Improving End-to-end Speech Translation by Leveraging Auxiliary Speech and Text Data
Preprint
Abstract
We present a method for introducing a text encoder into pre-trained end-to-end speech translation systems. It enhances the ability of adapting one modality (i.e., source-language speech) to another (i.e., source-language text). Thus, the speech translation model can learn from both unlabeled and labeled data, especially when the source-language text data is abundant. Beyond this, we present a denoising method to build a robust text encoder that can deal with both normal and noisy text data. Our system sets new state-of-the-arts on the MuST-C En-De, En-Fr, and LibriSpeech En-Fr tasks.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- SpeechUT: Bridging Speech and Text with Hidden-Unit for Encoder-Decoder Based Speech-Text Pre-training2022
- Unified Speech-Text Pre-training for Speech Translation and Recognition2022
- HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden UnitsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Improving Speech Translation by Understanding and Learning from the Auxiliary Text Translation Task2021
- Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoders2021
- Multilingual Denoising Pre-training for Neural Machine TranslationTransactions of the Association for Computational Linguistics, 2020
- Dual-decoder Transformer for Joint Automatic Speech Recognition and Multilingual Speech Translation2020
- BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension2020
- fairseq: A Fast, Extensible Toolkit for Sequence Modeling2019
- Hybrid CTC/Attention Architecture for End-to-End Speech RecognitionIEEE Journal of Selected Topics in Signal Processing, 2017