skinny.

Latest / Kana & Mari's SoundRepos (English)

hash2430/pitchtron | speech prosody transfer and style TTS | GitHub

Pitchtron is a Korean text-to-speech toolkit for transferring prosody from reference audio to synthesized speech, including dialects and emotional styles. It provides soft and hard prosody-transfer approaches, a Global Style Token baseline, multi-speaker training and inference tools, and audio preprocessing utilities. Related: hash2430/pitchtron / speech prosody transfer and style TTS / GitHub

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.