@inproceedings{zhao-etal-2022-redapt,
    title = "{R}ed{A}pt: An Adaptor for wav2vec 2 EncodingFaster and Smaller Speech Translation without Quality Compromise",
    author = "Zhao, Jinming  and
      Yang, Hao  and
      Haffari, Gholamreza  and
      Shareghi, Ehsan",
    editor = "Goldberg, Yoav  and
      Kozareva, Zornitsa  and
      Zhang, Yue",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, United Arab Emirates",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2022.findings-emnlp.142/",
    doi = "10.18653/v1/2022.findings-emnlp.142",
    pages = "1960--1967",
    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."
}