SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yang, Zeyu, Wei, Lai, Koshkin, Roman, Chen, Xi, Nakamura, Satoshi
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913984287866880
author Yang, Zeyu
Wei, Lai
Koshkin, Roman
Chen, Xi
Nakamura, Satoshi
author_facet Yang, Zeyu
Wei, Lai
Koshkin, Roman
Chen, Xi
Nakamura, Satoshi
contents This work proposes a grammar-based chunking strategy that segments input streams into semantically complete units by parsing dependency relations (e.g., noun phrase boundaries, verb-object structures) and punctuation features. The method ensures chunk coherence and minimizes semantic fragmentation. Building on this mechanism, we present SASST (Syntax-Aware Simultaneous Speech Translation), an end-to-end framework integrating frozen Whisper encoder and decoder-only LLM. The unified architecture dynamically outputs translation tokens or <WAIT> symbols to jointly optimize translation timing and content, with target-side reordering addressing word-order divergence. Experiments on CoVoST2 multilingual corpus En-{De, Zh, Ja} demonstrate significant translation quality improvements across languages and validate the effectiveness of syntactic structures in LLM-driven SimulST systems.
format Preprint
id arxiv_https___arxiv_org_abs_2508_07781
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation
Yang, Zeyu
Wei, Lai
Koshkin, Roman
Chen, Xi
Nakamura, Satoshi
Computation and Language
This work proposes a grammar-based chunking strategy that segments input streams into semantically complete units by parsing dependency relations (e.g., noun phrase boundaries, verb-object structures) and punctuation features. The method ensures chunk coherence and minimizes semantic fragmentation. Building on this mechanism, we present SASST (Syntax-Aware Simultaneous Speech Translation), an end-to-end framework integrating frozen Whisper encoder and decoder-only LLM. The unified architecture dynamically outputs translation tokens or <WAIT> symbols to jointly optimize translation timing and content, with target-side reordering addressing word-order divergence. Experiments on CoVoST2 multilingual corpus En-{De, Zh, Ja} demonstrate significant translation quality improvements across languages and validate the effectiveness of syntactic structures in LLM-driven SimulST systems.
title SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation
topic Computation and Language
url https://arxiv.org/abs/2508.07781