voxclone

voxclone clones your voice from a short recording — entirely on your device — then speaks any text you type, in your choice of language.

Record (or import) a short voice sample, review the auto-detected transcript, and save it as a voice profile. From there, type any text, pick a language, and voxclone's on-device AI speaks it back in your cloned voice. No image, no audio sample, and no generated speech ever leaves your phone.

WHY VOXCLONE

Cloud voice-cloning tools send your voice to a server, lock features behind a subscription, and can revoke access at any time. voxclone runs the entire clone-and-speak loop on your device: your voice sample never leaves your phone, and there's no account or subscription to lose access to.

FEATURES

• Record or import a reference clip — a short sample is all it takes
• Auto-detected transcript — accept it or edit it yourself
• On-device voice cloning — a neural TTS model runs directly on your phone; nothing is uploaded
• Multilingual generation — speak your cloned voice in the languages the model supports
• Save, replay, and share generated clips
• No account, no login, no cloud sync

PRIVACY, BY DESIGN

• Zero telemetry — voxclone doesn't phone home, ever
• Voice samples and generated audio are processed and stored locally, never transmitted
• The only network request voxclone ever makes is the one-time AI model download
• Open source — inspect exactly what the app does

voxclone is in active early development. Feedback and issue reports are welcome on GitHub.