arXiv cs.CL· Haotian Tan, Hiroki Ouchi, Sakriani Sakti·· 15 小时前AI 评分26
SimulSense：用“语义单元”驱动读写决策的高效同声传译框架
SimulSense: Sense-Driven Interpreting for Efficient Simultaneous Speech Translation
AI 导读
SimulSense 提出一种模仿人类译员的同声传译框架，持续读取输入语音，在感知到新的语义单元时触发写决策输出翻译，无需专门的交错训练数据，也不依赖昂贵的 LLM 推理做决策。对比两个 SOTA 基线，该方法取得更优的质量-延迟权衡并显著提升实时效率，决策速度最高快 9.6 倍。
正文 · 原文
Abstract:How to make human-interpreter-like read/write decisions for simultaneous speech translation (SimulST) systems? Current state-of-the-art systems formulate SimulST as a multi-turn dialogue task, requiring specialized interleaved training data and relying on computationally expensive large language model (LLM) inference for decision-making. In this paper, we propose SimulSense, a novel framework for SimulST that mimics human interpreters by continuously reading input speech and triggering write decisions to produce translation when a new sense unit is perceived. Experiments against two state-of-the-art baseline systems demonstrate that our proposed method achieves a superior quality-latency tradeoff and substantially improved real-time efficiency, where its decision-making is up to 9.6x faster than the baselines.
| Comments: | \c{opyright} 2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2509.21932 [cs.CL] |
| (or arXiv:2509.21932v3 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2509.21932 arXiv-issued DOI via DataCite |
Submission history
From: Haotian Tan [view email]
[v1]
Fri, 26 Sep 2025 06:18:10 UTC (176 KB)
[v2]
Fri, 30 Jan 2026 10:27:11 UTC (175 KB)
[v3]
Sat, 24 Oct 2026 15:12:00 UTC (177 KB)
来源：arXiv cs.CL · arxiv.org