{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T17:45:54Z","timestamp":1778694354102,"version":"3.51.4"},"reference-count":26,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,22]]},"DOI":"10.1109\/apsipaasc65261.2025.11249371","type":"proceedings-article","created":{"date-parts":[[2025,11,28]],"date-time":"2025-11-28T18:40:26Z","timestamp":1764355226000},"page":"95-100","source":"Crossref","is-referenced-by-count":1,"title":["Towards Robust Stereo 3-D SELD: A Study of Perceptual Features and Data Augmentation"],"prefix":"10.1109","author":[{"given":"Jun-Wei","family":"Yeow","sequence":"first","affiliation":[{"name":"Nanyang Technological University,Smart Nation TRANS Lab,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ee-Leng","family":"Tan","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,Smart Nation TRANS Lab,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Santi","family":"Peksi","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,Smart Nation TRANS Lab,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Woon-Seng","family":"Gan","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,Smart Nation TRANS Lab,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qirui","family":"Huang","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,Smart Nation TRANS Lab,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2018.2885636"},{"key":"ref2","article-title":"Real-time sound event localization and detection: Deployment challenges on edge devices","author":"Yeow","year":"2024","journal-title":"arXiv preprint"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.23919\/EUSIPCO63174.2024.10715220"},{"key":"ref4","first-page":"41","article-title":"Baseline models and evaluation of sound event localization and detection with distance estimation in dcase2024 challenge","volume-title":"Proceedings of the Detection and Classification of Acoustic Scenes and Events 2024 Workshop (DCASE2024)","author":"Diaz-Guerra","year":"2024"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2019.2900164"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TENSYMP52854.2021.9551007"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2015.2418571"},{"key":"ref8","first-page":"1347","article-title":"Blind estimation of the coherent-to-diffuse energy ratio from noisy speech signals","volume-title":"2011 19th European Signal Processing Conference","author":"Jeub","year":"2011"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.23919\/EUSIPCO55093.2022.9909581"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746132"},{"key":"ref11","article-title":"Starss23: An audio-visual dataset of spatial recordings of real scenes with spatiotemporal annotations of sound events","volume":"36","author":"Shimada","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.33682\/3qgs-e216"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3256088"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680"},{"key":"ref15","article-title":"Improved regularization of convolutional neural networks with cutout","author":"DeVries","year":"2017","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3173054"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747680"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10446118"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413609"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746384"},{"key":"ref21","article-title":"Squeeze-and-excite resnet-conformers for sound event localization, detection, and distance estimation for dcase2024 challenge","author":"Yeow","year":"2024","journal-title":"DCASE2024 Challenge, Tech. Rep."},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/jsen.2025.3583033"},{"key":"ref23","article-title":"Spatial and semantic embedding integration for stereo sound event localization and detection in regular videos","author":"Berghi","year":"2025","journal-title":"DCASE2025 Challenge, Tech. Rep."},{"key":"ref24","article-title":"The nerc-slip system for stereo sound event localization and detection in regular video content of dcase 2025 challenge","author":"Wang","year":"2025","journal-title":"DCASE2025 Challenge, Tech. Rep."},{"key":"ref25","article-title":"Improving stereo 3d sound event localization and detection: Perceptual features, stereo-specific data augmentation, and distance normalization","author":"Yeow","year":"2025","journal-title":"DCASE2025 Challenge, Tech. Rep."},{"key":"ref26","article-title":"Stereo sound event localization and detection with source distance estimation using data-driven resnet-conformer ensemble","author":"He","year":"2025","journal-title":"DCASE2025 Challenge, Tech. Rep."}],"event":{"name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","location":"Singapore, Singapore","start":{"date-parts":[[2025,10,22]]},"end":{"date-parts":[[2025,10,24]]}},"container-title":["2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11248853\/11248968\/11249371.pdf?arnumber=11249371","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:51:06Z","timestamp":1764399066000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11249371\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1109\/apsipaasc65261.2025.11249371","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}