Voice Pro is a free, open-source Gradio web interface that bundles a complete speech studio for creators: text-to-speech, zero-shot voice cloning, transcription, translation and audio cleanup in one place. The project has collected more than 13,200 stars on GitHub, is released under the GPL-3.0 license, and was actively maintained in 2026. It runs locally, so your audio never has to leave your machine.
Voice Pro wraps several best-in-class open models behind a single browser UI. For speech synthesis it offers Edge-TTS and Kokoro, plus zero-shot cloning engines such as E2 and F5-TTS and CosyVoice that can imitate a voice from a short reference clip. For understanding audio it uses Whisper and WhisperX for accurate speech-to-text and subtitle generation. Around that core it adds the utilities creators actually need: downloading from YouTube, isolating vocals with Demucs, translating subtitles into other languages, and assembling audiobooks, podcasts and karaoke tracks.
The all-in-one design matters in practice: a creator can take a video from download to transcribed subtitles, translated captions and a dubbed voice-over without leaving the interface or uploading material to a chain of separate web services. Long-form work such as chaptered audiobooks is handled with the same point-and-click simplicity, which is why the project has become a staple recommendation in the open-source voice community.
Voice Pro is free and open source under the GPL-3.0 license. Install it locally and use every feature without subscriptions or API fees; you only supply the compute.
Voice Pro is made for YouTubers, podcasters, audiobook producers, educators and developers who want studio-grade speech synthesis, cloning and transcription in a single self-hosted tool instead of juggling several paid services.