Neural Text Normalization with Subword Units

Courtney Mansfield, Ming Sun, Yuzong Liu, Ankur Gandhe, Björn Hoffmeister


Abstract
Text normalization (TN) is an important step in conversational systems. It converts written text to its spoken form to facilitate speech recognition, natural language understanding and text-to-speech synthesis. Finite state transducers (FSTs) are commonly used to build grammars that handle text normalization. However, translating linguistic knowledge into grammars requires extensive effort. In this paper, we frame TN as a machine translation task and tackle it with sequence-to-sequence (seq2seq) models. Previous research focuses on normalizing a word (or phrase) with the help of limited word-level context, while our approach directly normalizes full sentences. We find subword models with additional linguistic features yield the best performance (with a word error rate of 0.17%).
Anthology ID:
N19-2024
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Industry Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Anastassia Loukina, Michelle Morales, Rohit Kumar
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
190–196
Language:
URL:
https://aclanthology.org/N19-2024
DOI:
10.18653/v1/N19-2024
Bibkey:
Cite (ACL):
Courtney Mansfield, Ming Sun, Yuzong Liu, Ankur Gandhe, and Björn Hoffmeister. 2019. Neural Text Normalization with Subword Units. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Industry Papers), pages 190–196, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Neural Text Normalization with Subword Units (Mansfield et al., NAACL 2019)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/N19-2024.pdf