OllamaFrom the builder
Speech models in Ollama
Ollama can now turn text into speech, starting with Phonon, a family of small text-to-speech models from Gradium, available on macOS first with Windows and Linux to follow. Using MLX on Apple Silicon, Ollama says it ran 35 percent faster than Gradium's own runtime on an M5 Max MacBook Pro while matching the reference word error rate. Phonon covers English, French, German, Spanish and Portuguese with multiple voices and an OpenAI-compatible speech endpoint; the main model is Apache 2.0, while the 40M variants are limited to noncommercial use.
Ollama can now turn text into speech.
Saved . A shortened excerpt from the official blog; the original publisher retains authorship.