Audio Fingerprint Test
Audio fingerprinting uses the Web Audio API to generate a unique identifier based on how your browser processes sound internally. No microphone access required — the differences come from your OS, audio drivers, and hardware.
Audio Destination
DynamicsCompressor Defaults
Signal Analysis
Supported Audio Nodes ()
How Audio Fingerprinting Works
Audio fingerprinting has nothing to do with your microphone or speakers. It exploits how your browser's audio stack — the combination of your operating system, audio drivers, browser version, and hardware — processes sound signals internally. A triangle wave oscillator at 10,000 Hz is passed through a DynamicsCompressor node, and the resulting audio samples are analyzed. Microscopic differences in floating-point arithmetic, driver implementations, and audio processing pipelines produce unique output values for each system.
The waveform above shows the full rendered audio signal. The highlighted purple region (samples 4500-5000) is where the fingerprint is extracted — this section contains the most variance between different systems. The hash of these values creates a stable identifier that persists across sessions, even without cookies.
Research from Princeton University found audio fingerprinting scripts on major websites, contributing approximately 5.4 bits of entropy — enough to distinguish between roughly 42 different configurations. Combined with canvas, WebGL, and font fingerprints, it becomes a powerful tracking vector.