A white noise generator test online lets you produce a continuous, uncorrelated random audio signal directly in your browser tab, so you can verify speakers, headphones, audio routing, or a DAW input in seconds without installing software or uploading a file. The Lizely White Noise Generator does exactly that: a fresh two-second mono AudioBuffer is filled with uniformly distributed samples from -1 inclusive to 1 exclusive, then looped through a single GainNode for instant playback. The tool starts at a conservative 15% gain and accepts any slider value from 0% through 100% while the signal plays, with explicit Start and Stop buttons and zero microphone access. Because generation and playback both happen locally, your audio never leaves the current tab. That makes the test useful for quick confidence checks on any device with a modern browser — long enough to confirm a signal, short enough to abort the moment something sounds wrong.
The Goal Behind a White Noise Generator Test Online
When readers search for a white noise generator test online, they usually want one of three things: to confirm that browser-based audio playback works on their device, to verify a hardware path such as speakers, headphones, or an audio interface, or to evaluate a generator before relying on it for music work. The Lizely White Noise Generator supports all three because it produces the test signal itself rather than streaming a pre-recorded file. Each session creates one AudioContext, one looping AudioBufferSourceNode, and one GainNode, so the test signal is consistent and easy to clean up.
Generation begins with a fresh 32-bit seed requested at every Start, then runs the three-shift xorshift32 recurrence first described by George Marsaglia to produce unsigned 32-bit values. Those values are mapped linearly into [-1, 1) and placed into a two-second mono buffer that loops until Stop is selected or the component leaves the page. The browser's native sample rate is accepted only from 8 kHz through 192 kHz, which caps the allocation at about 384,000 float samples, or roughly 1.5 MiB of sample storage at the top rate. That fixed budget prevents an unusual AudioContext configuration from requesting an unbounded buffer.
How to Run the Online White Noise Test
- Set your device output low, then leave the page slider at its conservative 15% starting value or reduce it further before you begin.
- Select Start white noise. The buffer begins looping immediately. Adjust the 0–100% slider only as needed while you listen.
- Select Stop before navigating away, or whenever the sound feels uncomfortable. Stop closes the AudioContext and disconnects both the source and the gain node.
Each Start is guarded against duplicate or stale starts, so repeated clicks cannot create overlapping players. If Web Audio is unavailable or the browser rejects startup, the tool shows an error and leaves no stale playing state.
What the Test Signal Actually Is
White noise is defined by a power spectral density that is flat across the frequencies of interest. NI's reference on dithering and high-quality components describes wide-band noise as flat in the frequency domain, and MathWorks documents the same flat-PSD definition in its Band-Limited White Noise reference. The Lizely signal is a practical digital approximation rather than mathematically ideal infinite-bandwidth noise. It is band-limited by whatever sample rate the browser reports and uses a uniform amplitude distribution rather than Gaussian.
For a uniform random variable U on [-1, 1), the expected mean-square value can be derived in three steps:
- Formula: E[X²] = ∫ from −1 to 1 of x² · (1/2) dx
- Substituted: (1/2) · [x³/3] evaluated from −1 to 1 = (1/2) · (1/3 − (−1/3))
- Result: 1/3, so the RMS amplitude is sqrt(1/3) ≈ 0.577
Both uniform and Gaussian distributions qualify as white when samples are uncorrelated and the expected power spectral density is flat, so the uniform bounded approximation remains a valid test signal in practice.
Audio Chain Checks the Online Test Helps You Run
The signal is broadband and roughly time-invariant, which means it makes several common problems easy to hear. Dropouts, glitches, pops on Start or Stop, and channel imbalance all stand out against broadband hiss. If you route browser playback into a DAW, the test doubles as a routing check: a clean capture at the expected level means the input is wired correctly and not clipping, and the Web Audio API specification defines the nodes the tool uses to keep that path predictable.
| Test goal | How the tool helps | What to listen for |
|---|---|---|
| Confirm browser audio output | Start the signal and listen at the speakers | Any hiss at all, without crackle or silence |
| Verify left or right routing | Pan the destination speakers or DAW input | Signal present in only the expected channel |
| Check DAW input monitoring | Route the browser output into the DAW | Clean, non-clipping capture at a steady meter |
| Detect interface dropout | Run the loop while you wiggle cables | No clicks, glitches, or sudden silence |
What the Online Test Will Not Tell You
The signal is not calibrated. The 0–100% slider controls only the page's linear GainNode, so your operating system, browser, speakers, headphones, amplifier, and listening environment all change the actual sound pressure you receive. The conservative 15% starting value is a tool default rather than a measurement or safety guarantee, and 100% is not endorsed as safe.
The tool also does not produce Gaussian noise, does not measure room acoustics, does not record microphone input, and does not claim to diagnose, prevent, or treat hearing, tinnitus, sleep, anxiety, or any other health condition. It is a simple controllable signal source, not medical equipment, hearing protection, or a certified acoustic test instrument.
Test Volume Safely
The volume slider expresses linear gain as volumePercent / 100. At the 15% default the linear gain is 0.15, at 50% it is 0.5, and at 100% it is 1.0. None of those numbers is a safety threshold on its own. Begin with your device output low, leave the page slider at 15% or below, and raise either control slowly only after you have confirmed the signal is comfortable. If anything feels uncomfortable at any setting, press Stop right away and follow the guidance supplied with your audio equipment.
For a deeper walk-through of how the buffer, the seed, and the gain slider fit together, see the companion guide on how to generate white noise in your browser.