Repository navigation
DiffSampler 0.3.0
DiffSampler first public release, DiffSampler only contains a DiffSinger frontend (still very important to the core and handles most the actual rendering) currently. DiffSinger full backend will come soon.
Test with my Defoko inspired voicebank Naoko, made from the voice Naoko from the L&H TTS engine Japanese ver: