TY  - RPRT
TI  - SelfTTS: cross-speaker style transfer through explicit embedding disentanglement and self-refinement using self-augmentation
AU  - Lucas H. Ueda
AU  - João G. T. Lima
AU  - Pedro R. Corrêa
AU  - Flávio O. Simões
AU  - Mário U. Neto
AU  - Paula D. P. Costa
PY  - 2026
UR  - https://arxiv.org/abs/2603.22252
ID  - 2603.22252
ER  -