Free Online Audio Spectrum Analyzer

What is an audio spectrum analyzer?

Every sound is a mixture of frequencies. A low hum and a cymbal crash can be equally loud while being made of completely different ingredients. An audio spectrum analyzer separates those ingredients and shows you what is actually in the sound.

Most audio software draws a waveform, the jagged shape that grows taller when things get louder. A waveform answers how loud. It cannot answer what. A hum, a hiss and a voice all look like roughly the same wiggle.

waveform: one flat wigglea voice and its overtoneshum along the bottom

A spectrogram answers the other question. It is the same audio drawn with three axes:

• Left to right is time, like a video timeline.
• Bottom to top is pitch: low sounds sit at the bottom, high sounds near the top.
• Brightness is loudness.

A three-minute recording becomes one picture you can read at a glance. A constant hum appears as a bright horizontal line near the bottom, running the full width. A click is a thin vertical stripe. Tape hiss is a faint haze across the top. Something you could only vaguely hear becomes something you can point at. (The word is sometimes spelled spectogram; it means the same picture.)

That matters the moment your ears run out of road. A podcast editor can see whether a buzz runs through the whole take or only part of it. Someone working out a bass line can read the notes off the screen instead of guessing. A singer can watch a held note wobble. An engineer can show a client that the noise is real rather than imagined.

How to use this frequency analyzer online

Four steps, all of them in your browser.

1. Open something. Drag an audio file or a video onto the page, open your microphone, or click one of the three examples. The file is decoded on your own machine; nothing is uploaded.
2. Read the shapes. Horizontal lines are sustained tones: a note, a hum, a whistle. Vertical lines are sudden events, like a click, a door, or the start of a word. A flat wash spread over the whole height is noise.
3. Zoom into whatever looks wrong. Scroll to zoom, drag to move around. The image is recalculated from the audio every time the view changes rather than being stretched, so it stays sharp however far in you go.
4. Point at it. Move over any spot and you get three numbers for that exact pixel: the frequency in hertz, the nearest musical note with its octave, and the level in decibels.

Step four is the one that turns a picture into an answer. Instead of something low and constant is sitting under this recording, you get there is a line at 50 Hz, that is mains hum. Now you know what to cut.

Three things worth knowing once you are in:

• A stack of evenly spaced lines is usually one note, not several. A voice or an instrument playing a single pitch produces a fundamental with overtones above it. Point at the lowest line with Mark lines of the same note ticked, and dashed markers land on the lines that belong to it.
• Open the microphone to check yourself as you go. The picture scrolls live, so you can watch whether a sung note holds steady or drifts.
• Export a PNG when you find something. The image carries its scales and the analysis settings, so it still means something to whoever you send it to.

Free online spectrum analyzer, or an installed app?

Search for a spectrum analyzer and most of the results are phone apps and desktop software. They are not worse. They are for different moments.

A browser tool wins when you want to look at something now. There is nothing to install, no account, and no permission request until you ask for the microphone. It works on a computer you do not own, like a studio machine or a shared laptop, and it leaves nothing behind when you close the tab.

A phone app wins for measuring a room, because the phone is the microphone you already have with you. Checking a rehearsal space or a noisy vent is its job, not this one.

A plugin wins inside a mixing session. If you are already in a DAW and want a spectrum on one channel, a plugin sits in the signal chain where you need it.

Desktop software wins for volume. A hundred files to check, or a format a browser cannot decode, is work for something installed.

What you give up here: no batch processing, no place in a plugin chain, and very large files are analysed at half rate so they fit comfortably in a browser tab. What you get: it opens in about a second, it costs nothing, and your audio never leaves your machine.