We are looking for an experienced AI developer to build a system that continuously listens to conversations inside a specific room and generates a structured summary at the end of each day. This is not a basic transcription task. The system must: * Record and process audio throughout the day * Transcribe conversations accurately * Identify and distinguish between different speakers * Match voices to predefined names (speaker recognition) * Generate a clean daily summary of what was discussed * Optionally highlight key topics, decisions, conflicts, or repeated issues Core Requirements: * Strong experience with speech-to-text systems * Speaker diarization and voice identification * Ability to train the system to recognize specific individuals by voice * Automated daily report generation * Privacy-aware architecture and secure storage * Ability to deploy locally or on a secure server Nice to Have: * Experience with Whisper, Pyannote, or similar speech models * Experience with LLM-based summarization * Real-time processing capability * Dashboard interface for reviewing transcripts and summaries Deliverables: * Fully working AI audio monitoring system * Speaker recognition trained on specific individuals * Automated end-of-day summary report * Documentation for setup and scaling Please include: * Similar projects you’ve built * What tech stack you would use * Estimated timeline * Fixed price