Display real-time frequency spectrum analysis of audio with FFT visualization.
This tool displays live audio as a frequency plot: the horizontal axis runs from low to high frequency, and the vertical axis shows how much energy each frequency carries, redrawn continuously as the sound changes. It works by taking short slices of the incoming audio and computing a fast Fourier transform on each slice, revealing the frequencies inside it. The analysis runs entirely in your browser, so the audio never leaves your device. The thing most people misread is the tallest peak: voices and instruments produce harmonics, so the highest bump is often an overtone rather than the pitch. Read the whole pattern.
A concrete input and expected output from the current implementation.
Input
Play a steady 440 Hz test tone, for example from any online tone generator, and hold it while the analyzer runs.
Expected output
The display shows one dominant, narrow peak positioned near 440 Hz, well above a quiet floor elsewhere. Raise the pitch and the peak moves right; lower it and it moves left; stop playing and it collapses back to the floor.
A pure tone puts nearly all its energy at one frequency, so the transform produces a single tall band centered there. The exact position is the nearest frequency bin, which can sit slightly above or below 440 Hz depending on the transform size and sample rate.
The plot spans from near 0 Hz up to the Nyquist limit, which is half the sample rate of whatever audio feeds the analysis. The source sets that boundary: high-rate audio shows the full human hearing range, while low-rate recordings or streams stop at a few kilohertz and show no energy above that. Within that range, the transform measures energy at evenly spaced frequencies; the spacing is the sample rate divided by the transform size.
The shape of the display identifies the sound. A pure tone is one narrow bump. Speech and musical notes form a stack of evenly spaced bumps: the fundamental plus harmonics at double, triple, and higher multiples of its frequency. Broad noise, such as breath, hiss, or wind, fills a wide region at low height. When a harmonic stack appears, the spacing between the bumps, not the tallest one, is the actual pitch.
No. The analysis runs entirely in your browser on this device; the audio is processed locally as it plays and is not uploaded or stored. If the tool requests microphone access, the permission belongs to this page only and can be revoked at any time in the browser's site settings.
Most real sounds contain many frequencies at once: a voice or instrument produces a fundamental plus harmonics at whole-number multiples, and a harmonic is often louder than the fundamental. The tallest bump can therefore be an overtone while the actual pitch is lower. Read the pitch from the spacing between the evenly spaced bumps; that spacing is the fundamental frequency.
The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.
Convert between bytes, KB, MB, GB, and TB with precise binary and decimal calculations.
Open, view, and edit CSV files in a spreadsheet-style table with sorting and filtering.
Visualize JSON data as a collapsible tree with syntax highlighting and search.
Generate Code 128, EAN, UPC, and other barcode formats with customizable size and labels.
Convert images to Base64 data URLs and decode Base64 strings back to images.
Generate Apache .htaccess redirect rules for 301, 302, and rewrite configurations.