How to Make a Spectrogram of an Audio File
Updated 2026-06-21
To make a spectrogram of an audio file, open the file in a browser-based analyzer that runs an FFT on the sound, then read the resulting plot: time runs left to right, frequency runs bottom to top, and color shows loudness. You can do this instantly with the Audio Spectrogram & Waveform tool — drag in an MP3 or WAV, and the spectrogram renders locally. Nothing is uploaded.
Make the spectrogram in four steps
- Open the Audio Spectrogram & Waveform tool and drop in an audio file (MP3, WAV, and other common formats work) — or click to use your microphone for a live view.
- Play the clip. The waveform shows overall loudness over time, while the spectrogram below it shows which frequencies are present at each moment.
- Switch between a logarithmic and linear frequency scale. Log scale spreads out the low and mid frequencies the way human hearing does and is better for music and voice; linear scale gives equal spacing and suits technical analysis of high frequencies.
- Hover anywhere on the spectrogram to read the exact value at that point — frequency in Hz, the nearest musical note, and intensity in dB.
How to read what you see
A spectrogram maps three dimensions at once. The horizontal axis is time, the vertical axis is frequency, and brightness or color is amplitude (how loud that frequency is at that instant).
- Horizontal bands are sustained tones — a held vocal note, a synth pad, or a steady hum.
- Evenly stacked horizontal lines are harmonics: a fundamental pitch plus its overtones, the signature of voices and most instruments.
- Vertical streaks are transients — drum hits, consonants, clicks, or pops.
- A bright wall that stops at a hard ceiling (often around 16 kHz) is a telltale sign of a low-bitrate MP3 or a re-encoded file, since lossy compression discards the highest frequencies.
The color map matters for reading detail. Perceptual maps make small changes in loudness easier to see than a plain rainbow, so quiet harmonics and noise floors stand out.
Practical uses and exports
- Check audio quality. Compare a file against a known-good source to spot the frequency ceiling of a re-encoded or upsampled track.
- Find a pitch or note. Hover over a tone to get its frequency and nearest note — useful for transcribing, tuning, or identifying a hum's source (a 60 Hz or 50 Hz line often means electrical interference).
- Inspect a recording. Locate background noise, clipping, or a problem frequency before editing.
When you are done, export the spectrogram as a PNG image to share or annotate, or export the underlying data as a CSV for your own analysis in a spreadsheet or script.
Because the entire process — decoding, the FFT, and rendering — happens in your browser, your audio never leaves your device. There is no upload, no account, and no server processing, which keeps private recordings, unreleased music, and sensitive audio private.
Ready to see your sound? Open the Audio Spectrogram & Waveform tool and drop in a file to make your first spectrogram.