To check the result after generating white noise, start the browser-local signal at the tool’s 15% default gain, listen for steady broadband sound, adjust the 0–100% slider if needed, and select Stop to confirm immediate shutdown. The White Noise Generator does not create a finished file, waveform, spectrum display, or acoustic test report, so “checking” means auditioning the live output and confirming that its playback controls respond correctly. Internally, it creates a continuous, bounded random signal with a nominally flat power distribution across frequencies the browser can represent. This is a practical digital approximation, not mathematically ideal infinite-bandwidth noise. For a careful check, lower the device output first, leave the tool at 15% or reduce it further, then start playback. Listen briefly, make small gain adjustments only if needed, and stop the signal before leaving the page or whenever it feels uncomfortable. The 15% setting controls only this page’s gain; it does not represent a measured sound-pressure level or a safety guarantee. Your operating system, browser, speakers or headphones, amplifier, and listening environment also shape the final level and tonal impression.
What the Result Represents
The result is a live playback session rather than a saved audio file. The player fills a two-second mono buffer with independent-looking random samples uniformly distributed from -1 inclusive to 1 exclusive, then loops that buffer through a gain stage. Playback continues until Stop is selected or the page is left.
Checking the result therefore means separating an operational playback check from a technical review of the documented signal and from an acoustic measurement of your room. The page exposes the first, its implementation defines the second, and it does not measure the third.
- Playback behavior: Confirm that the sound starts, continues, responds to gain changes, and stops when requested.
- Signal behavior: Review the documented mono, bounded, nominally flat character of the generated signal.
- Acoustic behavior: Understand that the tool cannot measure sound pressure, room response, or equipment calibration.
For a closer look at the difference between nominal spectral flatness and real-world playback, see How Accurate Is a Browser White Noise Generator?
Run a Controlled Playback Check
- Set your device output low. Open the White Noise Generator and leave its gain at the conservative 15% starting level, or reduce it further. The percentage belongs to the page’s audio control; it is not a decibel reading or a measurement of sound pressure.
- Select Start white noise. Listen for a continuous, broad signal rather than a short clip. Adjust the 0–100% slider only as needed. At 0%, the page’s gain stage is silent; moving the slider upward increases its linear gain. A setting of 100% is accepted, but the tool does not endorse it as safe.
- Select Stop before leaving the page or whenever the sound feels uncomfortable. Confirm that playback ceases immediately. If the signal becomes uncomfortable, stop it first rather than spending time searching for a lower setting.
If no sound begins, check that the device output is audible and the browser tab is not muted. A startup rejection produces an error rather than leaving a stale playing state, so an error message is also part of the result check rather than something to ignore.
Confirm Continuity and Control Response
The most useful functional test is to observe each control in sequence. Because the two-second buffer loops, the generated signal should continue instead of ending after two seconds. Listen for the broad, even character expected of the generated noise, while remembering that your speakers or headphones can alter its tonal impression.
| Control action | Expected result | What it confirms |
|---|---|---|
| Select Start white noise | One continuous mono playback session begins | The audio path started successfully |
| Move the 0–100% slider | 0% is silent, while smaller and larger percentages change the page gain | The gain control responds |
| Select Stop | Playback ends immediately and the current audio graph is closed | The shutdown path works |
| Click Start repeatedly | No overlapping players or layered copies are created | The duplicate-start guard works |
| Leave the page | Cleanup stops, disconnects, and closes the active audio resources | Unmount cleanup completes |
Each successful start requests a fresh 32-bit seed, so later sessions can present a different random sequence even when the gain setting has not changed. That variation is expected. Repeated clicks while a start is pending or playback is active cannot create several simultaneous players.
Keep Signal Accuracy Separate from Loudness
The term “white” refers to a nominally flat expected power spectral density across the frequencies the browser can represent. It does not mean infinite bandwidth or mathematically perfect noise. This generator uses bounded uniform samples, whose amplitude distribution differs from Gaussian white noise, and its practical output is described as a band-limited digital approximation.
The implementation keeps generation bounded as well. The player accepts the current browser’s native sample rate only from 8,000 through 192,000 Hz. Its exactly two-second mono Float32 buffer reaches 384,000 float samples, or about 1.5 MiB of sample storage, at the maximum accepted rate. These are implementation boundaries, not controls exposed in the interface.
Developer-side verification is more detailed than a listening test. Eight exact Float32 samples catch recurrence or mapping drift, while a 200,000-sample fixture checks bounds, near-zero mean, expected uniform mean-square power, RMS, extremes, and low lag-one product. Each session uses a fresh 32-bit seed with the xorshift32 sequence. These tests support the implementation, but they do not appear as a spectrum panel that listeners can inspect.
The 15% default controls only this page’s linear gain. Your operating system, browser, speakers, headphones, amplifier, fit, and environment all affect actual loudness. A low percentage is not a measurement or safety guarantee, and 100% is not endorsed as safe. Begin low, lower the output or stop if the sound feels uncomfortable, and follow the guidance supplied with your audio equipment. For the distinction between tool gain and measured loudness, see Does the 15% Default Guarantee Safe White Noise Levels?
Verify Cleanup and Local Processing
Each start creates one AudioContext, one looping AudioBufferSourceNode, and one GainNode, using the browser processing model described by the W3C Web Audio API. Stop invalidates any start waiting for browser resume, stops and disconnects the audio nodes, and closes the context. The same cleanup runs if startup fails or the component unmounts, so navigation and repeated Start/Stop cycles do not intentionally leave live audio graphs behind.
If Web Audio is unavailable or the browser rejects startup, the tool shows an error and leaves no stale playing state. Treat that error as a failed playback check rather than assuming silence means the signal is working. Also check the browser’s current output path and device volume, because the generator cannot correct a muted tab or a disconnected headphone.
All generation and playback happen in the current browser tab. No audio, setting, seed, or listening data is uploaded to Lizely, and the tool does not access the microphone. It does not record the result, download a file, analyze a room, or calibrate sound pressure. It is a controllable signal source, not medical equipment, hearing protection, or a certified acoustic instrument.
Final Verification Checklist
- The device output was low before playback began.
- The tool remained at 15% or lower until a small adjustment was needed.
- The sound continued rather than stopping after the two-second buffer duration.
- The slider produced an audible gain change, and 0% silenced the page.
- Stop ended playback immediately, with no lingering sound or error state.