Hi, I'm Cosmas, building voice help for older users in Africa

I'm Cosmas Ondino, a developer in Nairobi. I'm working toward an MA in Philosophy at the University of Nairobi while studying computer science at the University of the People and software engineering at PLP Academy. I've also volunteered in policy and research with NGOs and taught philosophy to undergraduates.

I've taken part in three hackathons, using AI to look at problems facing much of Sub-Saharan Africa. My main project is SautiCare, a Python voice agent for older people who are new to smartphones, especially when they're alone in an emergency. It's designed to use local African languages. I'm now updating an earlier hackathon build after an invitation from Anna AI OS. For coding, I work mostly with Claude, Gemini, and ChatGPT.

What I'm still figuring out is how to make the voice interaction reliable across languages and accents. If anyone here has tested speech recognition or text-to-speech tools with African languages, I'd love your recommendations.