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
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1960–1967
- Language:
- URL:
- https://aclanthology.org/2022.findings-emnlp.142
- DOI:
- 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)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.findings-emnlp.142.pdf