Combine decibels, work the Sabine relationship, build a room's absorption, estimate speech intelligibility, and browse the material data: the routine Week 6 acoustics arithmetic in one place, so your time goes to measurement and interpretation.
Each tab opens with example values so you can see it working; replace them with your own readings.
Decibels measure sound energy on a logarithmic scale, so you cannot average or add them with plain arithmetic.
LAeq is the equivalent continuous level (the single steady level carrying the same energy as a fluctuating signal), and the "A" weighting discounts very low and very high frequencies to track how the ear hears, so a dBA figure is close to perceived loudness. Because the scale is logarithmic, two equal levels combine to just 3 dB above one of them, while a source 10 dB quieter adds almost nothing. Enter your levels separated by commas or spaces, then energy-average repeated readings of one source or combine independent sources into a total.
The calculator converts each level back to sound energy (10^(Li/10)), averages or sums the energies, then returns to decibels. Because the scale is logarithmic the answer is always pulled toward the louder reading rather than the arithmetic midpoint (doubling the energy is only +3 dB), which is why averaging dB values directly would mislead.
LAeq differs from the statistical levels a meter may also show: LA90 (exceeded 90% of the time) tracks the steady background floor, while LA10 (exceeded only 10% of the time) captures the louder intermittent peaks; LAeq sits between them.
Reverberation time (RT60) is how long a sound takes to decay by 60 dB after the source stops: the "tail" you hear in a hard, empty room.
It is the single most telling descriptor of a room's acoustic character, and the Sabine equation ties it to the room's size and its total sound absorption: A = 0.16·V/T. Absorption and reverberation are inversely related: roughly double the absorbing area and the reverberation time halves. Enter the volume plus one of T or A and the tool returns the other.
With V the room volume in m³ and T the average RT60 at 500 Hz and 1000 Hz. The 0.16 constant (≈ 24·ln(10)/c) carries the speed of sound c, which is why it is specific to metric units. Sabine assumes a diffuse, well-mixed sound field, so it is only approximate in very live or very dead rooms, or where the absorption is bunched on one surface rather than spread evenly around.
Measurements become design evidence once judged against a benchmark. AS/NZS 2107:2016 gives recommended design sound levels and reverberation times per interior type: find the closest category, then check that your energy-averaged LAeq and your mid-frequency RT60 (the 500 Hz and 1 kHz average) sit within range. These are indicative figures for orientation; the standard gives satisfactory-to-maximum ranges that vary with room size, so consult AS/NZS 2107:2016 for design work.
| Space type | LAeq (dBA) | Mid-freq RT60 (s) |
|---|---|---|
| Lecture theatre | 35–40 | 0.6–1.0 |
| Teaching space / classroom | 35–45 | 0.4–0.6 |
| Private office | 35–40 | 0.4–0.6 |
| Open-plan office | 40–45 | 0.4–0.6 |
| Meeting / conference room | 30–40 | 0.6–0.8 |
| Library (quiet study) | 40–45 | 0.4–0.6 |
| Corridor / lobby / foyer | 45–50 | n/a |
Build a room's total absorption surface by surface, to see where the acoustics come from.
A room reverberates the way it does because of its size, its surfaces, and its occupancy: people are strong absorbers. Add each major surface (ceiling, walls, floor, glazing, furniture, seating), pick its material, and enter its area in m²; the tool sums the octave-band absorption A = Σ(S·α) and predicts the reverberation time per band. Compare that against the RT60 you measured: where they diverge, the difference is usually people, furniture, or geometry the generic table cannot capture.
A = Σ(S·α) adds up each surface's area S (m²) times its absorption coefficient α: the fraction of incident sound energy the surface soaks up rather than reflects (0 = perfectly reflective, 1 = a fully open window). α is strongly frequency-dependent: soft, porous materials like carpet and acoustic tile absorb the highs well but little bass, whereas low frequencies are usually tamed by panel or membrane absorbers. That is why the tool works band by band, and why a room can end up bright or boomy rather than uniformly live or dead.
Absorption coefficients are per m² of surface; seating/audience values are per m² of floor area they occupy. Predicted RT uses T = 0.16·V/A in each band and will differ from measurement because the table is generic and does not model geometry.
The Speech Transmission Index scores how well the detail of speech survives its passage across a room, on a scale from 0 (unintelligible) to 1 (excellent).
It rolls the two things that wreck intelligibility (background noise and reverberation) into a single number. Two routes are offered here, matching the tutorial: a one-line quick read from the speech-to-noise ratio for a first feel in the room, and the full octave-band calculation (IEC 60268-16) ported from the class STI workbook. Use the qualitative bands to interpret the score.
This method reproduces the BPSD5030 STI workbook developed by Densil Cabrera (School of Architecture, Design and Planning, University of Sydney), verified cell-for-cell against the original spreadsheet.
Enter room volume, talker-to-listener distance, and the octave-band reverberation time and background noise. If you could not measure RT at 125 Hz or 8 kHz, substitute the 250 Hz and 4 kHz values respectively and note that you did so.
The quick estimate maps the A-weighted speech-to-noise ratio straight onto the STI scale: fast, but it sees neither reverberation nor the per-band structure of noise. The full method reproduces the class STI workbook: a statistical room-acoustics model (IEC 60268-16, with auditory masking) that computes the modulation transfer index in each octave band from the room volume, talker distance, reverberation time and background-noise spectrum, then weights the bands into one index.
Qualitative bands used here (as in the spreadsheet): <0.20 none · 0.20–0.40 poor · 0.40–0.60 fair · 0.60–0.80 good · >0.80 excellent.
The octave-band data behind the calculators: absorption coefficients for finishes and sound reduction indices for constructions.
Read a row across the octave bands (63 Hz–8 kHz) to see how a material behaves with frequency: a carpet soaks up highs but little bass, a masonry wall does the reverse. Switch between the two tables with the pills, and search or filter by category to find a material. On any absorption row, click + to push that material straight into the room-absorption builder in the Room absorption tab.
Tip: click + on an absorption row to add it to the room-absorption builder in the Room absorption tab.
Absorption coefficient α is the fraction of incident sound energy a surface soaks up (0 = perfectly reflective, 1 = a fully open window); it feeds A = Σ(S·α) in the Room absorption tab. Sound reduction index R (dB) is how much airborne sound a construction blocks per octave band, with Rw a single-number rating. All values are indicative generic data (BB93 / Adrian James), for rough calculations only, not for Building-Control submissions.