Addressing Accent Mismatch In Mandarin-English Code-Switching Speech Recognition
Zhili Tan, Xinghua Fan, Hui Zhu, Ed Lin
Abstract
Automatic speech recognition systems suffer from accuracy degradation when code-switching (multiple languages are spoken in a single utterance) is encountered. This is especially common for non-native speakers where there is a mismatch between speech and acoustic model. In this paper, we experiment on Mandarin-English code-switching audio spoken by native Chinese speakers and evaluate three techniques to improve accuracy-data adaptation, individual senone modeling and lexicon enrichment. Our results show the recognition of accented speech improves up to 12% on various code-switching datasets. We also propose several metrics to measure code-switching recognition quality, not captured in typical word error rate (WER) measurement.
BibTeX
@inproceedings{icassp2020_addressingaccent,
title = {Addressing Accent Mismatch In Mandarin-English Code-Switching Speech Recognition},
author = {Zhili Tan and Xinghua Fan and Hui Zhu and Ed Lin},
booktitle = {ICASSP 2020},
year = {2020}
}