He said the scripts built a WebAudio graph that introduced a sawtooth oscillator to generate a waveform, an analyzer to measure the result after the waveform passes through a browser’s audio implementation, and a script to read the associated frequency data.

This is a fairly interesting method of fingerprinting. Annoying too, as it hogs the audio channels as experienced by the discoverer

It’s actually an old one.

Add it to the pool of a thousand others.