CoolFace
Apppublic

haze-ai/youtube-multimodal-mcp

sourceHugging Facemitupdated 4mo agoView on Hugging Face
0likes
App README

YouTube Multimodal MCP

Remote multimodal video analysis backend. Analyzes YouTube videos for:

  • —Transcript and subtitles (faster-whisper)
  • —Audio analysis (librosa: BPM, key, energy, mood)
  • —Visual style (PIL: colors, brightness, aesthetics)
  • —Scene detection (ffmpeg)
  • —Presentation structure and pitch scoring
  • —Storytelling arcs and hooks

REST API

MethodEndpointDescription
POST/api/analyzeSubmit video for analysis
GET/api/status/{job_id}Poll job status
GET/api/result/{job_id}Get complete results
GET/healthHealth check