一个名为 Uh-Mazing 的新基准测试已被开发出来，用于研究话语中断对语音翻译的影响。目前的系统通常会移除这些语音不规则之处，导致意义丢失。该基准测试包含来自 London Lesbian and Gay Switchboard 的、经过人工翻译和话语中断标注的语音，揭示了虚假开始和自我修正会显著降低翻译质量。研究人员发现，模型倾向于省略话语中断，而不是错误翻译它们，并且在推理时解码可以帮助缓解这个问题，而无需重新训练。 AI
影响 强调了保留语音话语中断对于更准确的机器翻译的重要性。
排序理由 该集群描述了在 arXiv 上发布的一篇新学术论文和基准测试。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2608.02138v1 Announce Type: new Abstract: Current speech translation systems, including SpeechLLMs, are trained on cleaned text and tend to strip disfluencies like filled pauses and false starts rather than translate them. We show this comes at a cost: disfluencies carry me…
Current speech translation systems, including SpeechLLMs, are trained on cleaned text and tend to strip disfluencies like filled pauses and false starts rather than translate them. We show this comes at a cost: disfluencies carry meaning that gets lost when speech is cleaned up. …