math-rir

Room Impulse Response analysis using the SSIR (Spatial Segmentation of Impulse Response) method.

Based on: Pawlak & Lee, "Spatial segmentation of impulse response for room reflection analysis and auralization", Applied Acoustics 249 (2026).

What it does

Segments a measured Room Impulse Response (RIR) into consecutive, variable-length sound events (direct sound + early reflections), each with a constant direction of arrival (DOA). This preserves the full temporal energy profile while enabling per-reflection analysis and manipulation.

Key algorithms

• Direct sound onset detection (Algorithm 1 from paper): log-magnitude peak detection with 11 dB threshold
• Local Energy Ratio reflection detection: 1ms analysis windows, median energy threshold, DOA/TOA validation
• Onset refinement: per-segment onset detection within pre-onset windows
• Mixing time estimation: Abel & Huang echo density method

API

use math_rir::{analyze_rir, analyze_srir, SsirConfig};

// Mono RIR (energy-based detection only)
let result = analyze_rir(&rir_samples, &SsirConfig::new(48000.0));

// B-format SRIR (full DOA estimation + validation)
let result = analyze_srir(&[&w, &x, &y, &z], &SsirConfig::new(48000.0));

println!("{} events, mixing time {:.1}ms",
    result.num_events(), result.mixing_time_ms());

Used by

• sotf-plugin-binaural: SSIR-analyzed measured room reflections with per-reflection HRTF
• autoeq/roomeq: SSIR-informed correction weights for room EQ optimization