ROOM TONE · MEASURE · COMPARE

Audio Noise Profile Library

Keep a record of how your recording space sounds. Compare room tone, equipment settings and quiet passages before your next session.

MEASURED SNAPSHOT

Choose a profile

Frequency fingerprint

20 Hz–20 kHz · log scale

Windowed FFT snapshots from the selected region. These are digital levels, not calibrated sound-pressure measurements.

Listen & choose a quiet region

Start · secondsEnd · seconds

Choose room tone without speech or music. Measurements include all sound in the region; the tool cannot identify noise automatically.

Build a useful room-tone library

Capture. Record a few seconds of the room without speaking, using the same mic distance and input gain.
Compare. Try a fan, light or computer setting change. Add a second recording and compare the fingerprints.
Keep. Label the conditions and save the measured library. Original audio is not included.

Up to 12 profiles; audio files up to 25 MiB each, and eight million decoded channel samples per recording. Supported formats depend on your browser. Saved libraries hold measurements and notes, not original recordings. Browser decoding can resample compressed audio. Public pages may include advertising; see our privacy information.

Using Audio Noise Profile Library

Measure and label quiet audio regions, compare frequency fingerprints and keep a reusable library of recording conditions.

Try the studio example to compare fan-on, fan-off and hum demonstration signals. For your own measurements, open supported audio recordings.
Choose a quiet start/end region with no speech or music, then Measure region. Save a clear profile name, location, microphone and gain notes.
Select a comparison reference to overlay spectra and compare RMS levels. Keep gain, equipment and microphone distance consistent.
Save measured library to reopen profiles without original audio. Export a summary CSV or the selected profile report.

Example

The example uses three generated signals at a fixed digital gain: broadband fan-like noise, a quieter fan-off condition, and a 60 Hz hum. These illustrate measurements, not real microphone recordings.

Questions & answers

Does this remove background noise?

No. It measures selected regions and stores labeled profiles. Noise removal is a separate editing task.

What do the levels mean?

RMS and peak are uncalibrated digital dBFS values, not sound-pressure levels or LUFS. All sound in the selected region contributes to them.

How is the frequency plot measured?

Up to 16 evenly spaced Hann-windowed FFT snapshots are averaged in power across the selected region and all channels. Frequency resolution depends on sample rate and window size. It is not an exhaustive transient or hum detector.

Do saved profiles include audio?

No. Library JSON keeps notes, level statistics, a waveform envelope and the measured spectrum. Reopened profiles can be compared but cannot be auditioned or remeasured without opening audio as a new profile.

What are the limits?

Up to 12 profiles, 25 MiB per audio file and eight million decoded channel samples per recording; all loaded recordings share a sixteen-million-sample memory budget. Audio format support and resampling depend on your browser.