default voice: synthetic (weighted blend of five readers); 7-reader preset removed 934d64f verified mcPear commited on about 9 hours ago
default voice: Wojciech Masiak; 7-reader blend kept as a preset a328640 verified mcPear commited on about 24 hours ago
default voice: 7 readers equal share (Kopa 0), rescaled blending, temperature 0 recommended 4153160 verified mcPear commited on 1 day ago
RECIPE: new voice needs one to two hours of clean recordings 627141b verified mcPear commited on 1 day ago
default voice: equal blend of all 8 readers (only preset); card and samples updated ffc5410 verified mcPear commited on 1 day ago
ONNX with spk_emb input (any blend/style at run time); voices.json = embedding tables + trio preset; Kopa blends removed 29857f3 verified mcPear commited on 1 day ago
ONNX: every voice × style baked as speaker rows; default voice trio (Faszczewska+Masiak+Bielenia); structured voices.json 0cdc3ae verified mcPear commited on 1 day ago
v3: Polish Matcha-TTS + fine-tuned HiFi-GAN vocoder, ONNX with baked voices c24c082 verified mcPear commited on 1 day ago