Ajamall family site · Sound and music tools
Home › Mic & Measure › Spectrum View
// mic

Spectrum View

Shows microphone sound as live frequency bars (FFT) and reports the frequency and note name of the strongest peak.

What you enter and what your microphone hears are processed only in this browser. They are not sent to a server or stored. Sound only starts when you press a button. Start at a low volume. Results are for reference only.

How to use it

  • Press "Turn on mic" and allow access.
  • Make a sound; bars move from low (left) to high (right) frequencies, and below them you see the strongest peak frequency and the nearest note name.
  • A larger "FFT size" separates frequencies more finely but reacts more slowly. "Axis" chooses log (musical spacing) or linear, and "Max frequency" sets the visible range.
  • Press [Freeze] to hold the picture, and "Stop" to turn the mic off when done.

How the FFT bars are made

The microphone waveform is cut into short blocks, multiplied by a window function, transformed with the FFT (fast Fourier transform) to get the size at each frequency, and shown in dB. The bar values on screen come from the browser's analyser (AnalyserNode); the 96 bars group neighbouring frequency bins and draw the maximum inside each group. The log axis divides the range from 20 Hz up to the max frequency into equal ratios.

Peak frequency calculation

A parabola is fitted through the dB values of the largest bin and its two neighbours to estimate the position between bins. At 44,100 Hz and FFT 4096 the bin width is about 10.8 Hz, which would be coarse without interpolation. Noise, harmonics and the window can still make it differ from the true pitch, so treat it as a reference.

FAQ

What does a spectrum show?

It splits sound into pure tones of many pitches (frequencies) and shows how strong each one is. Higher to the right, and a taller bar means more of that frequency. The same note looks different on different instruments and voices because the pattern of harmonics (whole-number multiples of the frequency) differs, which is much of a sound's "color".

The peak frequency is different from the note I played.

The strongest component can be a harmonic instead of the fundamental (common with voice and wind instruments). For pitch, the "Pitch Detector" tool is a better fit. Also, background noise is possible, and anything below 20 Hz (wind, vibration) is skipped when looking for the peak.

What is the FFT size and how do I choose it?

It is the number of samples analyzed at once. Bin width = sample rate / FFT size, so a larger size gives narrower bins that separate close notes better, but each analysis covers a longer stretch of time, so quick changes look blurrier.

Is my voice or the sound around me sent to a server or stored?

No. Sound is analyzed only inside your browser on this page, is not sent to a server and is not stored. The microphone turns off when you stop or leave the page. Your browser or operating system will still show its own "microphone in use" indicator while it is on.

Sources and check date

This tool uses the plain definitions of mathematics and music theory and does not rely on an outside reference document. The definitions this site uses are written on the glossary page.

Date we opened the official documents ourselves: 2026-11-12. Standards and conventions can differ between references, so read the original again before an important decision.

How we calculate · Glossary · Sources