Outmake Beta Download Beta
Voice models

Kitten TTS on Mac: voices, downloads and setup

Kitten Nano and Mini offer eight English voices each. Outmake’s curated packages are 31 MB and 68 MB.

Outmake guides · Published

Published by Ambedo Labs, the company behind Outmake. Sources and limits are linked below.

When to choose Kitten TTS

Start with Nano if download size is your main constraint. Listen to the same passage in Mini before deciding whether the larger package is worth keeping.

Nano download
31 MB, int8 v0.8
Mini download
68 MB, v0.8
Voices per model
8
Language
English
Model licences
Apache 2.0

Where shown, download sizes are rounded from the app’s pinned packages. They exclude the shared speech engine and do not measure runtime memory.

Nano and Mini are separate downloads

Outmake lists Kitten Nano and Kitten Mini as separate models. Nano is the smaller package. Mini is a larger alternative from the same maker.

The app’s engine exposes eight speakers for each of these versions. Outmake labels them Voice 1 through Voice 8. These numbers identify the engine’s speakers; they do not rank them.

Use the supported package

The curated versions run through sherpa-onnx. They should not be confused with arbitrary MLX conversions carrying the Kitten name.

Outmake’s MLX voice browser excludes the Kitten architecture because the conversion checked by the app used voice files the runtime could not load. That does not prevent the curated sherpa-onnx packages from being offered.

Compare with your own text

After downloading a model, play a sentence in several of its voices. Then try a paragraph with the punctuation and names you expect to hear.

Download size alone cannot tell you how a voice will sound. This guide does not claim Mini beats Nano in a listening test, or give a speech-generation time that has not been measured.

Run Kitten TTS in Outmake

Outmake is free and requires macOS 26 or later on Apple silicon. These voice models run locally after their files have downloaded. You can also follow the upstream project links below to use another runtime.

  1. Open Models, then Voices.
  2. Open Kitten Nano or Kitten Mini. Check the licence shown on its card.
  3. Choose Get. Once downloaded, play a sentence in the voice’s page.
  4. Choose Use to make it the local voice. To hear a reply, open its menu and choose Read Aloud.

Voice settings and read-aloud help explain how to choose voices and stop playback. This page is about generating speech. For speech-to-text, use dictation or transcription instead.

Before you rely on it

Try a short passage before committing to a long one. Local speech keeps synthesis on your Mac; it does not make an online chat model local. Outmake’s privacy policy explains the difference.

Sources

Model cards and runtime documentation checked 11 November 2026. App-specific details were checked against Outmake’s voice catalog and runtime configuration. No new listening test or speed benchmark was performed for this guide.