核心概念
Wearable systems for sound event localization and detection require consideration of self-motion for improved performance.
統計資料
"Our dataset provides recordings of acoustic events around a subject moving at 6DoF."
"The proposed method utilizes dynamic cues by applying excitations to the acoustic features in accordance with the velocity and angular velocity extracted from the sensor signals."
"Validation experiments on our dataset showed that learning the system using a dataset that includes self-motion improves SELD performance during movement."
引述
"Conventional SELD tasks have dealt only with microphone arrays located in static positions."
"The proposed method effectively improves SELD performance with a mechanism to extract acoustic features conditioned by sensor signals."
"The dataset is divided into three subsets ('stat.', '3DoF', and '6DoF') in accordance with the self-motion condition."