End-to-end Automatic Speech Recognition and Speech Translation: Integration of Speech Foundational Models and LLMs

2025/10/11 by Luu, Nam, Bojar, Ondřej
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2510.10329

Abstract

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 COMET^DA₂₂ metric.

Citations

• Findings of the IWSLT 2024 Evaluation Campaign
• A Comprehensive Overview of Large Language Models
• On decoder-only architecture for speech-to-text and large language model integration
• Prompting Large Language Models for Zero-Shot Domain Adaptation in Speech Recognition
• SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities
• Robust Speech Recognition via Large-Scale Weak Supervision
• No Language Left Behind: Scaling Human-Centered Machine Translation
• HuBERT: Self-Supervised Speech Representation Learning by Masked\n Prediction of Hidden Units
• End-to-End Speech-Translation with Knowledge Distillation: FBK@IWSLT2020
• End-to-End Automatic Speech Translation of Audiobooks
• SGDR: Stochastic Gradient Descent with Warm Restarts

Related

• End-to-End Speech-to-Text Translation: A Survey
• Multilingual End-to-End Speech Translation
• Blending LLMs into Cascaded Speech Translation: KIT's Offline Speech Translation System for IWSLT 2024
• The IWSLT 2021 BUT Speech Translation Systems
• Hearing to Translate: The Effectiveness of Speech Modality Integration into LLMs