Measure, optimise, prove it with data
Convolve audio with before/after impulse responses
In professional audio, subjective impressions are unreliable. Memory of how a system sounded before adjustments fades within seconds, making it impossible to objectively evaluate whether EQ changes, acoustic treatment, or driver alignment actually improved performance. Before/After comparison solves this by capturing a quantitative baseline measurement, then overlaying the post-adjustment response to reveal exactly what changed and by how much.
The frequency response plot shows magnitude in decibels across the audible range from 20 Hz to 20 kHz on a logarithmic frequency axis. Four viewing modes let you examine the data from different perspectives. Before Only (red trace) shows the original system response. After Only (green trace) shows the post-adjustment response. Both Overlaid renders both traces simultaneously for direct visual comparison. Delta view computes the difference (After minus Before) and plots it against a zero reference line, making it immediately obvious which frequency ranges were boosted or cut and by how much.
Raw frequency response curves contain thousands of data points. The metrics dashboard distills this into five actionable numbers. Frequency Flatness measures the deviation from a perfectly flat response across the speech intelligibility band (200 Hz to 8 kHz), expressed in dB. Lower is better. Phase Coherence indicates how consistent the phase relationship is across frequency, which directly affects imaging and intelligibility in multi-speaker systems. System Health is a composite score from 0 to 100 that weighs flatness, coherence, noise floor, and distortion. SPL Change shows whether the adjustments affected overall level. Noise Floor tracks whether the signal chain cleanliness improved or degraded during the process.
Numbers tell part of the story, but sometimes you need to hear the difference. The A/B listening feature uses the Web Audio API ConvolverNode to apply the measured impulse response to a live audio signal. Switch between Before and After impulse responses to hear how your programme material (music, speech) would sound through each system state. This bridges the gap between measurement data and perceptual experience, which is especially valuable when presenting results to non-technical stakeholders who may not interpret frequency response graphs.
The Share Comparison feature generates a permanent link to your before/after results, complete with the frequency response overlay and all metrics. This is invaluable for system integrators who need to document their work for clients, consultants who need to demonstrate the impact of acoustic treatment, and sound engineers building a portfolio of system optimisations. Each shared comparison includes improvement percentages that quantify the value of the work performed.
For averaged frequency response measurements, 5 to 10 seconds of pink noise or swept sine excitation provides sufficient averaging to produce stable, repeatable results. The averaging reduces the influence of room noise and transient events. For impulse response capture (used in A/B listening), a single sweep or click is sufficient.
No. The microphone must remain in exactly the same position for before and after measurements. Any change in position introduces differences due to the room transfer function at the new location, which would be incorrectly attributed to your system adjustments. Use a microphone stand and mark the position if needed.
Delta view computes the difference in dB between the after and before measurements at each frequency. Positive values mean the after measurement is louder at that frequency; negative values mean it is quieter. A flat delta line at 0 dB means no change. This view is particularly useful for verifying that EQ adjustments landed at the intended frequencies and depths.
System Health is a weighted composite of frequency flatness (40%), phase coherence (25%), noise floor (20%), and distortion (15%). A score of 100 represents a theoretically perfect system. Scores above 80 indicate excellent performance. Between 60 and 80 is acceptable for most applications. Below 60 suggests significant issues that should be addressed.
The tool is designed for immediate before/after comparison in a single session. However, you can export baseline measurements and import them later for comparison with a new post measurement. Be aware that environmental factors (temperature, humidity, audience presence) affect room acoustics and may introduce differences unrelated to your system changes.