IT Home October 27 news: Ollama, the open-source platform for running AI models locally, announced on October 24 thatits Mac app now officially supports text-to-speech (TTS) generation,with Windows and Linux versions to follow.
Tech media outlet 9to5Mac considers this the first time Ollama has added voice generation capability, taking local AI from being able to "read and write" a step further toward "being able to speak."
The first supported voice model is the Phonon series developed by Gradium. The model is small in size and technically advanced; users do not need to rely on cloud services and can download it directly through Ollama and run it on Apple Silicon Macs, making it more suitable for privacy-conscious, offline-use scenarios.
On macOS,Ollama uses Apple's open-source machine learning framework MLX, making full use of Apple Silicon's unified memory architecture to improve running efficiency.
Official tests show that on a MacBook Pro equipped with an 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 voice generation, so users do not have to wait for the AI to finish outputting the full text; they can hear the speech in sync as the answer is being generated, making the interaction experience closer to a natural conversation.
IT Home, citing the blog post, reports that Phonon currently offers standalone models for five languages — English, French, German, Spanish, and Portuguese — plus a multilingual model supporting all five; users can also choose voices with different accents, such as American English or British English.
Ollama is one of the most popular AI open-source projects on GitHub, with nearly 185,000 Stars and about 18,600 Forks; users can download and run open AI models locally, or connect to more powerful cloud models as needed.