ATAcoustic Toolbox

Create Sounds

Sound Generator

Create calibrated pressure waveforms and coloured noise, listen safely in the browser and export the samples as mono 32-bit-float WAV.

Signal definition

Preview audio is peak-normalized to a conservative listening level. It is not an acoustically calibrated reproduction of the entered dB SPL or Pa value.

Equivalent level—
RMS pressure—
Peak pressure—
Crest factor—
Samples / file—

Generated waveform

Noise colours

Idealized one-sided power spectral density, normalized to 0 dB at 1 kHz.

WhitePinkBrownBlueViolet

Mathematical definition

S(f) ∝ fα

The exponent α describes the slope of the power spectral density S(f). A doubling of frequency changes the spectral level by approximately 3.01 α dB per octave.

White
α = 0 · 0 dB/oct
Pink
α = −1 · −3.01 dB/oct
Brown / red
α = −2 · −6.02 dB/oct
Blue
α = +1 · +3.01 dB/oct
Violet
α = +2 · +6.02 dB/oct

What the WAV contains

The downloaded mono IEEE-float WAV stores the generated instantaneous sound pressure directly in pascals. This is useful for numerical workflows, simulation and calibrated processing. Ordinary media players interpret floating-point samples as normalized audio and can clip pressure values outside ±1, so use the normalized browser preview for listening.

Signal characteristics

Square, triangle and sawtooth waveforms are synthesized from harmonics below Nyquist to reduce aliasing. Noise generation uses a repeatable seed: white noise has constant power per hertz, pink noise approximately per octave, brown noise emphasizes low frequencies, while blue and violet noise increasingly emphasize high frequencies. A short cosine fade prevents clicks while preserving the requested overall RMS pressure.