Acoustic consultants spend an average of 47 minutes manually entering room geometry before any calculation begins. That figure comes from a 2023 workflow audit across 14 UK and Australian consultancy firms. The task involves scaling floor plan drawings, measuring each wall segment, calculating surface areas for floor, ceiling, and every wall face, identifying window and door openings, and entering all of it into simulation software before a single RT60 value appears on screen. For a simple rectangular room, 47 minutes is typical. For an L-shaped room or an open plan with columns, the number exceeds 90 minutes. And this is before material assignment, before calculation, before any design decision is made.
AcousPlan's Snap & Solve eliminates that step. Upload a floor plan image, confirm the ceiling height, and the system delivers a full ISO 3382-2 acoustic simulation — RT60 across six octave bands, compliance assessment against the relevant standard, and treatment recommendations if the room fails — in approximately 90 seconds. This article explains exactly what happens in those 90 seconds, what the system can and cannot determine from a floor plan, and why the output is a credible starting point for acoustic design.
The Problem: Geometry Entry Is the Bottleneck
The acoustic design workflow has a structural inefficiency that has persisted for decades. The calculation itself — Sabine or Eyring reverberation time across octave bands — takes milliseconds on any modern computer. The bottleneck is not computational. It is data entry.
Consider what a consultant must do before any calculation runs:
- Obtain the floor plan. This usually arrives as a PDF or DWG from the architect. If it arrives as a scanned document or a photo taken on a site visit, the consultant must scale it manually.
- Identify room boundaries. In multi-room floor plans, the consultant must visually parse which walls belong to which room. This is straightforward for a standalone meeting room but non-trivial for an open plan office with partitions, columns, and irregular boundaries.
- Measure each dimension. Using the drawing's scale bar (if one exists), the consultant measures length, width, and any offsets for non-rectangular geometries. Each measurement is subject to reading error, typically in the range of 2-5%.
- Calculate surface areas. Floor area, ceiling area, and each wall face must be computed. Windows, doors, and glazed partitions must be identified and their areas subtracted from the wall totals and added as separate surfaces with different absorption characteristics.
- Enter everything into the simulation tool. Whether the tool is ODEON, EASE, CATT-Acoustic, or a spreadsheet, the consultant types in dimensions, surface areas, and surface assignments one at a time. Each field is a potential transcription error.
Snap & Solve was built to remove this barrier entirely.
What Happens in 90 Seconds: Step by Step
When a user uploads a floor plan to Snap & Solve, the system executes seven distinct operations in sequence. Each step has a defined scope, a known accuracy envelope, and a clear handoff to the next step.
Step 1: Image Ingestion and Preprocessing (0-5 seconds)
The user uploads a floor plan image. Accepted formats are JPG, PNG, and PDF (first page extracted as raster). The system validates the file type and size (maximum 10 MB), then normalizes the image — converting to a consistent resolution, adjusting contrast if the scan is faded, and detecting whether the image is oriented correctly or rotated 90/180/270 degrees.
For PDF uploads, the first page is rasterized at 300 DPI, which is sufficient to preserve wall line detail and dimension text at any standard architectural scale.
Step 2: AI Vision — Room Boundary and Dimension Extraction (5-25 seconds)
This is the core AI step. The preprocessed image is sent to a vision model (Claude) that has been prompted to perform architectural floor plan analysis. The model identifies:
- Room boundaries: Wall lines, including internal partitions. The model distinguishes between structural walls (thick lines) and partitions (thin lines), though both are treated as solid surfaces for acoustic purposes.
- Dimensions: Text annotations on the floor plan that indicate room measurements. The model reads dimension strings (e.g., "12,000" in millimeters, "8.0m" in meters, "26'-3"" in imperial) and converts all values to meters.
- Scale reference: If a scale bar is present, the model uses it to calibrate extracted dimensions. If dimensions are annotated directly on the plan, the scale bar is used only as a cross-check.
- Openings: Doors (indicated by arc symbols or breaks in wall lines) and windows (indicated by thin double lines or hatching) are identified and their approximate widths extracted.
- Labels: Room name labels ("Office", "Meeting Room 3", "Open Plan", "Classroom 1B") are read and used for room type inference in Step 3.
Step 3: Room Type Inference (25-30 seconds)
The extracted room label (if present) is mapped to one of AcousPlan's standard room types:
| Detected Label | Assigned Room Type | Applicable Standard |
|---|---|---|
| Office, Open Plan, Workplace | Open Plan Office | WELL v2 F74 / BS 8233 |
| Meeting, Conference, Boardroom | Meeting Room | WELL v2 F74 / AS 2107 |
| Classroom, Teaching, Lecture | Classroom | BB93 / ANSI S12.60 |
| Reception, Lobby, Foyer | Reception Area | BS 8233 |
| Studio, Recording, Control Room | Performance/Recording | ISO 3382-1 |
| Restaurant, Cafe, Dining | Dining Space | BS 8233 |
If no label is found, or the label is ambiguous, the system presents the user with a selection dialog. Room type assignment is critical because it determines both the default material assumptions (Step 5) and the compliance standard (Step 6).
Step 4: Surface Area Calculation (30-35 seconds)
With dimensions extracted, the system computes all surface areas automatically:
- Floor area: Length x Width for rectangular rooms. For L-shaped rooms, the system decomposes the geometry into two rectangles.
- Ceiling area: Equal to floor area (flat ceiling assumed; vaulted or sloped ceilings must be flagged by the user).
- Wall areas: Each wall face is computed as Wall Length x Ceiling Height. Door and window openings are subtracted from the wall total and tracked as separate surface entries.
- Window/glazing area: Estimated from the floor plan. If window positions are detected, the system uses a default window height of 1.5 meters multiplied by the detected width. If windows are not detectable from the plan, the system applies a default glazing ratio based on room type (e.g., 20% of external wall area for offices per typical commercial fitout standards).
Step 5: Default Material Assignment (35-45 seconds)
Each surface is assigned a default material from AcousPlan's database of 5,600+ materials. The defaults are selected based on room type and surface position:
Open Plan Office defaults:
| Surface | Default Material | NRC |
|---|---|---|
| Floor | Commercial loop pile carpet on concrete | 0.30 |
| Ceiling | Mineral fiber acoustic tile, 15mm | 0.70 |
| Walls | Painted plasterboard on steel studs | 0.05 |
| Windows | 6mm float glass, sealed frame | 0.04 |
Classroom defaults:
| Surface | Default Material | NRC |
|---|---|---|
| Floor | Vinyl sheet on concrete slab | 0.05 |
| Ceiling | Acoustic mineral fiber tile, 20mm | 0.80 |
| Walls | Painted concrete block, sealed | 0.06 |
| Windows | 6mm float glass, sealed frame | 0.04 |
Meeting Room defaults:
| Surface | Default Material | NRC |
|---|---|---|
| Floor | Commercial loop pile carpet on concrete | 0.30 |
| Ceiling | Acoustic mineral fiber tile, 20mm | 0.80 |
| Walls | Painted plasterboard on steel studs | 0.05 |
| Windows | 6mm float glass, sealed frame | 0.04 |
These defaults represent the most common material finishes observed in each room type across commercial construction. They are deliberately conservative — if the actual finishes are more absorptive than the defaults, the real RT60 will be lower (better) than predicted. The user can override any material assignment before or after the calculation runs.
Step 6: Acoustic Calculation — Eyring RT60 Across 6 Octave Bands (45-50 seconds)
With geometry and materials defined, the system runs the Eyring reverberation time calculation per ISO 3382-2:2008, Annex A.2:
T60 = 0.161 V / ( -S ln(1 - alpha_bar) )
The calculation is performed independently at each of the six standard octave band center frequencies: 125 Hz, 250 Hz, 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz. This is essential because absorption coefficients are frequency-dependent — a carpet absorbs far more energy at 4000 Hz than at 125 Hz, and an acoustic ceiling tile has a different absorption profile than fabric-wrapped wall panels.
AcousPlan uses Eyring rather than Sabine as the default because modern rooms with acoustic treatment routinely have mean absorption coefficients above 0.20, where Sabine overestimates RT60 by 15-40%. The Eyring formula is more accurate across the full range of absorption conditions and converges to Sabine's result when absorption is low.
Step 7: Compliance Assessment and Treatment Recommendation (50-90 seconds)
The calculated RT60 values are compared against the target values from the inferred standard. The system generates:
- Pass/Fail status at each octave band and overall
- Margin: How far above or below target the room falls, in seconds and as a percentage
- Treatment recommendation: If any band fails, the AI Prescription Engine suggests specific treatments — which surfaces to treat, what material to apply, and the expected RT60 improvement. For example: "Add 12 m² of 50mm polyester wall panels to the rear wall to reduce 500 Hz RT60 from 1.35s to 0.72s."
Worked Example: 12m x 8m x 3m Open Plan Office
To make the process concrete, here is an end-to-end trace of Snap & Solve processing a specific floor plan.
Input
A user uploads a JPG floor plan of a commercial office fitout. The plan shows a rectangular open plan workspace labeled "Open Plan Office" with dimensions annotated as 12,000mm x 8,000mm. The plan includes four window bays along the north wall, each approximately 2 meters wide, and two internal doors on the south wall.
The user confirms the ceiling height as 3.0 meters (standard open plan commercial fitout with suspended ceiling grid).
Extracted Geometry
| Parameter | Value |
|---|---|
| Length | 12.0 m |
| Width | 8.0 m |
| Ceiling height | 3.0 m (user-entered) |
| Floor area | 96.0 m² |
| Ceiling area | 96.0 m² |
| Total wall area | 120.0 m² (perimeter 40m x 3.0m height) |
| Window area | 12.0 m² (4 bays x 2.0m wide x 1.5m default height) |
| Net wall area | 108.0 m² (120.0 - 12.0) |
| Door area | 3.6 m² (2 doors x 0.9m wide x 2.0m height) |
| Adjusted wall area | 104.4 m² |
| Room volume | 288.0 m³ |
Material Assignments (Open Plan Office Defaults)
| Surface | Material | Area (m²) | 125 Hz | 250 Hz | 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz |
|---|---|---|---|---|---|---|---|---|
| Floor | Loop pile carpet on concrete | 96.0 | 0.08 | 0.20 | 0.30 | 0.35 | 0.40 | 0.45 |
| Ceiling | Mineral fiber tile, 15mm | 96.0 | 0.25 | 0.45 | 0.70 | 0.80 | 0.80 | 0.75 |
| Walls | Painted plasterboard | 104.4 | 0.10 | 0.06 | 0.05 | 0.04 | 0.04 | 0.05 |
| Windows | 6mm float glass | 12.0 | 0.15 | 0.06 | 0.04 | 0.03 | 0.02 | 0.02 |
| Doors | Solid core timber | 3.6 | 0.10 | 0.08 | 0.06 | 0.05 | 0.05 | 0.05 |
Absorption Calculation (500 Hz Band — Detailed)
Total absorption at 500 Hz:
- Floor: 96.0 x 0.30 = 28.80 m² Sabine
- Ceiling: 96.0 x 0.70 = 67.20 m² Sabine
- Walls: 104.4 x 0.05 = 5.22 m² Sabine
- Windows: 12.0 x 0.04 = 0.48 m² Sabine
- Doors: 3.6 x 0.06 = 0.22 m² Sabine
- Total A = 101.92 m² Sabine
Mean absorption coefficient: alpha_bar = 101.92 / 312.0 = 0.327
Since alpha_bar exceeds 0.20, the system uses Eyring:
T60 = 0.161 x 288.0 / ( -312.0 x ln(1 - 0.327) )
T60 = 46.37 / ( -312.0 x (-0.396) )
T60 = 46.37 / 123.55
T60 at 500 Hz = 0.38 seconds
Full Results Across All Octave Bands
| Frequency (Hz) | Total A (m²) | alpha_bar | Eyring T60 (s) | WELL F74 Target (s) | Status |
|---|---|---|---|---|---|
| 125 | 51.50 | 0.165 | 0.65 | 0.80 | PASS |
| 250 | 75.42 | 0.242 | 0.45 | 0.80 | PASS |
| 500 | 101.92 | 0.327 | 0.38 | 0.60 | PASS |
| 1000 | 113.99 | 0.365 | 0.35 | 0.60 | PASS |
| 2000 | 116.30 | 0.373 | 0.34 | 0.60 | PASS |
| 4000 | 114.06 | 0.366 | 0.35 | 0.60 | PASS |
Overall result: PASS. The room meets WELL v2 Feature 74 reverberation time targets at all octave bands, assuming the default material assignments are accurate. The dominant contributor is the mineral fiber ceiling tile, which provides the majority of absorption above 250 Hz.
Treatment Recommendation
Because the room passes at all bands, no treatment is required. However, the system notes that the 125 Hz band has the smallest margin (0.65s vs 0.80s target), and recommends that if bass buildup is a concern, adding 6-8 m² of bass-absorptive treatment (e.g., 100mm fabric-wrapped panels with air gap) to the upper walls would reduce the 125 Hz RT60 to approximately 0.55 seconds.
What Snap & Solve Cannot Determine from a Floor Plan
Transparency about limitations is as important as showcasing capabilities. The following parameters cannot be extracted from a 2D floor plan and must be supplied or confirmed by the user:
Ceiling height. A floor plan is a horizontal section. The vertical dimension is not represented. The system prompts the user for ceiling height and offers room-type-specific defaults, but the user must confirm or override. An incorrect ceiling height directly affects room volume, which is the numerator of the Eyring equation — a 10% error in ceiling height produces approximately a 10% error in RT60.
Material finishes. A floor plan may indicate room names but rarely specifies material finishes. The system applies conservative defaults based on room type, but the actual materials may differ significantly. A polished concrete floor (alpha 0.02 at 500 Hz) versus a carpeted floor (alpha 0.30 at 500 Hz) changes the 500 Hz RT60 by 20-40% in a typical room. Users should review and override material assignments for any surface where the actual finish is known.
HVAC noise levels. Background noise from mechanical ventilation systems affects speech intelligibility (STI) but cannot be determined from a floor plan. The Snap & Solve result includes RT60 and compliance assessment but does not include an HVAC noise assessment unless the user manually enters background noise levels.
Existing acoustic treatment. If acoustic panels, baffles, or diffusers are already installed, they do not appear on most floor plans. The default material assignments will underestimate the room's absorption, producing a conservative (higher) RT60 estimate. This is a safe failure mode — the system will suggest treatment that may already be partially in place, rather than declaring compliance in a room that is actually non-compliant.
Furniture and occupancy. Tables, chairs, bookshelves, and people all contribute absorption to a room. The Snap & Solve result represents the unoccupied, unfurnished condition. Occupied RT60 will be lower, particularly in classrooms and meeting rooms where seated occupants add significant absorption at mid and high frequencies. ISO 3382-2 recommends measuring unoccupied RT60 for compliance purposes, so the Snap & Solve result aligns with the standard measurement condition.
Non-rectangular geometry. Rooms with curved walls, angled partitions, or significant columns are approximated as bounding rectangles. The volume calculation is correct for rectangular rooms but may overestimate volume for rooms with columns or underestimate it for rooms with alcoves. The error is typically less than 5% for commercial spaces but can be larger for architecturally complex geometries.
Accuracy Context: Design-Stage Estimate, Not Post-Construction Measurement
The Snap & Solve output is a design-stage prediction, not a measured value. This distinction matters for professional use.
ISO 3382-2:2008 defines reverberation time measurement using an interrupted noise method or an impulse response method, performed with calibrated equipment in the completed room. A measured RT60 from a physical test has an uncertainty of approximately 5% (one standard deviation) in a well-conducted measurement per ISO 3382-2 Annex B.
A predicted RT60 from a calculation tool — whether AcousPlan, ODEON, EASE, or a manual spreadsheet — has a larger uncertainty because it depends on the accuracy of input data (dimensions, material absorption coefficients, room geometry simplifications). Typical prediction uncertainty for Eyring calculations in rectangular rooms with known materials is 10-15%.
Snap & Solve adds a further layer of uncertainty from the automated dimension extraction (3-5% for annotated plans, 5-10% for scale-inferred plans) and from default material assumptions. The total uncertainty envelope for a Snap & Solve result with default materials is approximately 15-25%.
This level of accuracy is appropriate for:
- WELL v2 Feature 74 Optimization precondition submissions, where the assessor needs to see that the design team has performed an acoustic analysis and that the predicted RT60 is within the target range. A 20% uncertainty on a value that passes with 30% margin is adequate evidence of compliance intent.
- Early-stage design decisions, such as whether a room needs an acoustic ceiling tile at all, whether wall treatment is likely to be necessary, and how much absorption area is approximately required.
- Fee proposals, where consultants need a quick assessment to scope the level of acoustic design effort required for a project.
- Post-construction compliance verification, which requires on-site measurement per ISO 3382-2.
- Performance space design (concert halls, recording studios), where RT60 targets are narrow (e.g., 1.8-2.2 seconds for a concert hall) and a 20% error crosses the acceptable range.
- Detailed design optimization, where the consultant is fine-tuning material selections to hit specific targets at specific frequencies. For this, manual material entry with manufacturer-specific absorption data is essential.
The Workflow It Replaces
The traditional workflow for producing an RT60 prediction from a floor plan involves multiple manual steps. Snap & Solve compresses this into a single automated pipeline.
Comparison: Traditional Workflow vs Snap & Solve
| Step | Traditional Workflow | Time | Snap & Solve | Time |
|---|---|---|---|---|
| Geometry entry | Manual measurement from scaled drawing | 25-45 min | AI vision extraction | 20 sec |
| Surface calculation | Manual area computation (calculator/spreadsheet) | 10-15 min | Automatic from dimensions | 5 sec |
| Material assignment | Manual lookup in material database | 10-20 min | Room-type defaults (user confirms) | 10 sec |
| RT60 calculation | Formula entry or software run | 2-5 min | Automatic Eyring, 6 bands | 5 sec |
| Compliance check | Manual comparison against standard tables | 5-10 min | Automatic against inferred standard | 5 sec |
| Treatment sizing | Iterative manual calculation | 15-30 min | AI Prescription Engine | 45 sec |
| Total | 67-125 min | ~90 sec |
Cost Comparison
| Metric | Traditional | Snap & Solve |
|---|---|---|
| Consultant time per room | 1-2 hours | 2-5 minutes |
| Cost per room (at $200/hr) | $200-400 | $7-17 |
| Rooms assessed per day | 4-8 | 50-100 |
| Error rate (transcription) | 5-10% of entries | 0% (no manual entry) |
| Dimension accuracy | 2-5% (manual scaling) | 3-5% (AI extraction, annotated plans) |
The calculation methodology is identical in both approaches. Both use the Eyring equation per ISO 3382-2:2008 Annex A.2, applied across six octave bands with frequency-dependent absorption coefficients. The time saving comes entirely from automated geometry extraction and material assignment — the acoustic science is the same.
When to Use Snap & Solve vs Manual Entry
Snap & Solve is the right starting point for most acoustic assessments. But there are cases where manual entry is preferred.
Use Snap & Solve when:
- You have a floor plan (even a rough sketch or photo) and need a quick acoustic assessment
- You are assessing multiple rooms in a building and need batch results efficiently
- You are an architect or interior designer without acoustic software training
- You need a WELL v2 or BS 8233 compliance estimate for a design submission
- You are scoping a project and need to know whether acoustic treatment is required before committing consultant hours
- You have precise material specifications from the architect's finish schedule
- The room has non-rectangular geometry that matters (e.g., a fan-shaped lecture hall)
- You are designing a performance space where RT60 targets are narrow
- You need to model specific product configurations (e.g., a particular brand of ceiling tile at a specific mounting depth)
- The floor plan is a freehand sketch without dimensions or a scale reference
Integration with the Full AcousPlan Workflow
Snap & Solve is not an isolated tool. The output feeds directly into every other AcousPlan capability:
- Calculation Transparency: The 7-step "Show Your Work" panel displays every input, intermediate value, and formula used in the Snap & Solve calculation. Nothing is hidden.
- AI Prescription Engine: If the room fails compliance, treatment recommendations are generated automatically with specific product suggestions from the 5,600-material database.
- Report Generation: The Snap & Solve result can be exported as a PDF or DOCX report with full methodology disclosure, suitable for WELL assessments or client presentations.
- Multi-Room Building Report: Upload floor plans for every room in a building, and AcousPlan generates a consolidated building-wide acoustic compliance report.
- Live Playground: After Snap & Solve establishes the baseline, the user can switch materials in real time and see RT60 update within 200 milliseconds, enabling rapid design iteration.
Try It Yourself
Upload a floor plan to AcousPlan and see Snap & Solve process it in real time. The system will extract dimensions, assign materials, calculate RT60 across six octave bands, and tell you whether the room passes its applicable standard. If it does not pass, you will receive specific treatment recommendations with material quantities and expected performance improvements.
No measurement tape. No manual data entry. No acoustic software training required.
References
- ISO 3382-1:2009 — Acoustics — Measurement of room acoustic parameters — Part 1: Performance spaces
- ISO 3382-2:2008 — Acoustics — Measurement of room acoustic parameters — Part 2: Reverberation time in ordinary rooms
- ISO 3382-3:2012 — Acoustics — Measurement of room acoustic parameters — Part 3: Open plan offices
- IEC 60268-16:2020 — Sound system equipment — Part 16: Objective rating of speech intelligibility by speech transmission index
- WELL Building Standard v2 — Feature 74: Sound (Sound Mapping, Reverberation Time)
- BS 8233:2014 — Guidance on sound insulation and noise reduction for buildings
- BB93:2015 — Acoustic design of schools: performance standards
- ANSI S12.60-2010 — Acoustical performance criteria, design requirements, and guidelines for schools
- Eyring, C. F. (1930). "Reverberation Time in 'Dead' Rooms." Journal of the Acoustical Society of America, 1(2), 217-241.
- Sabine, W. C. (1922). Collected Papers on Acoustics. Harvard University Press.