流匹配文本到语音系统实现了卓越的零样本质量，但在部署后保持静态: 词汇表外专有名词的发音错误仍然存在，除非重新训练模型。我们引入了 FlowEdit,，这是一个用于冻结流匹配 TTS 的终生适应框架，它可以将发音校正学习为潜在调节编辑而不是权重更新。当提供纠正反馈时, FlowEdit 会优化文本嵌入空间, 中的标记级扰动，然后将纠正存储在现代 Hopfield 网络中，作为内容可寻址情景存储器。在推理时, 校正是通过软注意力检索的，具有相似性门,，可实现模糊形态匹配。在我们策划的涵盖 18 个语系的 312 个多语言专有名词的基准测试中, FlowEdit 相对于零样本基线将目标词音素错误率降低了 92.7%，同时保持相同的一般语音质量。在单个 GPU 上，校正大约需要 15 秒即可完成。
Flow-matching text-to-speech systems achieve remarkable zero-shot quality but remain static after deployment: pronunciation errors on out-of-vocabulary proper nouns persist unless the model is retrained. We introduce FlowEdit, a life-long adaptation framework for frozen flow-matching TTS that learns pronunciation corrections as latent conditioning edits rather than weight updates. When corrective feedback is provided, FlowEdit optimizes a token-level perturbation in the text embedding space, then stores the correction in a Modern Hopfield Network serving as content-addressable episodic memory. At inference, corrections are retrieved via soft attention with a similarity gate, enabling fuzzy morphological matching. On our curated benchmark of 312 multilingual proper nouns across 18 language families, FlowEdit reduces target-word Phoneme Error Rate by 92.7% relative to the zero-shot baseline while maintaining identical general-speech quality. Corrections complete in approximately 15 seconds on a single GPU.
科目:人工智能(cs.AI)
Subjects: Artificial Intelligence (cs.AI)