Free to get started
Try a voice, make a take, and download it. No credit card needed.
Natural-sounding voiceovers, right in your browser. Type it. Hear it. Make it yours.
First time? A one-time model download gets you set up.
Compatibility mode — this may take a bit longer on your device.
Try a voice, make a take, and download it. No credit card needed.
Skip the forms and get to the good part. Your next voiceover is a few clicks away.
Runs in your browser. Nothing is uploaded. Your script stays between you and your screen.
FROM SCRIPT TO SOUND
No studio. No learning curve. Just you and your next idea.
Write a hook, tell a story, or give the example a spin.
Preview a few voices. Pick an accent. Set the pace.
Hit Generate, give it a listen, and download your MP3 or WAV.
Try scripts and tips for TikTok, Shorts, and your next big little idea.
Yep. You can make voiceovers for personal, non-commercial use, with up to 5,000 characters per take. No credit card needed. Paid plans with a commercial license are coming soon.
Free audio is for personal, non-commercial use. Commercial licenses will be included with paid plans, coming soon.
Try an up-to-date Chrome or Edge for the fastest experience on devices with WebGPU. Safari and Firefox can use compatibility mode. Speed depends on your device; keep this tab open while it works. Older phones may not have enough memory.
The voice model runs on your device. It needs a one-time download of about 90–330 MB, plus a few smaller runtime and voice files. Your browser caches it for next time, unless you clear site data or your browser frees up storage. On mobile connections or with Data Saver on, we wait until you click Generate.
No. Runs in your browser. Nothing is uploaded. Your script and generated audio stay on your device. Your browser connects to Hugging Face to download model files, but it never sends your script there. We don’t use analytics, ads, or tracking cookies.
This version is built for English voiceovers, with American and British voices. The Chinese interface makes the tool easier to use; it doesn’t change the language of the voices.