jamiepine/voicebox
The open-source AI voice studio. Clone, dictate, create.
A star delta has been computed for 61,707 of 62,640 tracked repositories in this period, and only those can be ranked in it.
A repository missing here has not been measured far enough back yet. That does not mean it did not move.
Star counts measured between 00:37 and 06:42 (BKK).
The open-source AI voice studio. Clone, dictate, create.
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System
A simple, high-quality voice conversion tool focused on ease of use and performance.
语音 AI 女友（Voice AI girlfriend for DeepSeek Harness）：Whisper 语音输入 + Qwen3-TTS 声音克隆 + 句子级流式朗读 + 数字人动画窗。插话/排队双模式，说话即打断。
Confucius4-TTS: a Multilingual and Cross-Lingual Zero-Shot TTS Engine
Turn PDFs and EPUBs into audiobooks; subtitles or videos into dubbed videos (including translation), and more. For free. Pandrator uses local models, including voice-cloning (instant, RVC-enhanced, XTTS fine-tuning) and LLM processing. It aspires to be a user-friendly app with a GUI, an installer and all-in-one packages.
Wunjo Make: Face Swap, Lip Sync, Control Remove Objects & Text & Background, Restyling, Audio Separator, Clone Voice, Video Generation. Open Source, Local & Free.
That's all 8 repositories in this period.