Free Text to Speech

Text to speech turns written text into spoken audio. This tool uses the Kokoro-82M voice model in your browser to read an English script aloud and gives you a WAV file you can play and download.

1. Type or paste English text into the “English script” box. 2. Click Create result and wait while the model loads. 3. Play the audio in the result area. 4. Click Download to save the WAV file.

There’s one voice, Kokoro’s American English female voice, and it reads English only. Scripts that contain Chinese characters are rejected with a prompt to enter English text.

Each run accepts up to 1,000 characters. For a longer script, split it into parts, create a WAV for each one and join them in your video or audio editor.

You get a mono WAV file at 24 kHz named voiceover.wav. WAV works in most video editors, and if you need a smaller MP3, you can run it through an audio converter.

Not in this tool, which always uses the same voice at normal speed. You can adjust pacing with shorter sentences and commas, or change the speed later in a video editor.

No. The voice model runs in your browser tab, so your script is turned into audio on your own device. The model files (about 95 MB) are downloaded once, the first time you create audio.

It’s free and there’s no signup or login. Apart from the 1,000-character limit per run, there’s no daily character quota, so you can create as many clips as you need.

The voice comes from Kokoro-82M, an open model released under the Apache 2.0 license. Check the rules of the platform you publish on, since some ask you to label AI-generated voices or content.

The model guesses pronunciation from spelling, so names, brands, acronyms and symbols can come out oddly. Try spelling the word the way it sounds, putting spaces between the letters of an acronym or replacing symbols with words, then create the audio again.