新式语音翻译模型会忽略说话人性别，导致翻译偏向男性
Different Speech Translation Models Encode and Translate Speaker Gender Differently
- 用探针方法分析多种语音翻译模型的性别编码能力
- 新架构模型几乎不编码性别信息，传统模型则能捕捉
- 性别编码弱的模型更倾向输出男性化译文，适合关注公平性的研究者
近期研究表明，语音模型的隐藏状态可捕捉说话人特征（包括性别）。这一现象是否适用于语音翻译（ST）模型？若成立，对翻译中的说话人性别指派有何影响？本文从可解释性角度出发，采用探针方法评估三种语言方向（英-法/意/西）下多种ST模型的性别编码能力。结果表明，传统编码器-解码器模型能捕捉性别信息，而融合语音编码器与机器翻译系统的新型适配器架构模型则无法做到。此外，我们发现性别编码能力弱的系统倾向于默认输出男性化译文，且该偏差在新架构中更为显著。
原文摘要 · Abstract (English)
Recent studies on interpreting the hidden states of speech models have shown their ability to capture speaker-specific features, including gender. Does this finding also hold for speech translation (ST) models? If so, what are the implications for the speaker's gender assignment in translation? We address these questions from an interpretability perspective, using probing methods to assess gender encoding across diverse ST models. Results on three language directions (English-French/Italian/Spanish) indicate that while traditional encoder-decoder models capture gender information, newer architectures -- integrating a speech encoder with a machine translation system via adapters -- do not. We also demonstrate that low gender encoding capabilities result in systems' tendency toward a masculine default, a translation bias that is more pronounced in newer architectures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。