Segmenting Subtitles for Correcting ASR Segmentation Errors

David Wan, Chris Kedzie, Faisal Ladhak, Elsbeth Turcan, Petra Galuscakova, Elena Zotkina, Zhengping Jiang, Peter Bell, Kathleen McKeown


Abstract
Typical ASR systems segment the input audio into utterances using purely acoustic information, which may not resemble the sentence-like units that are expected by conventional machine translation (MT) systems for Spoken Language Translation. In this work, we propose a model for correcting the acoustic segmentation of ASR models for low-resource languages to improve performance on downstream tasks. We propose the use of subtitles as a proxy dataset for correcting ASR acoustic segmentation, creating synthetic acoustic utterances by modeling common error modes. We train a neural tagging model for correcting ASR acoustic segmentation and show that it improves downstream performance on MT and audio-document cross-language information retrieval (CLIR).
Anthology ID:
2021.eacl-main.248
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Editors:
Paola Merlo, Jorg Tiedemann, Reut Tsarfaty
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2842–2854
Language:
URL:
https://aclanthology.org/2021.eacl-main.248
DOI:
10.18653/v1/2021.eacl-main.248
Bibkey:
Cite (ACL):
David Wan, Chris Kedzie, Faisal Ladhak, Elsbeth Turcan, Petra Galuscakova, Elena Zotkina, Zhengping Jiang, Peter Bell, and Kathleen McKeown. 2021. Segmenting Subtitles for Correcting ASR Segmentation Errors. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 2842–2854, Online. Association for Computational Linguistics.
Cite (Informal):
Segmenting Subtitles for Correcting ASR Segmentation Errors (Wan et al., EACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.eacl-main.248.pdf
Data
OpenSubtitles