Two-person AI duet video generator in an orange recording booth
Reverse-engineer any video, shot by shot, to get a production blueprint
Stage
Not given
Not given
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Paid
Open source
Pricing
Not given
Open-source and free to use offline. Optional third-party vision models (Gemini or Doubao) available for improved accuracy on re-compressed files; costs depend on the model provider's pricing, not Rea Video.
Key features
Upload two photos to generate a duet scene
Matte-orange recording booth setting with one microphone
AI scene composition using GPT Image 2
Motion transfer animation preserving original performance and audio
Vertical 9:16 MP4 output with audio
Generation time of 3–5 minutes
Optional styling notes for scene direction
Automatic credit refund on failed generation
Choice of 8-second or 15-second video length
No subscription required, pay-per-generation model
Shot-by-shot production classification into 7 categories with confidence scores
Reverse-engineered prompts for AI-generated shots
Filming notes with camera, lens, and lighting setup details
Caption extraction with timings exportable as SRT/ASS
Audio analysis: voice, music, BPM, and sound-effect placement
98%+ shot classification accuracy on original files
Optional vision model integration for re-compressed files
Support for multiple video types: live-action, AI, 3D, code, avatar, screen recording, mixed
What makes it different
Preserves identity and recognizability of both performers across the duet
Enables duets between people in separate photos without manual editing
No subscription required; credits spent only when videos are generated
Video saved automatically to account; no other apps or accounts needed
Provides actionable remake plans instead of just binary AI/real labels
Hotel Lobby AI
Rea Video