Room Acoustics Analysis Tool

Overview

The tool operates on an impulse response (IR) — a recording of how your room responds to a short, broadband sound event. An IR encodes the complete acoustic behaviour of the room at the measurement position: frequency response, modal decay times, reflections, and reverberation.

You can provide an IR in two ways: by taking a measurement directly in the browser using your speakers and microphone, or by uploading a WAV file exported from external measurement software such as REW.

Once an IR is submitted, the server performs deconvolution, frequency analysis, peak detection, and EQ filter design. Results are returned as interactive plots and tables.

Room dimensions (optional)

At the top of the tool, three optional fields accept room dimensions in metres: length, width, and height. These are not required for the analysis to run — if left blank, the tool identifies peaks purely from the measured frequency response.

If dimensions are provided, the tool computes theoretical axial modal frequencies for each axis using:

f(n) = n × c / (2L)

c = speed of sound (~343 m/s at 20°C)
n = mode order (1, 2, 3...)
L = room dimension (m)

Each detected peak in the measured response is then matched to the nearest theoretical mode, annotated with the axis (length, width, or height) and order. This cross-referencing helps distinguish genuine room modes from other spectral features such as loudspeaker resonances or HVAC noise.

Taking a measurement

The measurement panel plays an exponential swept-sine signal through your speakers, records the acoustic response via your microphone, and uses Farina deconvolution to extract a calibrated impulse response. No external software is required.

Before you start

• — Place your microphone at the primary listening position, at ear height.
• — Use an omnidirectional microphone if possible. A measurement microphone (e.g. miniDSP UMIK-1) will give more reliable results than a laptop microphone.
• — Close doors and windows. Switch off noisy HVAC equipment during the sweep.
• — Set playback volume high enough to achieve adequate signal-to-noise ratio in the low frequencies, but not so high as to cause distortion. The tool will warn you if SNR is low.

Measurement parameters

Procedure

1. 1.Enter room dimensions if required.
2. 2.Adjust measurement parameters if needed. Default values are appropriate for most rooms.
3. 3.Click "Measure Room". The browser will request microphone permission — grant it.
4. 4.The tool plays the sweep. Do not move or make noise during playback.
5. 5.After the sweep ends, the tool captures the room decay for a further 2 seconds.
6. 6.The recording is sent to the server. Deconvolution and analysis run automatically.
7. 7.Results appear below the measurement panel.

Uploading an impulse response

If you have an existing IR measurement — from REW, Smaart, or another measurement application — you can upload it directly as a WAV file.

File requirements

• — Format: WAV (PCM integer or IEEE float, 16–32 bit)
• — Sample rate: 44.1 kHz or 48 kHz
• — Channels: mono preferred; stereo files are accepted and the left channel is used
• — The file should contain the impulse response only, not the raw swept-sine recording

Exporting from REW

1. 1.Take a measurement in REW at your listening position.
2. 2.In the measurement list, select the measurement you want to export.
3. 3.Go to File → Export → Export Impulse Response as WAV.
4. 4.Save the file, then drag it into the upload zone or click to browse.

Interpreting the results

Once an IR is submitted, the tool returns several outputs.

SNR indicator

Signal-to-noise ratio is estimated from the IR. A warning is shown if SNR is below approximately 20 dB, which indicates the measurement may be unreliable — particularly at low frequencies where background noise is often highest. Increase playback level or repeat the measurement in a quieter environment.

Frequency response plot

The magnitude of the transfer function from source to measurement position, derived from the IR via FFT. Displayed in dB over the analysis frequency range. Detected peaks are marked. This shows what the room is doing to the frequency response at your listening position.

Trend analysis plot

A smoothed version of the frequency response showing the low-frequency trend. Used as the baseline against which peaks are identified — a peak is flagged when it exceeds the local trend by a threshold amount.

Decay plot

Energy decay per octave band, derived from the IR. Shows how quickly sound energy decays at each frequency. Bands with significantly longer decay than the room average are likely to contain modal resonances.

Waterfall (CSD) plot

Cumulative spectral decay displayed as a colour map. The horizontal axis is frequency, the vertical axis is time after excitation, and colour represents energy level. Horizontal ridges — energy that persists long after the source stops — indicate modal resonances. The colour map saturates at the peak level and fades as energy decays; a ridge that extends far down the time axis at a specific frequency identifies a problem mode.

Detected peaks table

A table of frequency peaks identified in the measured response, sorted by severity. Each entry shows the peak frequency and estimated level above the local trend.

Room modes table

Shown only if room dimensions were provided. Lists predicted axial modal frequencies for each room dimension, with each matched to the nearest measured peak where one exists within a tolerance band.

EQ filters table

Suggested parametric EQ cuts for each detected peak. Each filter is specified by centre frequency, gain (negative, in dB), and Q factor. These are starting points — apply them in your DAW or room correction software and re-measure to verify the result.

Saving files

After a browser measurement, two download buttons appear:

• Download IR — the extracted impulse response WAV. Save this to track changes over time or to import into REW for further analysis.
• Download Recording — the raw microphone capture including the sweep and room decay. Useful for debugging a failed measurement.

Saving the IR WAV allows you to re-upload it later to compare results before and after treatment or EQ changes without needing to re-measure.