NeuralPrint
liveSee your brain react to anything.
View live demo ↗
Deep LearningNeuroscience
About
NeuralPrint takes any stimulus - a YouTube link, video, audio, or text - and uses a real neural encoding model (TRIBE v2) to predict how 368 brain regions would respond, millisecond by millisecond, across vision, language, and sound. The predicted activity renders as an interactive 3D brain.
Key features
- Submit video, audio, or text
- Predicts 368 brain regions
- TRIBE v2 neural encoding model
- Millisecond-level BOLD response
- Interactive 3D brain visualization
Technologies
Next.jsThree.jsFastAPIModal GPU
How it works
- 01A stimulus - video, audio, or text - is submitted.
- 02The content is encoded using the TRIBE v2 neural encoding model.
- 03BOLD responses are predicted across 368 brain regions.
- 04An interactive Three.js brain renders where and when activity occurs.
Why it matters
A multimodal deep-learning build - vision, language, and audio modelled together against measured brain responses.
ML research made interactive, not a CRUD app with an API bolted on.
Want something like this?
Same approach, scoped to your problem.