Spectrogram Compare

Drop or click to load two audio files — File A on the left and File B on the right. Both spectrograms will appear side by side with identical settings, so you can visually compare the frequency content. You can load one file first and add the second later. Use the shared settings panel to adjust FFT size, color palette, window function, frequency scale, and dB range — all changes apply to both spectrograms simultaneously for a fair comparison.

Look for frequency cutoff differences — a lossy file (MP3/AAC) will have a sharp cutoff in the high frequencies while a lossless file (FLAC/WAV) will show content all the way up. Check for differences in detail — lossy compression removes quiet sounds, so you may notice less texture in the spectrogram. Also compare the overall brightness and distribution — level differences between files are easy to spot when viewed side by side with identical color settings.

Upsampled audio is a lower-quality file (like a 128 kbps MP3) that has been re-encoded at a higher sample rate or into a lossless format to appear better than it is. To detect it, compare the suspected file against a known genuine hi-res file. The upsampled file will show a hard frequency cutoff (e.g., nothing above 16 kHz) despite being saved as 96 kHz FLAC. A real hi-res file will have natural frequency content extending to the Nyquist frequency. The Difference mode makes this even more obvious — the region above the cutoff will show large differences (red).

Difference mode replaces the two side-by-side spectrograms with a single heatmap that shows the absolute dB difference between File A and File B at every point in time and frequency. It computes |dB_A - dB_B| for each pixel. Where the two files are identical, the difference is zero (shown as black). Where they differ, the color indicates how large the difference is. This is the quickest way to see exactly where and how two files are different.

Black = identical (0 dB difference). Green = similar (less than 3 dB difference) — this is within the range of minor level variations and is generally inaudible. Yellow = moderate (3-10 dB difference) — this indicates noticeable differences, often caused by lossy compression artifacts or EQ changes. Red = large (more than 10 dB difference) — this indicates significant differences, such as missing frequency content, added noise, or completely different audio material.

No. All processing — decoding, FFT analysis, comparison, and rendering — happens entirely in your browser using the Web Audio API and JavaScript. Your audio files never leave your device and no data is sent to any server. You can even use this tool offline once the page has loaded.