Ollama's Mac app adds offline text-to-speech with Gradium's Phonon

According to IT之家, Ollama announced on October 24 that its Mac app now officially supports text-to-speech (TTS) generation, with versions for Windows and Linux to follow. Tech outlet 9to5Mac views this as the first time Ollama has added speech generation capability, extending its local AI platform beyond text input and output. The first supported model is the Phonon series developed by Gradium. It is described as compact with advanced technical metrics. Users can download it directly through Ollama and run it on Apple Silicon Macs without relying on cloud services, which fits privacy-conscious and offline-use scenarios. On macOS, Ollama uses Apple's open-source machine learning framework MLX, taking advantage of Apple Silicon's unified memory architecture for efficiency. Official tests show that on a MacBook Pro with the M5 Max chip, Phonon maintains a word error rate comparable to the Hugging Face Transformers reference implementation while generating speech 35% faster than Gradium's own implementation. The new feature also supports real-time speech generation, allowing users to start listening without waiting for the full text output. Current status: the capability is live in the Ollama Mac app, while the Windows and Linux versions have not yet followed.