Audio analyzers

Audio analyzer tools

These analyzers turn sound into pictures you can read: which frequencies are there, how loud they are and how they change. Start with the live spectrum analyzer, or pick a more focused tool below. All of them run in your browser, and your sound never leaves your device.

Which audio analyzer do I need?

Each tool answers a different question about the same sound. Pick the one that matches yours.

  • “What is in this sound right now?” Use the live spectrum analyzer. It shows one moment: every frequency from 20 Hz to 20 kHz and how strong it is. Pluck a guitar’s open A string and you see a peak at 110 Hz, with smaller ones at 220 and 330 Hz.
  • “When did that sound happen?” Use the Spectrogram. Time runs across the screen and colour shows level, so a bird chirp that rises from 3 to 6 kHz in a fifth of a second shows up as a bright slanted line.
  • “Is my room or speaker balanced?” Use the Real-Time Analyzer (RTA). It sums the sound into 31 bands of 1/3 octave. With pink noise playing, a band that sits 8 dB above the rest is the one to cut with EQ.
  • “What does the wave look like?” Use the Online Oscilloscope. It draws the wave itself, so you can see flat tops from clipping or count that a 1 kHz tone repeats every 1 ms.
  • “What is inside this recording?” Use the Audio File Spectrum Analyzer for the average spectrum of a whole file, its noise colour and its highest real content.

Three more tools look closer at one question each: the Harmonics & Overtone Analyzer for the harmonics of one note, the Voice Formant Analyzer for vowels, and Compare Two Audio Spectra for checking a mix against a reference.

Frequency, time and bands: three ways to see sound

A microphone only measures air pressure going up and down, thousands of times a second. To see what that sound is made of, an analyzer cuts it into short pieces and runs a Fast Fourier Transform (FFT) on each one. The FFT splits the piece into narrow slices of frequency called bins. With 8,192 samples at 48 kHz, each bin is 5.9 Hz wide.

Spectrum: one moment 110 Hz lowhigh Hz Spectrogram: over time whistle 2 notes 0 stime RTA: bands for EQ boomy band lowhigh bands
The same kind of sound seen three ways: one moment as a spectrum, seconds of sound as a spectrogram, and octave bands as an RTA.

Frequency. A spectrum draws one FFT as a line: low notes on the left, high on the right, louder higher up. It is the sharpest view for finding an exact pitch or a hum at 50 or 60 Hz.

Time. A spectrogram puts many spectra side by side and turns level into colour. There is a trade-off: a long FFT gives fine frequency detail but smears quick sounds, because 8,192 samples last 171 ms. A short one shows clicks sharply but blurs pitch.

Bands. An RTA adds the bins into octave or 1/3-octave bands, which grow wider as the pitch rises, much like hearing does. The band around 100 Hz is about 23 Hz wide, while the one around 10 kHz is about 2.3 kHz wide. That is why pink noise, which has equal energy in every octave, reads flat on an RTA.

Getting good results from a phone or laptop mic

Built-in mics are better than most people expect, if you help them a little:

  • Watch for voice processing. The tools ask the browser to switch off auto gain and noise cleaning. Some phones keep them on anyway, and a badge tells you when the level is being changed for you.
  • Know the limits. Phone mics fall away below about 50 to 100 Hz, and headset mics often stop near 8 kHz. Readings outside that range will look low.
  • Keep it steady and close. Hold the phone still, about 30 to 50 cm from the source, with the mic hole facing it. On a laptop, fan noise shows up as low humps, so test in a quiet moment.
  • Do not clip. Keep the peaks below about −6 dBFS. If the clipping badge lights up, move back or turn the source down.

FAQ

What is an audio analyzer?

A tool that measures sound and shows it as a graph: which frequencies are present, how loud they are and how they change.

Spectrum or spectrogram?

A spectrum shows one moment. A spectrogram shows the last seconds as a picture, so you can see sounds start and stop.

What is an RTA used for?

An RTA groups sound into octave or 1/3-octave bands, which is the easiest view for room and speaker EQ.

Can I analyze a song from YouTube?

On Chrome or Edge desktop choose “Tab audio” as the source and share the YouTube tab. On phones play it through a speaker and use the mic.

Written and tested by the SpectrumAnalyzer.app team · Last checked