AES-2id: Proper Impulse Response Measurement for Sound Reinforcement

What AES-2id Covers

AES-2id:2023 "AES information document for room acoustics and sound reinforcement systems — characterization and measurement of surface scattering using an impulse response" is the AES's guidance document for acoustic measurement in sound reinforcement applications. Unlike ISO 3382 (focused on room acoustic parameters), AES-2id addresses the practical needs of sound system engineers: system alignment, EQ, coverage verification, and troubleshooting.

The document is an "information document" (id), not a standard — it provides recommendations and best practices rather than mandatory requirements. However, it represents the consensus of the professional audio engineering community.

Recommended Excitation Signals

AES-2id recommends the exponential sine sweep (ESS) as the preferred excitation signal for impulse response measurement:

• Frequency range: 20 Hz to 20 kHz minimum (full audible range)
• Duration: ≥5 seconds for general measurement, ≥10 seconds for high-precision RT60 work
• Level: At least 10 dB above ambient noise across the full frequency range
• Deconvolution: Frequency-domain convolution with the inverse filter (time-reversed sweep with amplitude compensation)

MLS is acceptable as an alternative for rapid multi-position surveys where time-variance is controlled. Impulsive sources (balloon pops) are mentioned as field expedients but not recommended for formal measurement due to poor SNR and spectral inconsistency.

Source and Receiver Placement

Source (Speaker) Placement

For system measurement, the source is the installed sound system itself. The measurement signal is sent through the system processor, amplifiers, and speakers — measuring the complete signal chain.

For room-only measurement (without the sound system), an omnidirectional source is recommended. The AES document references ISO 3382-1 placement guidelines: minimum 2 m from any wall, 1.5 m height.

Receiver (Microphone) Placement

• Ear height: 1.2 m for seated audience, 1.5 m for standing
• Minimum positions: 5 across the primary coverage area (center, left, right, near, far at minimum)
• Distance from boundaries: At least 1 m from any wall to avoid near-field effects
• Microphone type: Omnidirectional measurement microphone calibrated per IEC 61094

Measurement Validation

AES-2id emphasizes validation of measurement quality before interpreting results:

INR (Impulse-to-Noise Ratio)

The ratio between the peak of the impulse response and the noise floor, measured in the octave band of interest. Requirements:

• ≥20 dB for basic frequency response measurement
• ≥35 dB for T20 extraction
• ≥45 dB for T30 extraction
• ≥50 dB for precise C80 and Ts measurement

SonaVyx reports INR per octave band after each measurement. The impulse response tool color-codes bands by INR: green (≥45 dB), amber (35-45 dB), red (<35 dB).

Coherence Validation

For transfer function measurements, coherence validates the quality at each frequency bin. AES-2id recommends:

• γ² > 0.8 for EQ decisions
• γ² > 0.5 for coverage assessment
• Report coherence alongside all TF measurements

Linear Regression Quality

For RT60 extraction from the Schroeder curve, the linear regression r² should exceed 0.995. Values below 0.99 indicate non-exponential decay or noise contamination. The RT60 tool displays r² per octave band.

Derived Parameters

From a valid impulse response, AES-2id recognizes the following derived parameters as relevant to sound reinforcement:

• Frequency response: Magnitude of the FFT of the windowed IR
• Propagation delay: Time of the first significant IR peak (used for system alignment)
• RT60: From Schroeder integration per ISO 3382
• C50/C80: Clarity ratios from early/late energy partition
• STI: From IR per IEC 60268-16 clause A.4
• Early reflection pattern: Individual reflections visible in the IR waveform

SonaVyx Implementation

SonaVyx follows AES-2id recommendations throughout:

• Default excitation: log sine sweep, 10 seconds, 20 Hz-20 kHz
• Frequency-domain deconvolution with inverse filter in Rust WASM
• INR computation and display per octave band
• Coherence display on all transfer function measurements
• r² quality indicator on RT60 per octave band
• Delay finder with millisecond and distance readout
• All derived parameters (RT60, C50, C80, D50, Ts, STI) computed on-device

The room analysis workflow and PA tuning workflow both incorporate AES-2id best practices as default settings, ensuring measurement quality even for users unfamiliar with the document.