Adapting ASR for under-resourced languages using mismatched transcriptions

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,...

Full description

Saved in:
Bibliographic Details
Main Authors: Chunxi Liu, Jyothi, Preethi, Hao Tang, Manohar, Vimal, Sloan, Rose, Kekona, Tyler, Hasegawa-Johnson, Mark, Khudanpur, Sanjeev
Format: Conference Proceeding
Language:English
Subjects:
Acoustics
Adaptation models
ASR adaptation
ASR for under-resourced languages
Buildings
Conferences
Electronics
Entropy
Error analysis
Errors
Hidden Markov models
Languages
mismatched transcriptions
Signal processing
Speech
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:2379-190X
DOI:10.1109/ICASSP.2016.7472797