When to choose Pocket TTS
Use Pocket if you want spoken replies with the default local setup. Choose another model if its supported language or pronunciation suits your text better.
- Outmake model
- Pinned MLX conversion
- Model download
- 236 MB
- Voices offered by Outmake
- 12
- Language in this version
- English
- Model licence
- CC BY 4.0
Where shown, download sizes are rounded from the app’s pinned packages. They exclude the shared speech engine and do not measure runtime memory.
The model and the voice are separate
Pocket generates speech using a voice representation. Outmake includes twelve prepared voices rather than asking you to upload a recording. The model files download when needed.
The names in the app include Nora, Wren, Miles, Avery, Finn, Mia, Rory, Jack, Ellis, Peter, Sage and Quinn. Their source recordings come from Kyutai’s donated voices, the VCTK corpus and the Voice-Zero collection.
The model has its own licence. Each prepared voice also carries the licence of its source recording, shown in Outmake’s voice information. Keep both sets of terms in mind if you use speech in something you publish.
English in Outmake’s pinned version
The Pocket entry inspected for this guide points to a specific MLX conversion and is listed as English. Newer upstream Pocket releases can have different language coverage.
Use the app’s version and voice information when deciding whether your text is supported. An upstream announcement does not update the model files installed on your Mac.
Check names before reading a long passage
Install the model, choose a voice and play a sentence with the names you need. Check the final words as well as the opening. A model can produce fluent speech while misreading a proper noun.
Outmake offers Pocket as a local voice. Selecting an online provider such as ElevenLabs is a separate choice with that provider’s own processing and billing.
Run Pocket 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.
- Open Models, then Voices.
- Open Pocket. Check the licence shown on its card.
- Choose Get. Once downloaded, play a sentence in the voice’s page.
- 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
- The twelve voices are Outmake’s selection, rather than a promise that the upstream model has only twelve.
- This page does not promise arbitrary voice cloning in Outmake.
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.