I'm working on my Open Source speaking clock (mostly for myself): https://github.com/jcubic/speaking-clock https://github.com/jcubic/speaking-clock It uses lo

I'm working on my Open Source speaking clock (mostly for myself): https://github.com/jcubic/speaking-clock https://github.com/jcubic/speaking-clock It uses local AI models for the voice.