speak - MadBomber/aia GitHub Wiki

The --speak option is intended to audibly say the response returned by the LLM process. At this time it is an experimental option only working on the MacOS platform using the say command line utility. The intent is to take advantage of the more powerful text-to-speech tools available in the generative AI.

For now its Siri and the default options you have set in your system preferences.