ICASSP 2016accepted0 citations

Adapting ASR for under-resourced languages using mismatched transcriptions

Chunxi Liu, Preethi Jyothi, Hao Tang, Vimal Manohar, Rose Sloan, Tyler Kekona, Mark Hasegawa-Johnson, Sanjeev Khudanpur

Abstract

Mismatched transcriptions of speech in a target language refers to transcriptions provided by people unfamiliar with the language, using English letter sequences. In this work, we demonstrate the value of such transcriptions in building an ASR system for the target language. For different languages, we use less than an hour of mismatched transcriptions to successfully adapt baseline multilingual models built with no access to native transcriptions in the target language. The adapted models provide up to 25% relative improvement in phone error rates on an unseen evaluation set.

BibTeX
@inproceedings{icassp2016_adaptingasrforun,
  title = {Adapting ASR for under-resourced languages using mismatched transcriptions},
  author = {Chunxi Liu and Preethi Jyothi and Hao Tang and Vimal Manohar and Rose Sloan and Tyler Kekona and Mark Hasegawa-Johnson and Sanjeev Khudanpur},
  booktitle = {ICASSP 2016},
  year = {2016}
}