TY  - RPRT
TI  - STCTS: Generative Semantic Compression for Ultra-Low Bitrate Speech via Explicit Text-Prosody-Timbre Decomposition
AU  - Siyu Wang
AU  - Haitao Li
AU  - Donglai Zhu
PY  - 2025
UR  - https://arxiv.org/abs/2512.00451
ID  - 2512.00451
ER  -