Building the dataset together

Every conference room deserves a score.

The Waze for Acoustics

We are building the world's first crowdsourced conference room acoustic dataset. The more facilities teams contribute, the smarter everyone's benchmarks become. Measurement works offline today, and community context unlocks as the dataset grows.

Phase 1 contributors help unlock benchmarks faster and shape the baseline everyone compares against.

Phase 1
Dataset foundation
0 -> 1,000
Scans to first milestone
Early access
Help set the standards
  • Now: RT60, noise, and C50 clarity scoring works offline
  • Expert intelligence (AI-assisted): Equipment fit, budget scenarios, and recommendations
  • Portfolio view: Group rooms by site, building, and floor
  • Benchmarks: Unlock at 100+ scans (equipment) and 500+ scans (percentiles)
RoomScore app screenshot placeholder
App preview (placeholder)
Early Access

Phase 1: Building the foundation.

RoomScore works offline today. Community benchmarks unlock as scans arrive.

0
Scans so far
0
This week
0 of 1,000 scans Phase 1 target

Seeding phase. Benchmarks unlock at 100 scans, percentiles at 500.

Why Now

Every conference room should have a score.

You already track WiFi, occupancy, and air quality. Acoustic performance should be just as measurable.

10 seconds per room

Fast enough to scan a full portfolio in one afternoon.

No extra equipment

LiDAR and on-device acoustics. No calibration gear required.

Free core workflow

Measurements and offline scoring stay accessible to every team.

For lean IT teams: scan 12 rooms in one afternoon and walk into budget reviews with defensible data.

Measurement Credibility

Phone-based measurement, with professional discipline.

RoomScore uses industry-standard acoustics, then enforces strict quality gates. The app would rather reject a scan than show a weak score.

Method (physics-first)

T20 decay analysis

RT60 is computed from the -5 dB to -25 dB decay window, then extrapolated to full decay.

Three-clap median

Three captures per room, median value reported to reduce variance.

150 Hz high-pass filter

Removes low-frequency HVAC rumble to focus on speech-band energy.

C50 speech clarity

Clarity index is computed alongside RT60 for intelligibility context.

Quality gates (auto reject)

R2 fit threshold

Measurements with R2 below 0.80 are rejected.

Noise stability checks

Ambient noise spread must stay within 12 dB, with a quiet window within 8 dB.

Device stability

Gyroscope + accelerometer detect movement and prompt a retry.

Clipping detection

Overloaded microphone signals are rejected before scoring.

Screening-grade, not certification.

RoomScore is validated in internal testing against reference tools and is designed for portfolio screening, prioritization, and budgeting. For certification or compliance work, use calibrated Class 1 meters.

Physics-based Measurements are immutable Expert intelligence adds context
Community Powered

Your scans compound into better context.

RoomScore improves as more teams contribute. Your data comes back as smarter benchmarks for your rooms.

What You Share (Default)

Volume, RT60, noise level, and equipment type. No room names or GPS.

Anonymous by default

What You Get Back

Percentile context, equipment benchmarks, and treatment validation as data grows.

Compounding value

Why It Matters

Community context turns measurements into budget-ready insights.

Early phase

How the network effect works

1

You scan a room. We anonymize and add it to the dataset.

2

Other teams scan their rooms. Benchmarks grow smarter.

3

Your original rooms gain percentile context without extra work.

Workflow

Scan. Measure. Score. Benchmark. Simulate. Act.

Deterministic physics first, contextual guidance second.

Scan

RoomPlan captures volume, surface area, and glass-to-wall ratio.

Measure

10-second noise test, 3-clap RT60 (T20), and C50 speech clarity with quality gating.

Score

Composite 0-100 score with star rating and confidence indicators.

Benchmark

Compare against similar rooms once the dataset reaches 100 scans.

Simulate

Add virtual panels to predict RT60 and score changes instantly.

Act

Save treatment plans and export PDF reports for your portfolio.

Expert Intelligence

Expert intelligence, grounded in physics.

Measurements are physics-based and immutable. Expert Intelligence (AI-assisted) adds context and recommendations without altering the measurement data.

What the expert layer delivers

  • Equipment fit score: Does the current setup match the room?
  • Budget scenarios: Minimal, recommended, and comprehensive plans.
  • Photo context (if shared): Visual checks for placement and treatment cues.
  • Plain-language guidance: Rationale tied to measurements.

Multi-room portfolio

Group audits by Site -> Building -> Floor, track completion progress, and compare rooms side-by-side.

Completion tracking: Know what is scanned and what is next.
Comparison view: Benchmark rooms within the same portfolio.
Export-ready: Share PDF reports with stakeholders.
Portfolio

See the full picture, not just one room.

Facilities teams manage dozens of rooms. RoomScore keeps your portfolio organized and measurable.

Dataset Roadmap

What unlocks as the dataset grows.

RoomScore works offline from day one. Community features unlock as scan milestones are reached.

Day 1
Available now

Core measurement + scoring

RT60, noise floor, C50 clarity, expert intelligence (AI-assisted), treatment simulation, PDF export. Works offline.

100
Scans

Equipment benchmarks

Compare mic and video systems against real-world performance.

500
Scans

Percentile rankings

High-confidence context for prioritizing treatment budgets.

Research
1,000+ scans

Outcome and spatial research

Treatment outcome analytics and room shape insights as the dataset scales.

Scale Vision

What becomes possible at scale.

These are future possibilities unlocked by large, anonymized datasets. They are not promises today.

10,000 scans

  • Regional and industry benchmarks
  • Seasonal performance trends
  • Equipment lifecycle insights

50,000 scans

  • Predictive maintenance alerts
  • Anomaly detection across portfolios
  • Placement optimization patterns

100,000+ scans

  • Automated treatment design
  • Equipment ROI benchmarks
  • Workplace platform API access
Phase 1

Why contribute early?

We're grateful to teams helping build the foundation. The value is reciprocal: the faster the dataset grows, the faster your rooms gain context.

Your benchmarks get better faster

At 100 community scans, equipment benchmarks unlock. At 500, percentile rankings appear. The sooner you contribute, the sooner your rooms have context.

Shape the standards

The first 1,000 scans define what "good" looks like. Your rooms help set the baseline everyone compares against.

Priority feedback

We run quarterly surveys with active contributors. You help decide which features to prioritize.

The reward is the dataset. Scan five rooms and benefit from five hundred as the community grows. That is the network effect.

Join Early Access
Benchmarks

Example benchmark output.

Illustrative output from internal validation. Live benchmarks unlock after 100+ community scans.

1

Scan a room and capture RT60, noise, and volume.

2

Match against similar rooms by volume and equipment.

3

Unlock percentile context once the dataset reaches critical mass.

Illustrative Example

Sample benchmark layout

Your RT60: 0.88 s (sample)
Target range: 0.40 to 0.60 s (sample)
Percentile: Bottom 20% (sample)
Room cohort: 40 to 60 m3 (sample)

Example values only. Live benchmarks appear after community data is collected.

Example data
Privacy by Design

Fair trade: your data for better context.

You choose what to share. Default is anonymous metrics only, enough to power benchmarks without identifying rooms.

Metrics Only

Volume, RT60, noise level, and equipment type. Default for every scan.

Research

Adds impulse response audio to validate measurement accuracy.

Full Research

Shares anonymized 3D geometry after rotation and centering.

Support

Direct support.

Replies come from the founder.

Contact

Email me at support@roomscore.tech.

Typical response time: 1 business day.

What devices are supported?

iPhone 15 Pro or newer with LiDAR. iOS 17+ required.

How accurate are the measurements?

RoomScore is validated in internal testing against reference tools (typical RT60 agreement within +/- 15% in conference rooms). The app rejects low-quality measurements automatically.

How is RT60 calculated?

RoomScore captures three claps, applies a 150 Hz high-pass filter, fits a T20 decay from -5 dB to -25 dB, and requires an R2 fit of at least 0.80 before reporting the median RT60.

Why trust phone-based measurement?

RoomScore enforces strict quality gates: device stability, clipping detection, and noise stability (spread at or below 12 dB, quiet window at or below 8 dB). If a scan fails, the app asks you to retry.

How does RoomScore compare to professional meters?

RoomScore is for fast screening and portfolio prioritization. Class 1 meters are required for certification, compliance, or critical acoustic design work.

What is the difference between data sharing tiers?

Metrics (default) shares RT60, noise, volume, and equipment. Research adds impulse audio. Full Research adds anonymized 3D scans. Private mode keeps everything on device.

Can I use this for acoustic certification?

No. RoomScore is for screening and planning. For certification, use calibrated Class 1 equipment and a licensed acoustical engineer.

Will core features stay free?

The plan is to keep core measurement and offline scoring free. Optional paid tiers may add exports, team workflows, or advanced analytics later.

Conference Room Acoustic Dataset, Built Together

RoomScore brings professional acoustic measurement to iPhone while building a shared conference room acoustic dataset. Whether you are a facilities manager optimizing meeting spaces, an AV integrator validating room readiness, or an acoustic consultant doing pre-site screening, RoomScore provides fast, reliable measurements you can trust.

How It Works: RT60 Measurement on iPhone

RoomScore captures room impulse responses via a simple 3-clap test. A 150 Hz high-pass filter removes low-frequency rumble, then a T20 decay fit (from -5 dB to -25 dB) estimates RT60. The app rejects measurements with poor linear fit (R2 below 0.80), unstable noise, or phone movement.

Crowdsourced RT60 Benchmarks for Facilities Teams

Do not settle for raw numbers. RoomScore compares your room against similar spaces as the community dataset grows, unlocking percentile context once enough rooms are scanned. This is facilities management acoustic data you can use to justify budgets and prioritize upgrades.

Ready to trust the measurement?

Build a reliable acoustic baseline for every room.

Request App Store Link