Unsupervised Domain Adaptation by Adversarial Learning for Robust Speech Recognition
Audio and Speech Processing
2018-07-31 v1 Artificial Intelligence
Computation and Language
Sound
Abstract
In this paper, we investigate the use of adversarial learning for unsupervised adaptation to unseen recording conditions, more specifically, single microphone far-field speech. We adapt neural networks based acoustic models trained with close-talk clean speech to the new recording conditions using untranscribed adaptation data. Our experimental results on Italian SPEECON data set show that our proposed method achieves 19.8% relative word error rate (WER) reduction compared to the unadapted models. Furthermore, this adaptation method is beneficial even when performed on data from another language (i.e. French) giving 12.6% relative WER reduction.
Keywords
Cite
@article{arxiv.1807.11284,
title = {Unsupervised Domain Adaptation by Adversarial Learning for Robust Speech Recognition},
author = {Pavel Denisov and Ngoc Thang Vu and Marc Ferras Font},
journal= {arXiv preprint arXiv:1807.11284},
year = {2018}
}
Comments
5 pages, 2 figures, the 13th ITG conference on Speech Communication