About Neural Storm
How it works
A graph of nodes is generated using a force-directed layout algorithm seeded deterministically. Each node is assigned a frequency band and an activation threshold. During playback, nodes activate when their band energy exceeds their threshold, triggering a visual pulse and sending activation signals along connected edges with a configurable propagation delay. Edge signals are rendered as traveling particles along cubic Bezier curves. The force-directed layout updates slowly over time based on node activation patterns, causing the network to organically restructure. All rendering uses instanced meshes for nodes and line segments for edges.
Audio Features
- ▸ Frequency Bands
- ▸ Onset Detection
- ▸ Spectral Complexity
- ▸ Harmonic Content
Best For
- ▸ Experimental & Glitch
- ▸ Neuroscience Content
- ▸ Tech Brand Videos
- ▸ Intelligent Dance Music
MusicMind Engine
Tension Analysis (L4) fires high-voltage arcs when harmonic surprise peaks, Timbral Fingerprint (L10) assigns each node a unique spectral identity for frequency-specific activation, and Temporal Memory (L12) shapes signal propagation speed based on the song's tempo evolution.
Learn how MusicMind analyzes music →Science Behind This Effect
Neural Storm's electrical arc patterns fire on detected chord changes, with arc intensity proportional to the harmonic surprise — unexpected chords create bigger storms.
Learn more: Tension & Resolution Science →