Skip to main content

Resources

Free AI Voice Generator

A small text-to-speech tool you can use for anything, entirely free of charge. Paste your text, pick a voice, and click Generate — your browser turns it into audio entirely on your device. It's the same tool I use to record the audio versions of my essays.

It runs the open-source Kokoro model directly in your browser, using WebGPU when it's available. Despite its lightweight architecture, it delivers quality comparable to larger (and premium) models while being significantly faster — and free. Kokoro is distributed under the Apache-2.0 license.

Checking whether the voice model is ready…

How to use it

  • Paste your text, then pick a primary Voice from the dropdown — there are twenty-eight. For a hybrid, pick a second one under Blend with and adjust the mix.
  • Add a pause anywhere by writing [0.5s] in the text — change the number to whatever you prefer.
  • The first time, your browser downloads the Kokoro model (~300 MB), so the first run can take a minute or two. After that it's cached and fast — a couple of minutes of audio takes about as many seconds on a modern machine.
  • When it finishes, play it back, download the MP3, or grab the WAV.

Voice samples

Hear each voice

Every clip reads the same line: “Some mornings, the coffee tastes better than usual — and I never quite know why.”

American English

British English

Good to know

Browser. Chrome, Edge, and Brave support WebGPU, which is what makes this quick. Safari and Firefox fall back to a slower WASM mode — it still works, just be patient.

Privacy. Everything runs locally. No telemetry, no server calls, no third parties, no cookies. Your text and the generated audio stay in your browser and never reach a server.

Quality. Some voices are better than others, depending on how long they were trained. Blending two voices within the same language usually produces a more interesting result than a single one. I recommendHeart (Female, American) — it's the most trained of all.

Audio files. WAV is uncompressed PCM (24 kHz, 16-bit, mono, ~2.9 MB per minute). MP3 is encoded with lamejs (24 kHz, mono, 128 kbps CBR, ~960 KB per minute). The model itself only produces 24 kHz, so the WAV is speech-quality without lossy compression on top — not studio-quality.