Text to speech for YouTube voiceovers
Narrate explainers, top-10s, tutorials, and faceless channels without recording a word. Scripts stay on your machine, and a ten-minute video doesn't cost more than a sentence.
One-time download, stored in your browser and reused offline. Manage or remove it any time on the Models page.
Preloaded with Kokoro-82M — The quality benchmark for in-browser speech.
Long scripts don't cost extra
Cloud TTS charges by the character. Here the model runs on your computer, so a documentary cut costs the same as a short intro: nothing.
WAV that drops into the editor
Download uncompressed audio and put it on the timeline in Premiere, Resolve, or CapCut — no watermark.
Same narrator every episode
Deterministic models stay consistent. Episode 40 can sound like episode 1.
YouTube videos — questions
Can I monetize YouTube videos that use these voices?
Licensing depends on both the model and the selected voice or source dataset. Review the current linked terms and any attribution requirements before monetizing the result.
How long can my script be?
About 5,000 characters per run (roughly 6–7 minutes). For longer videos, generate section by section and stitch the WAVs.
Will YouTube flag AI narration?
Platform rules change and can depend on the content. Review YouTube's current synthetic-content and disclosure rules before publishing.