End-to-end Automatic Speech Recognition and Speech Translation: Integration of Speech Foundational Models and LLMs
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918159131344896 |
|---|---|
| author | Luu, Nam Bojar, Ondřej |
| author_facet | Luu, Nam Bojar, Ondřej |
| contents | Speech Translation (ST) is a machine translation task that involves converting speech signals from one language to the corresponding text in another language; this task has two different approaches, namely the traditional cascade and the more recent end-to-end. This paper explores a combined end-to-end architecture of pre-trained speech encoders and Large Language Models (LLMs) for performing both Automatic Speech Recognition (ASR) and ST simultaneously. Experiments with the English-to-German language pair show that our best model not only can achieve better translation results than SeamlessM4T, a large foundational end-to-end, multi-modal translation model, but can also match the performance of a cascaded system with Whisper and NLLB, with up to a score gain of 8% in $\text{COMET}^{\text{DA}}_{22}$ metric. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_10329 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | End-to-end Automatic Speech Recognition and Speech Translation: Integration of Speech Foundational Models and LLMs Luu, Nam Bojar, Ondřej Computation and Language Speech Translation (ST) is a machine translation task that involves converting speech signals from one language to the corresponding text in another language; this task has two different approaches, namely the traditional cascade and the more recent end-to-end. This paper explores a combined end-to-end architecture of pre-trained speech encoders and Large Language Models (LLMs) for performing both Automatic Speech Recognition (ASR) and ST simultaneously. Experiments with the English-to-German language pair show that our best model not only can achieve better translation results than SeamlessM4T, a large foundational end-to-end, multi-modal translation model, but can also match the performance of a cascaded system with Whisper and NLLB, with up to a score gain of 8% in $\text{COMET}^{\text{DA}}_{22}$ metric. |
| title | End-to-end Automatic Speech Recognition and Speech Translation: Integration of Speech Foundational Models and LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.10329 |