Real-time speaker diarization and recognition system combining pyannote.audio for speaker segmentation, faster-whisper for transcription, and scikit-learn for speaker matching using cosine similarity on voice embeddings. Features live recording capabilities, automatic speaker enrollment from audio segments, retroactive speaker identification that updates past segments, and GPU acceleration with CUDA support. Exposes tools for identifying unknown speakers, managing speaker profiles, searching conversation history by speaker, and accessing real-time transcripts with speaker labels.
Cognium trust score
0%
Tier
Scanned
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
Last scanned 2026-09-28.
Returns 7 tools: search_skills, get_skill, list_leaderboard, get_trust_breakdown, resolve_composition, plus the ChatGPT-connector search and fetch. Every tool is annotated read-only.
Resolve this skill directly via MCP tools/call get_skill.