RedApt: An Adaptor for wav2vec 2 EncodingFaster and Smaller Speech Translation without Quality Compromise

Jinming Zhao, Hao Yang, Gholamreza Haffari, Ehsan Shareghi


Abstract
Pre-trained speech Transformers in speech translation (ST) have facilitated state-of-the-art (SotA) results; yet, using such encoders is computationally expensive. To improve this, we present a novel Reducer Adaptor block, RedApt, that could be seamlessly integrated within any Transformer-based speech encoding architecture. Integrating the pretrained wav2vec 2 speech encoder with RedAptbrings 41% speedup, 33% memory reduction with 24% fewer FLOPs at inference. To our positive surprise, our ST model with RedApt outperforms the SotA architecture by an average of 0.68 BLEU score on 8 language pairs from Must-C.
Anthology ID:
2022.findings-emnlp.142
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1960–1967
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.142
DOI:
10.18653/v1/2022.findings-emnlp.142
Bibkey:
Cite (ACL):
Jinming Zhao, Hao Yang, Gholamreza Haffari, and Ehsan Shareghi. 2022. RedApt: An Adaptor for wav2vec 2 EncodingFaster and Smaller Speech Translation without Quality Compromise. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 1960–1967, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
RedApt: An Adaptor for wav2vec 2 EncodingFaster and Smaller Speech Translation without Quality Compromise (Zhao et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.142.pdf
Video:
 https://aclanthology.org/2022.findings-emnlp.142.mp4