BPM Detection with JavaScript
summary
A JavaScript pipeline implements onset detection via short‑time Fourier transforms, four detection functions (spectral difference high‑frequency content phase deviation euclidean distance), peak picking and tempo from inter‑onset intervals, and runs in the browser using dsp.js and Web Audio. It reliably finds percussive beats, picks up tempo dips when analyzed in 10 second windows, runs in seconds (comparable to MATLAB on short clips), but stresses Chrome and benefits from web workers and maybe asm.js.
AI-written summary · about
chapters (15)
- 0:00Start
- 3:03IN THE BROWSER. WHY?
- 6:44WHAT IS BPM? WHAT IS A BEAT?
- 8:02DETECTION METHOD.
- 9:53DETECTION FUNCTIONS.
- 11:21BPM CALCULATION.
- 11:53MUSIC CHARACTERIZATION
- 14:03MODULES
- 14:34JS CHALLENGES
- 17:16STFT. SHORT-TIME FOURIER TRANSFORM
- 18:57PEAK PICKING
- 19:53OPTIMIZATIONS & PERFORMANCE
- 22:34WEAKNESSES
- 25:06THE WEB IS CHANGING
- 25:57RETHINKING WEB ENGINEERING
related talks
31mBeats in the Browser - Coding Music with JavaScript
26mBeats in the Browser - Coding Music with JavaScript
44mBeats in the Browser - Coding Music with JavaScript
30mJavaScript Patterns For Contemporary Dance Music
24mDrum 'n' JS
32mBeats in the Browser
01h 10mMusic 101 with JS Bach
40mBeats in the Browser - Coding Music with JavaScript