Text to Speech Voiceover

What this does

Type or paste a script, pick from over three hundred voices in more than seventy languages, and download the result as an MP3. There is no account and no fee.

Where the text goes, honestly

Every other tool on this site works entirely inside your browser. This one cannot, because speech synthesis needs an engine, so your text is sent to our server, turned into audio, and discarded. The audio file itself is deleted automatically a few hours after it is made. Nothing is kept, and nothing you type here is used for anything except making the file you download.

The voices come from two engines, and the difference matters for privacy. Six voices are synthesised entirely on this site's own server. The rest use a Microsoft neural speech service, which means for those voices the text is relayed to Microsoft to be spoken and the audio comes back; their privacy terms apply to that step. If you would rather your text never leave this site's server, use one of the six server voices, which are the first ones listed under English (United States) and English (United Kingdom).

Getting a natural read

The engine reads punctuation, so the fastest way to improve a stiff result is to edit the script rather than fight the voice. Short sentences read better than long ones. A comma is a breath, a full stop is a pause, and a paragraph break is a longer one. Spell out anything ambiguous: 2026 reads as a year, but a model number like X100 is often better written as X one hundred.

Numbers, currencies and units are read sensibly in most cases, but if a particular phrase matters, listen to that sentence before building an hour of audio around it. The character limit is generous enough to test a paragraph first.

What the voices are

These are neural voices, synthesised from trained models, and the page labels them as exactly that. They are not recordings of voice actors, and this tool does no voice cloning: it cannot imitate you, a celebrity, or anyone else. That is a deliberate line, not a missing feature.

What you may use the audio for

The speech generated from your own script is yours to use, including in videos you monetise. The usual boundary applies to the script itself: it has to be text you wrote or have the right to use. Reading someone else's article aloud does not make it yours.