Semantic Parsing of Disfluent Speech

Priyanka Sen, Isabel Groves


Abstract
Speech disfluencies are prevalent in spontaneous speech. The rising popularity of voice assistants presents a growing need to handle naturally occurring disfluencies. Semantic parsing is a key component for understanding user utterances in voice assistants, yet most semantic parsing research to date focuses on written text. In this paper, we investigate semantic parsing of disfluent speech with the ATIS dataset. We find that a state-of-the-art semantic parser does not seamlessly handle disfluencies. We experiment with adding real and synthetic disfluencies at training time and find that adding synthetic disfluencies not only improves model performance by up to 39% but can also outperform adding real disfluencies in the ATIS dataset.
Anthology ID:
2021.eacl-main.150
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1748–1753
Language:
URL:
https://aclanthology.org/2021.eacl-main.150
DOI:
10.18653/v1/2021.eacl-main.150
Bibkey:
Cite (ACL):
Priyanka Sen and Isabel Groves. 2021. Semantic Parsing of Disfluent Speech. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1748–1753, Online. Association for Computational Linguistics.
Cite (Informal):
Semantic Parsing of Disfluent Speech (Sen & Groves, EACL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/auto-file-uploads/2021.eacl-main.150.pdf