We are seeking an experienced AI engineer to build a voice-based AI system for a low-resource language. The system will support audio-only interaction (no text interface) and must process voice input and return voice responses (text interface may be added later). This is a feasibility and MVP project (MVP = Minimum Viable Product, the simplest working version of the system used to test if the idea works). Phase 1 Scope: • Evaluate existing speech-to-text models for a low-resource language • Build an end-to-end audio pipeline (Audio → AI → Audio) • Integrate speech recognition with LLM and text-to-speech • Provide feasibility report and recommended architecture Note: This project was previously posted from a personal account and has now been moved to our business account for collaboration purposes.