Add Intensity plugin: Audio dynamics classification

[Sangarshanan]

Summary

Adds a new Intensity_projucer plugin that uses Essentia's Intensity algorithm to perform real-time audio dynamics classification on an incoming signal.

The plugin follows the same structure and visual style as the existing RMS_projucer template

What it does

• Classifies the audio signal into one of three dynamic states: Relaxed (-1), Moderate (0), or Aggressive (1)
• Uses Essentia's Intensity algorithm internally.
• Displays the current classification as a labelled text state and colour-coded bar meter in a live UI.

Key design decisions

Accumulation buffer: Intensity is a composite temporal algorithm that runs PoolAggregator statistics (mean, dmean, dmean2) across multiple frames internally. Unlike RMS, it cannot be called on per-block DAW micro-frames. The plugin accumulates 4096 samples (~92 ms at 44.1 kHz) before triggering each classification.

reset() before compute() the algorithm's internal FrameCutter retains a read cursor between calls. Without an explicit reset, it stops reading audio after the first window. This is called on every processing cycle.

Discrete UI because the output is an integer with only three possible values, the continuous float display from the RMS template was replaced with a bold text label (RELAXED / MODERATE / AGGRESSIVE) that changes colour dynamically.

Tested on

• macOS (VST3 format, REAPER)
• Essentia built from source (build_external/)
• JUCE via Projucer (Xcode export)

UI