Spatial Annotation-Free Sound Event Localization and Detection via Spatial Instance Classification

Sound Event Localization and Detection (SELD) typically relies on strongly labeled data that specify when, where, and which sounds occur, resulting in high annotation costs. To address this challenge, we present a framework that enables SELD training without spatial labels, termed spatial annotation...

Бүрэн тодорхойлолт

Номзүйн дэлгэрэнгүй

Үндсэн зохиолчид:	Masahiro Yasuda, Noboru Harada, Shoichiro Saito, Nobutaka Ono
Формат:	Artigo
Хэл сонгох:	Inglês
Хэвлэсэн:	IEEE 2025-01-01
Цуврал:	IEEE Access
Нөхцлүүд:	Sound event localization and detection DOA estimation weakly-supervised learning multiple instance learning beamforming
Онлайн хандалт:	https://ieeexplore.ieee.org/document/11162519/
Шошгууд:	Шошго нэмэх Шошго байхгүй, Энэхүү баримтыг шошголох эхний хүн болох!