Spectrogram Viewer

See a whole file's frequencies at once: hums, whines, encoding cutoffs, and everything your ears suspected. 2048-point FFT, log frequency axis, rendered locally.

Drop an audio file here, or click to choose

MP3, WAV, AAC/M4A, OGG - rendered locally, never uploaded

Reading the picture

Horizontal lines are sustained tones - a line at 50/60 Hz that runs the whole file is mains hum; thin lines high up are electronic whines. Vertical stripes are clicks and transients. A hard ceiling where all energy stops (commonly 15-16 kHz) is the fingerprint of lossy encoding - the fastest way to catch a fake "lossless" file. Stacked parallel lines are harmonics: voice, strings, brass - each instrument's spacing and brightness pattern is its timbre made visible.

For a live spectrogram of your microphone, use the spectrum analyzer; to measure the file's loudness numbers, the audio analyzer; and to learn the frequency map itself, the frequency chart guide.

Spectrogram FAQ

What is a spectrogram?

A picture of sound: time runs left to right, frequency bottom to top, and brightness shows how much energy each frequency has at each moment. A pure tone is a horizontal line, a click is a vertical stripe, speech is stacked harmonic bands, and hiss is diffuse fog. It is the standard way to see what your ears hear.

What can I find with a spectrogram?

The classics: mains hum (a hard line at 50 or 60 Hz), high-pitched whines from electronics, the sharp cutoff that betrays low-bitrate MP3 encoding (a hard ceiling around 15-16 kHz), bandwidth of a recording, ultrasonic content you cannot hear, and hidden images that artists have embedded in tracks (famously in Aphex Twin's "Windowlicker").

How do I spot a fake "lossless" file?

Look at the top of the picture. A genuine lossless or high-bitrate file has energy reaching toward 20 kHz that fades naturally. A file that was once a 128 kbps MP3 shows a razor-sharp horizontal cutoff around 15-16 kHz - re-encoding it as FLAC cannot restore what the lossy encoder deleted, and the spectrogram exposes it instantly.

What are the settings behind this view?

2048-sample FFT with a Hann window and 50% overlap, log-scaled frequency axis (matching how hearing works), 60 dB dynamic range mapped to brightness. Long files are downsampled in time to keep the image sharp; the frequency resolution is about 23 Hz per bin at 48 kHz.

Is my file uploaded?

No - decoding and the FFT run in your browser. Nothing leaves your device, and the page works offline once loaded. For a live view of your microphone instead of a file, use the spectrum analyzer.