AINeutralarXiv – CS AI · 6h ago5/10
🧠
Adaptive Oscillatory Inductive Bias for Modeling Sharp Prosodic Dynamics in Diffusion-Based TTS
Researchers introduce OscillaTTS, a diffusion-based text-to-speech system that uses adaptive oscillatory nonlinearity to better model sharp prosodic transitions and rapid pitch variations in expressive speech. The approach improves upon existing methods that rely on fixed periodic activation functions, demonstrating consistent improvements in both objective metrics and subjective evaluations on standard speech datasets.