Live audio spectrum analyser for your microphone or a sound file

Pick Microphone or Audio file and watch the frequencies in the sound drawn live as bars from 20 Hz to 20 kHz, with the strongest frequency and nearest musical note shown underneath; for files you can also average the whole track.

Live demo · real result from this tool
Source

Press Start to see the spectrum.

Processed on your device — nothing is uploaded.

How to use the spectrum analyser

Live from the microphone

  1. Under Source, choose Microphone and press Start.
  2. Allow microphone access when your browser asks. The page must be on https for this to work.
  3. Make a sound. Bars rise at the frequencies present, and the line below reads, for example, “Strongest: 440 Hz · A4 · −20 dBFS”.
  4. Change FFT size at any time; the display adapts while running. Press Stop to release the microphone.

From a file

  1. Choose Audio file and drop a file (MP3, M4A, WAV, OGG, FLAC, WebM, or a video’s sound).
  2. Press Play & analyse live to watch the spectrum as it plays, or Analyse whole file for an averaged result: Strongest frequency, Nearest note, Resolution, a line chart and a table of 32 bands you can download as spectrum.csv.

What it does and when to use it

A spectrum shows how much of each frequency a sound contains. Use it to find the pitch of a hum (mains hum sits near 50 or 60 Hz), check whether a microphone or speaker handles bass, spot a sharp cut-off in the highs that can point to a low-bitrate MP3 source, or teach how notes and overtones work.

Make a reference tone with the tone generator, see a recording’s shape over time in the waveform viewer, or check only your mic level with the microphone tester.

How it works

The live view uses your browser’s Web Audio analyser, which runs a Fast Fourier Transform on the latest block of samples (the FFT size). The page groups the result into 64 bars on a logarithmic scale from 20 Hz to 20 kHz, each showing the strongest bin in its range.

Analyse whole file decodes the file in a Web Worker on your device, mixes it to mono, and averages Hann-windowed FFTs of up to 2,000 blocks spread across the file. A full-scale sine reads 0 dBFS. The strongest bin above 20 Hz is refined by interpolation and named with A4 = 440 Hz. Resolution = sample rate ÷ FFT size.

Worked examples

From Analyse whole file on test signals at 44.1 kHz:

  • A 440 Hz tone with a quieter 1,320 Hz overtone, FFT 4096: Strongest frequency 439.9 Hz, Nearest note A4, Resolution 10.77 Hz per bin. In the band table the 458 Hz band reads −6.1 dBFS and the 1,346 Hz band −20.9 dBFS, matching the two tones.
  • A 60 Hz hum, FFT 16384: Strongest 60 Hz, Nearest note B1 (−48 cents), Resolution 2.69 Hz per bin.
  • The same hum, FFT 1024: Strongest 61.3 Hz, Resolution 43.07 Hz per bin. The coarse bins blur the low end, which is why low sounds need a large FFT.

Limits and tips

  • Microphones and speakers colour the result, so the display shows what your equipment hears, not an absolute measurement.
  • Whole-file analysis needs WebCodecs for formats other than WAV and has no ffmpeg fallback; AIFF or WMA must be converted with audio to WAV first.

Frequently asked questions

Is the microphone sound recorded or uploaded?
No. The browser asks for permission, then passes the live sound to this tab, which analyses it frame by frame and draws it. Nothing is stored or sent. Press Stop and the microphone is released.
Which FFT size should I choose?
Larger sizes separate close frequencies better but react more slowly. At 44.1 kHz, 1024 gives 43 Hz per bin and 16384 gives 2.7 Hz per bin. Use 1024–2048 for a lively music display and 8192–16384 to pin down hum or a tuning note.
Can I use it as a tuner?
Roughly. The strongest frequency and note name, such as 'A4' or 'B1 (−48 cents)', are a good guide for a single sustained note. Rich instruments can have a stronger overtone than the fundamental, so the reading may jump an octave or more.
Why do I see sound when the room is quiet?
Every microphone hears some noise: fans, traffic, electrical hum at 50 or 60 Hz. The analyser turns off the browser's noise suppression and echo cancellation on purpose so you see the real input.

Written by the ToolsRift team · Last updated

Pages are drafted with AI assistance and checked by automated tests. How we write and check pages.

Was this tool useful?

Report a problem

Start typing to search every tool.

↑ ↓ to moveEnter to openEsc to close