Ollama adds local text-to-speech for Apple Silicon Macs
Technology · First reported 26 Oct, 22:49 · Updated 07:54

Ollama version 0.40.11 adds Phonon text-to-speech models on Apple Silicon Macs, with speech available in five languages. The reports describe real-time playback and say Windows and Linux support is planned, but no release date is given.
- Ollama added text-to-speech support through Phonon, a family of models developed by Gradium, for Apple Silicon Macs running version 0.40.11 or later.
- The models support English, French, German, Spanish and Portuguese, with separate language models and a multilingual option.
- Speech can begin as text arrives, including while a chat model is still generating its response.
- Ollama says its MLX-based implementation was 35% faster than Gradium’s runtime in tests on an M5 Max MacBook Pro; Notebookcheck notes the comparison used different model precision.
- The English model is available under Apache 2.0; the other language models are reported as restricted to non-commercial use. Windows and Linux support is expected, but no date is specified.
Summary written by Lux AI from 3 articles
Key facts
- Minimum version
- Ollama 0.40.11
- Supported languages
- English, French, German, Spanish and Portuguese
- Mac hardware
- Apple Silicon
- Speed figure
- Ollama reports 35% faster speech generation than Gradium’s runtime in an M5 Max MacBook Pro test
- Other platforms
- Windows and Linux support is planned; no date is stated
Source spectrum
Left 0 · Centre 3 · Right 0 of 3 rated sources
Images from the coverage
All coverage
Coverage over time
LeftCentreRight
Articles per hour since first reported, by side.
Ownership
- Corporation 67% (2)
- Unknown 33% (1)
Share of the 3 sources by owner type.

