I want to bring an AI-driven live-call translation system to life, which works web to mobile without downloading anything software or app to mobile as normal calls. delivered as companion web interface. There will 2 web dashboard, 1 for admin and 2 for user. Core functionality is clear: during any phone conversation—whether dialed through Asterisk, VICIDIAL, SIP carrier—the System should capture each side’s speech, convert it into real time in both native languages, run an AI engine such as Deep Gram, Google Voice AI, etc., and speak the translated audio back to the listener with minimal latency. Two-way, real-time translation is non-negotiable; speech-to-text, text-to-speech, and airtight VOIP integration all need to feel seamless to the user. On the technical side I expect: • Secure, scalable web application to manage language packs, call logs, user accounts, and billing hooks • SIP registration or API-level hooks that let me plug and play the system into any existing VOIP service without custom rewrites • Latency under one second per translation cycle in normal network conditions • Source code, build scripts, and deployment documentation so my in-house team can maintain the service after launch If you have hands-on experience with calling-app development, VOIP stacks, Asterisk dial-plans, or leveraging AI speech services, I’d love to see how we can push this forward together. Feel free to mention past projects or demos that prove you can achieve reliable, real-time performance at scale.