Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

From General-Purpose Audio Tagging to Spatially Grounded Sound Event Localization and Detection

About

This report investigates the extension of pretrained General-Purpose Audio Tagging (GP-AT) models toward spatially grounded Sound Event Localization and Detection (SELD). The proposed AT2SELD framework couples a pretrained AT backbone with compact First-Order Ambisonics (FOA) spatial processing, track-wise SED and Cartesian DOA estimation, permutation aware supervision, and calibration. It characterizes how semantic audio priors support localization-aware scene analysis under data, computation, and deployment constraints. The framework is developed through informed multi-stage Neural Architecture Search (NAS). Stage 1 shows that spectral FOA descriptors, based on magnitude, phase, and Intensity Vectors (IVs), provide the most reliable interface for semantic-to-spatial transfer. Stage 2 identifies early residual spatial encoding as the main capacity-sensitive component, while late track-wise abstraction and recurrent smoothing act mainly as refinement stages. Stage 3 shows that late cross-stitch coupling improves semantic-spatial interaction, whereas early fusion is costlier and less effective. Diagnostic evaluation analyzes the selected architecture under class balancing, focal loss, activity-conditioned DOA supervision, threshold calibration, and transfer across STARSS23, TAU2019, TAU-NIGENS2020, and TAU-NIGENS2021. Focal loss improves the activity point, active-only DOA supervision mitigates inactive target dominance, and validation-selected thresholds recover calibration without replacing spatial learning. Cross-dataset and oracle-activity analyses indicate strong fixed source localization on TAU2019, transferable representations from TAU NIGENS2021, and meaningful but uncertain behavior on STARSS23. Overall, GP-AT priors appear promising for SELD design when embedded in spatial-aware architectures and optimized through integrated calibration and deployment oriented strategies.

Stefano Giacomelli, Stefano Damiano, Claudia Rinaldi, Fabio Graziosi, Toon van Waterschoot• 2026

Related benchmarks

TaskDatasetResultRank
Sound Event Localization and DetectionTAU 2019 (test)
Precision (SED)95.6
1
Sound Event Localization and DetectionTAU-NIGENS 2021 (test)
Precision (SED)64.8
1
Sound Event Localization and DetectionTAU-NIGENS 2020 (test)
Precision (SED)81.1
1
Sound Event Localization and DetectionSTARSS23 (test)
P-SED84.2
1
Showing 4 of 4 rows

Other info

Follow for update