1 min voice data can also be used to train a good TTS model! (few shot voice cloning) - GitHub - ayutaz/GPT-SoVITS at uv_env