Fetching the paper…

CoMoSpeech: One-Step Speech and Singing Voice Synthesis via Consistency Model · Around