Controlled signal
The tool generates the test inside your browser with a bounded app-level gain. Your operating-system and hardware volume remain separate.
The analyzer turns microphone samples into a real-time FFT display in the browser. It is useful for seeing relative frequency energy, not for calibrated acoustic measurement.
No microphone audio is transmitted or stored. The display is relative and depends on the microphone, browser processing and room.
The tool generates the test inside your browser with a bounded app-level gain. Your operating-system and hardware volume remain separate.
Output tests do not upload audio. Microphone tools request permission only when started and explicitly state what stays in memory.
The analyzer converts a local microphone stream into FFT frequency bins and draws their relative energy. It can reveal a changing speech spectrum, a persistent narrow peak or a silent input path without transmitting the samples.
Keep the level moderate and change one thing at a time so the result remains useful.
Start the analyzer and grant microphone permission.
Play or speak near the microphone at a safe level.
Observe broad frequency regions and peaks rather than treating one frame as a measurement.
Stop when finished to release the microphone immediately.
Match what you heard or saw, change one variable, then use the linked check to verify the result.
| Observation | Likely cause | Next step | Retest |
|---|---|---|---|
| 01The display stays flat after permission is granted. | The wrong input, hardware mute, system privacy control or disconnected microphone may be preventing samples. | Confirm the system input meter moves, then restart the browser analyzer. | Test the microphone path |
| 02A narrow peak stays near the same frequency. | A steady tone, electrical hum or persistent device noise can concentrate energy in a small set of FFT bins. | Turn likely sources off one at a time and confirm whether the peak disappears or moves. | Generate a reference tone |
| 03Energy is spread broadly and changes with speech or room noise. | Speech, fans and environmental noise contain many simultaneous frequencies rather than one stable sinusoid. | Move closer to the intended source or reduce background noise, then compare at the same microphone gain. | Read relative microphone level |
| 04A visible peak does not match the expected frequency exactly. | FFT bin spacing, sample rate, window length, microphone response and nearby noise limit the displayed precision. | Use a longer stable reference and interpret the nearest band rather than claiming laboratory accuracy. | Review browser sample rate |
The generated signal stays controlled; the physical device, room and system path can still change what reaches you.
Built-in and consumer microphones emphasize or attenuate different bands before the browser receives samples.
Operating-system or device gain control, noise suppression and echo processing may reshape the spectrum.
Distance, reflections and background sources alter the energy arriving at the microphone.
A finite analysis window groups frequencies into bins; it does not produce an infinitely precise line.
The active AudioContext sample rate determines bin spacing and the highest representable frequency.
No. Samples go from the microphone to the Web Audio analyzer inside the page and are not transmitted.
It controls the frequency-analysis window. Larger values give finer frequency bins but update over a longer time window.
FFT bin spacing, microphone response, background noise and sample rate all affect the displayed peak.