audio-analysis

Reads real-time amplitude and 8-band frequency data from Decentraland SDK7 audio entities.

Updated Sep 11, 2026
One-click install
npx skills add https://github.com/ansonj-dev/neon-reverse --skill audio-analysis-ansonj-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audio-analysis
Source: https://github.com/ansonj-dev/neon-reverse/tree/main/agent/skills/audio-analysis
Command: npx skills add https://github.com/ansonj-dev/neon-reverse --skill audio-analysis-ansonj-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @dcl/sdk, and includes references (resource) components.

What problem does it solve? Building music visualizers, beat-reactive geometry, or audio-driven lighting in a Decentraland SDK7 scene requires access to live audio signal data, which the base AudioSource component does not expose. This Skill explains how to attach and read the AudioAnalysis component to obtain per-frame amplitude and 8-band frequency data from any audio-emitting entity. ## Core Features & Use Cases - Real-time signal reading: Use readIntoView or tryReadIntoView to pull amplitude and 8 frequency bands from entities with AudioSource, AudioStream, or VideoPlayer into a pre-allocated AudioAnalysisView. - Visualizer patterns: Canonical one-read-many-consumers system structure for driving equalizer bars, amplitude-pulsing meshes, lights, and particles from band data. - Mode and gain configuration: Choose between MODE_LOGARITHMIC (perceptual, gain-scaled) and MODE_RAW (linear FFT magnitudes), with tunable amplitudeGain and bandsGain. - Use Case: Build a club scene where a sphere pulses to overall volume and eight bars rise and fall with the frequency spectrum of a looping music track, using the included full reference example. ## Quick Start Ask the AI to add an AudioAnalysis component to an entity that already has an AudioSource and drive an equalizer bar scale from the band values each frame.

Frequently Asked Questions about audio-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a music visualizer in a Decentraland SDK7 scene?

Attach AudioSource and AudioAnalysis to the same entity, then call AudioAnalysis.readIntoView each frame into a pre-allocated view. Use view.amplitude for overall volume effects and view.bands[0..7] to scale equalizer bars or drive other visuals.

How do I read frequency band data from an AudioSource entity?

Create the component with AudioAnalysis.createAudioAnalysis(entity), allocate an AudioAnalysisView with an 8-element bands array once, and call readIntoView inside an engine system. The bands array fills with log-spaced frequency bins from sub-bass to treble.

Does AudioAnalysis work with VideoPlayer audio in Decentraland?

Yes, AudioAnalysis works on VideoPlayer entities when the video is a progressive MP4 or WebM file with a direct URL. HLS streams (.m3u8) play audible sound but write zeros to the component because their audio decodes on a path the analyzer does not tap.

Why does AudioAnalysis return zero values in my scene?

Zeros occur when the audio is paused, muted, or not yet loaded, since values come from live audio frames. They also appear on Bevy and mobile Godot explorers, which ignore the component entirely; only the Unity explorer writes analysis data.

What is the difference between MODE_LOGARITHMIC and MODE_RAW?

MODE_LOGARITHMIC maps bands to a perceptual log scale with amplitudeGain and bandsGain applied, suiting visualizers. MODE_RAW returns linear FFT magnitudes where low frequencies dominate and the gain settings are ignored.