← 研踪 PaperTrailarXiv:2611.24125 · v1
Location Matters: Layer-wise Prompt Conditioning for Voice--Style Disentanglement in Text-to-Speech
首次提交 2026-11-18 17:42
版本更新 2026-11-18 17:42
雷达观测 2026-11-19 15:04
摘要
Independent control of voice and emotional style characteristics of speech is a long-standing problem in text-to-speech (TTS) systems. We investigate how the conditioning location of voice and style prompts across TTS decoder layers affects the trade-off between preservation and expressiveness for voice identity and emotional style. We introduce a layer-dependent gated fusion mechanism and systematically study which layers should be influenced by which prompt type. Our findings show an asymmetric relationship. Voice benefits from early conditioning for identity preservation, while style benefits from early conditioning for expressiveness. Our findings provide practical guidance that helps neural network architects make an appropriate trade-off for the architecture of TTS systems according to their use cases.
作者备注:Submitted to ICASSP 2027
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 17:42 · 最新观测版本
Location Matters: Layer-wise Prompt Conditioning for Voice--Style Disentanglement in Text-to-Speech
Independent control of voice and emotional style characteristics of speech is a long-standing problem in text-to-speech (TTS) systems. We investigate how the conditioning location of voice and style prompts across TTS decoder layers affects the trade-off between preservation and expressiveness for voice identity and emotional style. We introduce a layer-dependent gated fusion mechanism and systematically study which layers should be influenced by which prompt type. Our findings show an asymmetric relationship. Voice benefits from early conditioning for identity preservation, while style benefits from early conditioning for expressiveness. Our findings provide practical guidance that helps neural network architects make an appropriate trade-off for the architecture of TTS systems according to their use cases.
查看此版本解读 →查看此版本原文 ↗