# IIITH-BUT system for IWSLT 2025 low-resource Bhojpuri to Hindi speech translation

arXiv: https://arxiv.org/abs/2506.04714 · HTML: https://mlsanity.com/paper/2506.04714/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 22 references that are in the map.
- First posted on arXiv 5 Jun 2025.

## Builds on (most cited first, 8 of 8)

- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [fairseq S2T: Fast Speech-to-Text Modeling with fairseq](https://mlsanity.com/paper/2010.05171/) (2020)
- [Direct speech-to-speech translation with a sequence-to-sequence model](https://mlsanity.com/paper/1904.06037/) (2019)
- [Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation](https://mlsanity.com/paper/1612.01744/) (2016)
- [Cascade versus Direct Speech Translation: Do the Differences Still Make a Difference?](https://mlsanity.com/paper/2106.01045/) (2021)
- [NAVER LABS Europe's Multilingual Speech Translation Systems for the IWSLT 2023 Low-Resource Track](https://mlsanity.com/paper/2306.07763/) (2023)
- [Strategies for improving low resource speech to text translation relying on pre-trained ASR models](https://mlsanity.com/paper/2306.00208/) (2023)

## Built on by (most cited first, 1 of 1)

- [TeluguST-46: A Benchmark Corpus and Comprehensive Evaluation for Telugu-English Speech Translation](https://mlsanity.com/paper/2512.07265/) (2025)

## Abstract (arXiv)

This paper presents the submission of IIITH-BUT to the IWSLT 2025 shared task on speech translation for the low-resource Bhojpuri-Hindi language pair. We explored the impact of hyperparameter optimisation and data augmentation techniques on the performance of the SeamlessM4T model fine-tuned for this specific task. We systematically investigated a range of hyperparameters including learning rate schedules, number of update steps, warm-up steps, label smoothing, and batch sizes; and report their effect on translation quality. To address data scarcity, we applied speed perturbation and SpecAugment and studied their effect on translation quality. We also examined the use of cross-lingual signal through joint training with Marathi and Bhojpuri speech data. Our experiments reveal that careful selection of hyperparameters and the application of simple yet effective augmentation techniques significantly improve performance in low-resource settings. We also analysed the translation hypotheses to understand various kinds of errors that impacted the translation quality in terms of BLEU.