SpecEnvoy
EN·中文

Your Voice's Spectrum

Everything so far used made-up signals. This one uses you. Your phone is doing a real Fourier transform thousands of times a second — point it at your own voice and watch the frequencies light up live.

See the sound you're making

Tap Start, allow the microphone, then hum a steady note. The graph is the spectrum of whatever your mic hears: frequency runs left to right, and the height shows how much of each frequency is present right now. The tallest peak is your pitch — we'll name the musical note for you.

Hum, sing or whistle 🎤

Nothing is recorded or sent anywhere — the audio never leaves your device; it's analysed in the browser and thrown away. Tap Stop to release the mic.

What you're looking at

One note, many spikes

Hum a single steady note and you won't get one spike — you'll get a tall one plus a row of evenly spaced shorter ones. Those are harmonics: the pitch you hear is the lowest, and the rest are what make your voice sound like you and not a flute.

This is Build-a-Wave, backwards

In Build a Wave you added harmonics with sliders to make a shape. Here the transform reads them straight off a real sound. Same recipe of sines — measured instead of chosen.

Try it: whistle and you'll see almost a single clean spike (a whistle is nearly a pure sine — few harmonics). Then sing "aah" and watch a whole forest of harmonics appear. Slide your pitch up and the whole pattern marches to the right. You're watching ĝ(f) update in real time on your own voice.

This is the same maths as the Winding Machine — your phone just does the winding-and-centre-of-mass millions of times a second, for every frequency at once. That's an FFT (Fast Fourier Transform), and it's running in your pocket right now.