44100Hz日本語HuBERTに対応した QuickVC: Any-to-many Voice Conversion Using Inverse Short-time Fourier Transform for Faster Conversion です。

Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis

リアルタイムボイスチェンジャー Realtime Voice Changer

Easily train a good VC model with voice data <= 10 mins!

論文や海外ブログについてまとめるリポジトリ