# On Target Segmentation for Direct Speech Translation

arXiv: https://arxiv.org/abs/2009.04707 · HTML: https://mlsanity.com/paper/2009.04707/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Sep 2020.

## Built on by (most cited first, 12 of 12)

- [CTC-based Compression for Direct Speech Translation](https://mlsanity.com/paper/2102.01578/) (2021)
- [Attention as a Guide for Simultaneous Speech Translation](https://mlsanity.com/paper/2212.07850/) (2022)
- [Under the Morphosyntactic Lens: A Multifaceted Evaluation of Gender Bias in Speech Translation](https://mlsanity.com/paper/2203.09866/) (2022)
- [Does Simultaneous Speech Translation need Simultaneous Models?](https://mlsanity.com/paper/2204.03783/) (2022)
- [Beyond Voice Activity Detection: Hybrid Audio Segmentation for Direct Speech Translation](https://mlsanity.com/paper/2104.11710/) (2021)
- [Efficient yet Competitive Speech Translation: FBK@IWSLT2022](https://mlsanity.com/paper/2205.02629/) (2022)
- [When Good and Reproducible Results are a Giant with Feet of Clay: The Importance of Software Quality in NLP](https://mlsanity.com/paper/2303.16166/) (2023)
- [How to Split: the Effect of Word Segmentation on Gender Bias in Speech Translation](https://mlsanity.com/paper/2105.13782/) (2021)
- [Who Are We Talking About? Handling Person Names in Speech Translation](https://mlsanity.com/paper/2205.06755/) (2022)
- [SPES: Spectrogram Perturbation for Explainable Speech-to-Text Generation](https://mlsanity.com/paper/2411.01710/) (2024)
- [No Pitch Left Behind: Addressing Gender Unbalance in Automatic Speech Recognition through Pitch Manipulation](https://mlsanity.com/paper/2310.06590/) (2023)
- [How To Build Competitive Multi-gender Speech Translation Models For Controlling Speaker Gender Translation](https://mlsanity.com/paper/2310.15114/) (2023)

## Abstract (arXiv)

Recent studies on direct speech translation show continuous improvements by means of data augmentation techniques and bigger deep learning models. While these methods are helping to close the gap between this new approach and the more traditional cascaded one, there are many incongruities among different studies that make it difficult to assess the state of the art. Surprisingly, one point of discussion is the segmentation of the target text. Character-level segmentation has been initially proposed to obtain an open vocabulary, but it results on long sequences and long training time. Then, subword-level segmentation became the state of the art in neural machine translation as it produces shorter sequences that reduce the training time, while being superior to word-level models. As such, recent works on speech translation started using target subwords despite the initial use of characters and some recent claims of better results at the character level. In this work, we perform an extensive comparison of the two methods on three benchmarks covering 8 language directions and multilingual training. Subword-level segmentation compares favorably in all settings, outperforming its character-level counterpart in a range of 1 to 3 BLEU points.