Latest / Kana & Mari's SoundRepos (English)

HojoAI/Hojo-TTS-Light | bilingual lightweight TTS engine | GitHub
Hojo-TTS-Light is an open-source lightweight bilingual Chinese and English text-to-speech system based on token language models. It provides 40M and 80M models, ONNX Runtime inference on CPU or GPU, preset voices, voice cloning with short reference audio, and simple CLI and Python APIs. Related: HojoAI/Hojo-TTS-Light / bilingual lightweight TTS engine / GitHub
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.