Acouvero
Download
DEEP READ · 10 MIN

Why your
room rewrites
the music
you hear.

Speakers do half the work; your room does the other half. It decides which low frequencies in the 20–500 Hz range survive, how much the 4–80 ms reflections blur the image, and how long the room keeps ringing after each note.

This piece walks through all of that in plain language — and how Acouvero turns it into a few things you can actually do.

CH · 01

How does sound
reach your ears?

TL;DR

Your speakers emit one sound — you hear dozens: the direct version, plus dozens of copies bouncing back from walls, floor, ceiling, furniture and windows, all layered on top of each other.

Picture your speakers as a stone dropped in a pond. Waves spread out from the splash. With nothing in the way, you see clean concentric ripples. But your living room isn't a pond — it has walls. Waves hit the walls, bounce back, and pile onto the next outgoing waves. What you see isn't neat ripples anymore — it's a tangle.

Your ears work in a similar way. The "direct sound" from the speakers reaches you within 3–10 ms; in the 4–80 ms that follow come the early reflections — copies bouncing off the nearest walls and furniture. After that, the reverberant tail: hundreds of reflections layered into a soft hum that can last 200–800 ms before it dies away.

Your brain handles these reflections in a curious way: below 5 ms, you don't hear them as separate — they just colour the timbre; in the 4–30 ms range, your brain fuses them with the direct sound into one source, but the position drifts; only past 50 ms do you actually hear an "echo". Acouvero cares about the middle band — the reflections you don't hear but that completely reshape what you feel.

Analogy It's like lighting a candle in a room. The direct light lets you see things, but the light bouncing off the walls sets the mood. Paint the walls black or white and the candle hasn't changed — yet what you see is completely different.
IMPULSE RESPONSE · TIMELINE DIRECT + EARLY + LATE DIRECT 0 MS EARLY REFLECTIONS 4 MS 28 MS LATE REVERBERATION 500 MS THE REGION ACOUVERO CARES ABOUT AMP 0
DIRECT SOUND
0–3ms · direct
EARLY REFLECTIONS
4–80ms · early
LATE REVERB
80–800ms · tail
CH · 02

The three
things a room does.

TL;DR

Boost: some low frequencies "lock in" and get amplified by 10+ dB.
Smear: reflections fight the direct sound — image floats.
Drag: the note ends, but the room keeps ringing.

They're unrelated — and you have to measure and treat each separately.

First: boost (room modes). Below 300 Hz, sound wavelengths are the same order of magnitude as the room itself — a 3.5 m living room maps roughly to a 49 Hz half-wavelength. At that frequency, sound bounces between walls and stacks on top of itself, getting amplified by 8–15 dB at some positions and cancelled at others. That's why bass slams your chest in one chair and vanishes one metre away.

Second: smear (early reflections). Sound from the speakers comes back almost instantly off walls, ceiling and floor. The earliest copies (within 4–30 ms) get fused with the direct sound by your brain, into a single source — but the position drifts. A stereo system should put vocals dead-centre between the speakers; you might hear them slightly left, with a halo around them. That's early reflections at work.

Third: drag (reverberation time / RT60). After a sound stops, the reflected field doesn't vanish — it decays. The professional name is RT60: how long it takes the energy to fall by 60 dB. Bedrooms run 0.3–0.4 s. Carpeted living rooms 0.4–0.6 s. Hard-surfaced open-plan spaces 0.8–1.2 s. When RT60 is too long, one note's tail drags into the next note's attack — and everything blurs.

Why 60 dB Human hearing has roughly a 60 dB dynamic range — drop 1000× quieter than the source and you basically can't hear it anymore. RT60 was proposed in 1922 by Wallace Sabine, the founder of architectural acoustics. A hundred years and counting.
THREE SYMPTOMS · COMPARED SCHEMATIC 01 · BASS BOOM 0 +10 +9 DB @ 49 HZ 20 HZ 500 HZ FREQUENCY RESPONSE 02 · BLURRED IMAGING +5 MS REFLECTION PATHS 03 · MUDDY DECAY RT60 = 0.78 S 0 MS 1000 MS DECAY ENVELOPE
CH · 03

How do we
measure all this?

TL;DR

Play a smooth 20 Hz → 20 kHz sweep, record it on the iPhone, then use math (deconvolution) to extract what the room did from the recording.

There are three classic measurement methods: impulse (a gunshot), white noise (a "shhh"), and sweep (a slow rise from low to high). The first two are fast but fragile — your AC unit kicking in ruins the data. Sweep is the opposite: slower (60 s) but the signal-to-noise ratio is excellent at every frequency — an elevator rumble won't derail it. That's why Acouvero uses sweep.

The math is called log-sine sweep deconvolution (Farina, 2000). Plain version: you know what the speakers played (your sweep), you know what you recorded (sweep + room), divide one by the other and what's left is pure room response — the technical term is impulse response (IR).

Once you have the IR, the rest is math:

  • FFT the IR → frequency response curve, showing which frequencies got boosted or buried
  • Find reflection peaks in the first 80 ms of the IR → early-reflection map showing which wall each bounce came from
  • Reverse-integrate the IR envelope → RT60 decay time
Why iPhone Since iPhone 12, the built-in mic is flat to ±1.5 dB between 80 Hz and 18 kHz omni. We've built a calibration curve per generation; Acouvero applies it automatically. Pro measurement mics (e.g. Earthworks M30) plug in over USB-C / Lightning and push accuracy a notch higher.
DECONVOLUTION · LOG SINE SWEEP FARINA · 2000 01 · INPUT · The sweep you play 20 HZ → 20 KHZ · 60 S 02 · RECORDED · What the mic hears SWEEP × ROOM = ? ÷ INPUT 03 · IMPULSE RESPONSE · The room's "fingerprint" IR — where all analysis starts
SAMPLE RATE
48kHz
SNR · SIGNAL-NOISE
80dB+
ON-DEVICE PROCESSING
0bytes uploaded
CH · 04

The order of fixes
actually matters.

RULE OF ORDER

First move, then absorb, finally EQ.
Reverse the order and you waste money for worse results.

Step 1: move (placement, $0). Low-frequency standing waves are decided mostly by where the speakers stand and where you sit. Every room has "bass dead zones" (cancellation) and "bass hotspots" (reinforcement). Find a better spot and most of the problem disappears. Acouvero computes an overall optimum — usually a 20–60 cm nudge, no renovation needed.

Step 2: absorb (treatment, €30–€3,000). After the move, early reflections are still there. Now you target absorption at the reflection points — the bounces arriving in the 4–80 ms window. Acouvero computes the exact spots (mirror method between walls) so a 60×60 cm panel is usually enough — no need to wrap the entire wall.

Step 3: EQ ($0, if you have a digital source). With the first two steps done, the leftover low-frequency standing waves (≤300 Hz) can be EQ'd. Important: only fix what's left, only in the bass, only standing-wave-shaped peaks. EQ can't fix early reflections, and it can't fix reverberation time.

Why order matters A lot of people start with EQ and flatten the bass. The catch: the curve you measured is at your current position — move one metre and it changes, and your EQ is now garbage. Lock in the physical setup first; then the EQ stays useful.
FIX ORDER · MOVE → PANEL → EQ EFFECTIVENESS PER COST 01 · MOVE · $0 BEFORE AFTER Fixes: 60% of bass standing waves 02 · PANEL · tens to thousands ¥ BEFORE AFTER Fixes: 80% of early reflections 03 · EQ · $0 (with a DSP) BEFORE AFTER Fixes: 100% of the remaining bass

Each step's "%" is its share of the matching problem class: moving furniture handles ~60 % of the bass-mode mess, the rest waits for panels and EQ. Doing all three takes you from "3-point speakers, 5-point room" to "5-point speakers, 8-point room".

iOS · App Store

Done reading?
Now go measure your room.

That's all the theory. The rest is putting the iPhone where you usually sit, tapping once, and seeing — 60 seconds later — what you're actually hearing.