White Noise Generator

What it does

The White Noise Generator produces uniform random noise samples at a configurable amplitude — up to 10,000 values per run — with selectable noise colour variants. White noise has equal power at all frequencies and no correlation between samples: the flattest, most structureless signal there is, which is precisely what makes it useful.

Common use cases

• Signal-processing tests — filters and FFT code verified against input with a known flat spectrum.
• Dithering — uniform noise is the textbook dither signal for audio and image quantization.
• Baseline comparisons — anomaly detectors and smoothers should do nothing interesting on white noise; that’s a valuable test.
• Simulation inputs — uncorrelated perturbations where Gaussian tails aren’t wanted.

Settings

• Samples — 1 to 10,000 values, exportable as CSV, JSON, or text.
• Amplitude — the noise scale; values are uniform within it.
• Noise colour — white or tinted variants that shape the spectrum.
• Seed — identical seed + settings = identical noise, so DSP tests stay deterministic.

Privacy note

Samples are computed locally in your browser; nothing is uploaded — synthetic mathematics, not recordings of anything.

FAQ

White noise vs Gaussian noise? Orthogonal concepts: “white” describes the spectrum (flat, uncorrelated samples), Gaussian describes the amplitude distribution. This tool draws uniform amplitudes; the Gaussian Noise tool draws bell-curve amplitudes — both are spectrally white.

What are noise colours? Spectral tilts: pink noise rolls off high frequencies (natural-sounding), brown/red rolls off harder (that’s a random walk!), blue emphasizes highs. Different test signals for different filters.

How do I verify my FFT with this? Transform a large white-noise sample — the spectrum should be flat within statistical wobble. Structure in the output means the bug is yours.