AT2SELD extends pretrained audio tagging backbones to SELD via FOA descriptors, track-wise processing, permutation-aware supervision, and staged NAS on multiple datasets.
Permutation Invariant Recurrent Neural Networks for Sound Source Tracking Applications,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From General-Purpose Audio Tagging to Spatially Grounded Sound Event Localization and Detection
AT2SELD extends pretrained audio tagging backbones to SELD via FOA descriptors, track-wise processing, permutation-aware supervision, and staged NAS on multiple datasets.