What is RT60 and why does it matter?

RT60 is the time a sound takes to decay by 60 decibels — in practice, how long the room rings. Below about 0.5 s a room sounds dry and speech is clear; above roughly 1.0 s in a small room, speech intelligibility falls away and the space feels harsh. It is the single most useful number for judging whether a room needs absorption.

Why does it measure T20 rather than a true RT60?

A genuine RT60 needs 60 dB of clean dynamic range above the noise floor, which phone hardware effectively never has. Acousticians solve this by measuring a shorter, cleaner part of the decay — from −5 to −25 dB — and extrapolating by three. That is what this does, and it is the standard method, not a shortcut.

Why is the level shown as dBFS instead of dB SPL?

Because a phone microphone is uncalibrated and its automatic gain control moves the reference continuously. dBFS is relative to digital full scale and is honest; a dB SPL figure from the same signal would be an invented number. Use it to compare before and after treating a room, not to certify a noise level.

What makes a good impulse for the decay measurement?

A single sharp hand clap, or a burst balloon in a larger room. It must be loud enough to sit well above the background and short enough not to smear the decay. Stand away from walls and corners, where reflections arrive too quickly to separate.

How does the STC estimate work, and how far can I trust it?

It uses the mass law at a 500 Hz reference — TL = 20·log10(m·f) − 47, with m the surface mass in kg/m². That captures the dominant effect for a single solid leaf, and it is a reasonable first pass. It does not model cavities, decoupling, resilient channels or coincidence dip, all of which matter enormously in real partitions. Use it to compare options, and use a tested assembly rating for anything that has to satisfy a code.

Is my audio recorded or uploaded?

No. The analyser reads frames from the browser's audio node, computes on them, and discards them. There is no recording, no file and no server.

Why does the analyser not start?

Microphone access requires HTTPS and an explicit permission grant. If you dismissed the prompt previously the browser remembers that for the site, and you will need to reset the permission in the site settings.

Field tool

Acoustic Analyzer

Measure room reverberation with your microphone, or estimate reverberation and wall sound isolation from geometry alone.

MicrophoneWhat it works from
MicrophoneMicrophone permission · HTTPS

Timber-framed house under construction with ladders, sawhorses and lumber

This browser does not expose microphone access, so the live measurement section is unavailable. The Sabine and mass-law calculators below work regardless.

Sabine reverberation estimate

RT60 = 0.049 × V ÷ A — from room volume and total absorption.

Room volume (ft³)Total absorption (sabins: sq ft × α)

0.79 s

Sabine assumes a diffuse field and evenly distributed absorption; it over-predicts in very dead or very irregular rooms (where Eyring’s formula is preferred).

Mass-law STC estimate

TL = 20·log₁₀(m × 500 Hz) − 47, from the wall’s surface mass — the formula's m is in kilograms per square metre, so a figure typed in pounds per square foot is converted before it goes in.

Surface mass (lb/sq ft)

≈ STC 32

Low confidence by design. Mass law models a single limp panel. It ignores stud decoupling, cavity depth, insulation and resilient channels — which in a real partition often matter more than mass. Use it to compare options, never to certify a party wall. Rated assemblies must come from a tested design (e.g. a UL or NRC assembly listing), not from this number.

How to use the Acoustic Analyzer

Allow microphone access
Start the analyser and accept the browser prompt. Audio is processed frame by frame and never recorded.
Watch the live spectrum
The spectrum and relative level show what the room is doing now — useful for spotting a dominant hum or a plant noise.
Capture a decay
Start a capture and make a sharp impulse — a single hard hand clap or a balloon burst. Stand away from walls and corners.
Read RT60
The decay slope is measured between −5 and −25 dB and extrapolated, which is the standard T20 method.
Compare against Sabine
Enter room volume and absorption to compute the theoretical Sabine RT60, and surface mass to estimate transmission loss. Neither uses the microphone.

Where this tool stops

A phone microphone is uncalibrated and applies automatic gain control, so levels are relative dBFS and never absolute dB SPL. Presenting them as SPL would be a fabrication.