To generate an AI voiceover, paste your script into the tool above (up to 3,000 characters, about 500 words or three minutes), choose a voice and press Convert; with a cloud voice you can download the result as an MP3, and with an on-device voice in English, Spanish, Portuguese or French as a WAV file. It is free and needs no sign-up.
Free text to speech: paste up to 3,000 characters, pick a voice and listen in your browser. The voices load in a moment.
Write the script, paste it, choose the language and a voice, press Convert, then listen from the start.
If a sentence sounds wrong, change the sentence rather than fighting the voice.
When you are happy, press Download and drop the file into your video editor.
A voice marked as a browser voice is listen-only, so for a file you can save, pick a cloud voice (MP3) or an on-device voice (WAV).
Keep the script file next to your project, with a version number, so you can regenerate one scene when the edit changes.
Narration runs at about 150 words per minute, so a 60-second video needs roughly 150 words and one conversion of 3,000 characters covers around three minutes. For a longer video, write the script scene by scene, convert each scene separately and place the clips on your timeline. That way a change to one scene never forces you to regenerate the whole voiceover. Leave a little room: a voiceover that fills the video to the last frame feels rushed, so aim for about 90 percent of the running time.
Preview the first twenty seconds with two or three voices before you commit. A calm, steady voice suits tutorials and explainers; a brisker one suits tips and short clips. Under Advanced settings you can nudge the speed, and small changes work better than big ones. Once you pick a voice and a speed, keep them for every video in a series so viewers recognise it. One conversion uses one voice, so for a conversation between two speakers convert each speaker's lines separately and combine the clips in your editor.
Use short sentences, one idea each, and write the words the way they should sound: spell out symbols, write a year as you would say it and avoid long lists in parentheses. Use Insert pause for a deliberate beat before a key line. Listen once with your eyes closed; every place where the voice stumbles is a place where viewers will stumble too. Prices, times and dates can be read several ways, so if the voice picks the wrong one, write the words you want, such as twelve thirty instead of 12:30.
The tool reads text with cloud voices that download as MP3, with on-device voices that save as WAV and with the voices available on your device. Background music mixing is not available, so add music and cut silences in your video editor. The voices you see depend on your device and browser.
Yes. It is free, needs no sign-up and has no credit card step. Each conversion takes up to 3,000 characters.
Yes. Pick a cloud voice and download the MP3, or use an on-device voice in English, Spanish, Portuguese or French and press Download WAV. Browser voices are listen-only.
One conversion covers about three minutes. For longer videos, write and convert the script in scenes.
Not yet. Music mixing is not available, so add it in your video editor after you download the voiceover.
Cloud voices cover English, Chinese, Spanish, French, Japanese and Korean. English, Spanish, Portuguese and French can also be generated on your device, and other languages play with the voices your browser provides.
Paste it into the free tool, pick a voice and press play. Cloud voices download as MP3, free within the daily limits.