PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding - ACL Anthology

PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding

Trang Le, Daniel Lazar, Suyoun Kim, Shan Jiang, Duc Le, Adithya Sagar, Aleksandr Livshits, Ahmed A Aly, Akshat Shrivastava


Abstract
Spoken Language Understanding (SLU) is a critical component of voice assistants; it consists of converting speech to semantic parses for task execution. Previous works have explored end-to-end models to improve the quality and robustness of SLU models with Deliberation, however these models have remained autoregressive, resulting in higher latencies. In this work we introduce PRoDeliberation, a novel method leveraging a Connectionist Temporal Classification-based decoding strategy as well as a denoising objective to train robust non-autoregressive deliberation models. We show that PRoDeliberation achieves the latency reduction of parallel decoding (2-10x improvement over autoregressive models) while retaining the ability to correct Automatic Speech Recognition (ASR) mistranscriptions of autoregressive deliberation systems. We further show that the design of the denoising training allows PRoDeliberation to overcome the limitations of small ASR devices, and we provide analysis on the necessity of each component of the system.
Anthology ID:
2024.findings-emnlp.820
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
14027–14038
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.820
DOI:
10.18653/v1/2024.findings-emnlp.820
Bibkey:
Cite (ACL):
Trang Le, Daniel Lazar, Suyoun Kim, Shan Jiang, Duc Le, Adithya Sagar, Aleksandr Livshits, Ahmed A Aly, and Akshat Shrivastava. 2024. PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 14027–14038, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding (Le et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.820.pdf