Skip to content

dsp-ark/graphverb

Repository files navigation

Graphverb

Graphverb is a JUCE-based audio plugin that performs dynamic, spectral-aware reverb processing using graph analysis and K-Means clustering.

GraphVerb


How It Works

  1. Spectral Analysis
    Uses a short-time Fourier transform (STFT) to extract magnitude information.

  2. Graph Construction
    A SpectralGraph is built from the spectral bins, treating each bin as a node.

  3. Clustering
    The graph is clustered via CommunityClustering (a simple k-means variant).

  4. Per-Cluster Reverb
    Each frequency cluster is passed through a reverb unit with its parameters determined by the energy magnitude of that cluster.

  5. Mixing & Output
    A weighted sum of wet reverbs (based on normalized cluster energy) is blended with the dry signal.


TODO

  • Replace k-means with spectral or graph community detection
  • Host automation support (DAW parameter control)

About

Spectral-aware Reverb Processing

Resources

License

Stars

Watchers

Forks

Packages

No packages published