White Noise
Definition
White Noise
White noise is a random signal with equal power spectral density across all frequencies — every hertz of bandwidth contains the same energy. Named by analogy with white light (equal energy at all visible wavelengths), white noise sounds bright and hissy because higher octaves contain proportionally more energy. SonaVyx generates white noise and uses it as a basis for pink noise generation and acoustic testing.
Power spectral density: S(f) = constant for all f; energy doubles per octave (+3 dB/octave)
How It Is Measured
White noise is generated digitally using a pseudorandom number generator producing uniformly distributed samples. Each sample is statistically independent, resulting in a flat power spectrum from 0 Hz to the Nyquist frequency. SonaVyx generates white noise in the AudioWorklet and can filter it to produce pink noise (-3 dB/octave slope) for system measurement applications.
Practical Example
An acoustician uses white noise to test the frequency response of a 1/3 octave filter bank for IEC 61260 compliance verification. Because white noise has equal energy per hertz, each successive third-octave band should show exactly 1 dB more energy than the previous band (since each band is wider by a factor of 2^(1/3)). Any deviation from this 1 dB/band staircase indicates a filter bank error.
White Noise vs Pink Noise
White noise has equal energy per hertz, which means each successive octave contains twice the energy of the previous one (+3 dB/octave). Pink noise is filtered white noise with a -3 dB/octave slope, producing equal energy per octave. For audio measurement with octave-band displays (RTA), pink noise appears flat while white noise appears as a rising slope. Pink noise is therefore preferred for system tuning with SonaVyx.
Applications
White noise is used as a reference for testing filter characteristics, calibrating measurement equipment, generating pink noise (by applying a -3 dB/octave filter), sound masking in offices (sometimes with spectral shaping), and acoustic testing where linear-frequency analysis is used. For auditory masking research, white noise provides a known, predictable masking spectrum.
Perceptual Character
White noise sounds bright and slightly hissy to human ears because the equal-energy-per-hertz distribution places disproportionate perceived energy at high frequencies. Pink noise sounds more balanced because its equal-energy-per-octave distribution matches the logarithmic frequency perception of human hearing. For sleep and relaxation applications, brown noise (-6 dB/octave) sounds even warmer and is sometimes preferred.
Try It Now
Generate white noise — free audio signal generation in your browser