Prosodic Feature
Prosodic features, encompassing aspects of speech like pitch, intensity, and rhythm, are crucial for conveying meaning and emotion beyond the literal words spoken. Current research focuses on accurately modeling and manipulating these features in applications such as speech synthesis, editing, and voice conversion, often employing deep learning models like diffusion models, variational autoencoders, and actor-critic reinforcement learning. This work is significant for improving the naturalness and expressiveness of synthetic speech, enhancing accessibility for individuals with communication disorders, and advancing our understanding of human communication itself.
Papers
Controllable Prosody Generation With Partial Inputs
Dan Andrei Iliescu, Devang Savita Ram Mohan, Tian Huey Teh, Zack Hodari
Improving Prosody for Cross-Speaker Style Transfer by Semi-Supervised Style Extractor and Hierarchical Modeling in Speech Synthesis
Chunyu Qiang, Peng Yang, Hao Che, Ying Zhang, Xiaorui Wang, Zhongyuan Wang
Prosodic features improve sentence segmentation and parsing
Elizabeth Nielsen, Sharon Goldwater, Mark Steedman
ProsAudit, a prosodic benchmark for self-supervised speech models
Maureen de Seyssel, Marvin Lavechin, Hadrien Titeux, Arthur Thomas, Gwendal Virlet, Andrea Santos Revilla, Guillaume Wisniewski, Bogdan Ludusan, Emmanuel Dupoux