to joint CTC-attention end-to-end ASR systems. We also introduce three novel distillation strategies. The core intuition behind them is to integrate the error rate metric to the teacher selection rather than solely focusing on the observed losses. In this way, we directly distill and optimize the student toward the relevant metric for speech recognition. We evaluate these strategies under a selection of training procedures on different datasets (TIMIT, Librispeech, Common Voice) and various languages (English, French, Italian). In particular, state-of-the-art error rates are reported on the Common Voice French, Italian and TIMIT datasets.

Materials

  • PDF
BibTeX
@inproceedings{gao2020distilling,
  title = {{Distilling Knowledge from Ensembles of Acoustic Models for Joint CTC-Attention End-to-End Speech Recognition}},
  author = {Gao, Yan and Parcollet, Titouan and Lane, Nicholas D.},
  year = {2021},
  booktitle = {IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)},
  pages = {138--145},
  publisher = {IEEE},
  doi = {10.1109/asru51503.2021.9688302},
  url = {https://doi.org/10.1109/asru51503.2021.9688302}
}
Cambridge Machine Learning Systems Group
Department of Computer Science and Technology
University of Cambridge
ContactGitHub