Exploring Mojo for Real-Time AI Voice Call Solutions
Dear Mojo Community,
I hope you’re all doing well! My name is Juan Manuel Castaño, and I’ve recently started exploring Mojo programming language, captivated by its potential for high-performance computing. I’m working on a project that involves building an AI-driven solution for real-time voice calls.
Given the critical need for low latency and high-speed thread connections in this setup, I’m curious to learn if Mojo could help optimize this process. Specifically, I’d like to know if Mojo is suitable for enhancing the speed and efficiency of thread management in such a real-time environment.
I’d be incredibly grateful for any guidance, suggestions, or resources from this amazing community. If anyone has experience or ideas to share, I’d love to hear from you!
4 Replies
Yes Mojo would be a very good fit for that use case. However, it is still in its very early stages. I would not recommend building any company or product around it just yet. We don't have networking nor very good multi threading, and stdlib is still rapidly changing and evolving (breaking changes are very normal currently)
Thank you martin!
I'm campaigning for DPDK to be more or less part of the standard library for networking, so "low latency" should be doable unless you need times on the order of <100 nanoseconds.
Welcome to the Mojo community, Juan Manuel. I am also interested in real-time voice calls backed by AI, and I actually work on a company that provides these kinds of products, but using Python instead of Mojo. In my spare time, I try to play with Mojo, but my approach is just using websockets and rely on the big AI companies underneath. I initially started the mojo-openai-realtime-api repo but I then realized that I don't want to rely on Python websockets but a Mojo alternative, so I started mojo-websockets, but I cannot put a lot of hours per week yet so it's evolving pretty slow.