TR2021-039

Unsupervised Domain Adaptation For Speech Recognition via Uncertainty Driven Self-Training


Abstract:

The performance of automatic speech recognition (ASR) systems typically degrades significantly when the training and test data domains are mismatched. In this paper, we show that self-training (ST) combined with an uncertainty-based pseudo-label filtering approach can be effectively used for domain adaptation. We propose DUST, a dropout-based uncertainty-driven self-training technique which uses agreement between multiple predictions of an ASR system obtained for different dropout settings to measure the model’s uncertainty about its prediction. DUST excludes pseudo-labeled data with high uncertainties from the training, which leads to substantially improved ASR results compared to ST without filtering, and accelerates the training time due to a reduced training data set. Domain adaptation experiments using WSJ as a source domain and TED-LIUM 3 as well as SWITCHBOARD as the target domains show that up to 80% of the performance of a system trained on ground-truth data can be recovered.

 

  • Related Publication

  •  Khurana, S., Moritz, N., Hori, T., Le Roux, J., "Unsupervised Domain Adaptation for Speech Recognition Via Uncertainty Driven Self-Training", arXiv, December 2020.
    BibTeX arXiv
    • @article{Khurana2020dec,
    • author = {Khurana, Sameer and Moritz, Niko and Hori, Takaaki and Le Roux, Jonathan},
    • title = {Unsupervised Domain Adaptation for Speech Recognition Via Uncertainty Driven Self-Training},
    • journal = {arXiv},
    • year = 2020,
    • month = dec,
    • url = {https://arxiv.org/abs/2011.13439}
    • }