Semi-supervised training in low-resource ASR and KWS
Florian Metze, Ankur Gandhe, Yajie Miao, Zaid Sheikh, Yun Wang, Di Xu, Hao Zhang, Jungsuk Kim
Abstract
In particular for “low resource” Keyword Search (KWS) and Speech-to-Text (STT) tasks, more untranscribed test data may be available than training data. Several approaches have been proposed to make this data useful during system development, even when initial systems have Word Error Rates (WER) above 70%. In this paper, we present a set of experiments on low-resource languages in telephony speech quality in Assamese, Bengali, Lao, Haitian, Zulu, and Tamil, demonstrating the impact that such techniques can have, in particular learning robust bottle-neck features on the test data. In the case of Tamil, when significantly more test data than training data is available, we integrated semi-supervised training and speaker adaptation on the test data, and achieved significant additional improvements in STT and KWS.
BibTeX
@inproceedings{icassp2015_semisupervisedtr,
title = {Semi-supervised training in low-resource ASR and KWS},
author = {Florian Metze and Ankur Gandhe and Yajie Miao and Zaid Sheikh and Yun Wang and Di Xu and Hao Zhang and Jungsuk Kim and Ian R. Lane and Wonkyum Lee and Sebastian Stüker and Markus Müller},
booktitle = {ICASSP 2015},
year = {2015}
}