White Noise Generator - White, Pink & Brown Noise Online

Colored noise is random sound with a defined spectral shape, and the color names describe where the energy sits. White noise spreads equal power across every hertz from bass to treble, which the ear hears as a bright hiss because each higher octave contains twice as many hertz as the one below. Pink noise rolls the level off at 3 decibels per octave, giving every octave equal energy — this matches how human hearing groups frequencies, so pink noise sounds naturally balanced, closer to steady rain. Brown noise (also called red or Brownian noise, after Brownian motion rather than the color) falls at 6 decibels per octave, concentrating energy in the low end for a deep rumble like surf heard through a wall. This generator synthesizes each color in your browser with the Web Audio API. White noise is plain random samples; pink noise is shaped with Paul Kellet's well-known filter approximation, accurate to within a fraction of a decibel across the audible band; brown noise integrates white noise sample by sample with a leak term so it never drifts or clips. The two-second buffer loops seamlessly, so playback is continuous, gapless, and works offline once the page has loaded — nothing is streamed or recorded. Why people use it: steady broadband noise raises the acoustic floor around you, so sudden sounds — a door, a colleague's call, traffic — stand out less against the background. That masking effect is the mechanism behind most reported benefits for focus in open offices and for falling asleep in noisy apartments. Research on noise and sleep is mixed overall, but masking of intermittent disturbances is well established, and many people simply find the lower colors less fatiguing: if white noise feels harsh after a few minutes, try pink; if you want the treble gone entirely, brown is the popular choice for deep-focus and sleep playlists. The sleep timer stops playback after 15, 30, or 60 minutes, which covers the typical 10 to 20 minutes most adults need to fall asleep without leaving noise running all night. Keep the level moderate: effective masking usually needs only 50 to 60 dBA, about the loudness of a quiet conversation, and hearing-safety guidance recommends staying at or below that for hours-long listening — especially for children, where 50 dBA at the ear is the commonly cited ceiling for infant sound machines. For professional use, the export section renders 1, 5, or 10 minutes of the selected color offline at 44.1 kHz stereo — each channel filled with independent random samples, so the file has a wide, decorrelated stereo image instead of a mono signal copied to both sides — and downloads it as a 16-bit WAV or a 192 kbps MP3. Pink noise files are the workhorse here: because pink noise carries equal energy per octave (the minus 3 dB per octave slope matches how analyzers band their filters), it is the standard stimulus for calibrating speaker levels with an SPL meter, ringing out monitors with a real-time analyzer, and setting surround channels to a common reference. White noise files suit burn-in and hiss-balance checks, and brown noise makes a portable low-frequency masker. The size readout shows the exact WAV footprint before you commit — a minute of stereo 16-bit audio at 44,100 Hz is a shade over 10 MB, so the MP3 is the right pick for phone playlists.