⚙️ Process Media
Upload audio or video and run the full interaction packet pipeline. Steps 6–10 are now wired: Voxtral transcription, pyannote diarization, librosa acoustic analysis, OpenCV visual observation, merge, and local LLM synthesis.
Adapter Status
Green = adapter is reachable. Configure backends via environment variables (VOXTRAL_BASE_URL, PYANNOTE_BASE_URL, etc.).
Loading…