Pick an English voice, paste your text and download the audio — the voice model runs on your device, so your text stays private.
Up to 4,000 characters of English. Longer texts are read in parts.
Pick from US and UK, female and male voices, and set the speed.
Listen to it, then download a WAV or an MP3.
The model runs in your browser, so your script never leaves your device.
US and UK, female and male, with adjustable speed.
Download lossless WAV or a small MP3.
The model is fetched when you press Generate, then cached.
Voiceovers for short videos, narration for slides and tutorials, a quick audio version of a script, or a way to hear how your text sounds. It reads English only and works best on clear, well-punctuated sentences.
Write the way people speak: short sentences, commas where you would pause, and full stops at the end of each thought. Spell out numbers, abbreviations and unusual names the way you want them said. If a word is mispronounced, try writing it phonetically.
Female and male, American and British voices are available. Speed 1.0 is natural; 0.85–0.95 suits tutorials and explanations, and 1.1 works for fast social clips.
WAV is lossless and best if you are going to edit the audio. MP3 is smaller and fine for sharing and for adding to video. You can add music or trim silence with the other audio tools.
Do not present generated speech as a real person’s voice. Follow the rules of the platform where you publish — many require you to label AI-generated audio — and respect the rights to any text you read.
The voice model runs on your device, so nothing you type is sent to a server. The first run downloads about 90 MB, which is cached afterwards.
Resize, crop, convert and clean up images, audio and video — all in your browser.
Browse all tools →