We introduce a data augmentation technique based on byte pair encoding and a BERT-like self-attention model to boost performance on spoken language understanding tasks. We compare and evaluate this method with a range of augmentation techniques encompassing generative models such as VAEs and performance-boosting techniques such as synonym replacement and back-translation. We show our method performs strongly on domain and intent classification tasks for a voice assistant and in a user-study focused on utterance naturalness and semantic similarity.
@article{arxiv.2104.08268,
title = {Data Augmentation for Voice-Assistant NLU using BERT-based Interchangeable Rephrase},
author = {Akhila Yerukola and Mason Bretan and Hongxia Jin},
journal= {arXiv preprint arXiv:2104.08268},
year = {2021}
}