TAU-NIGENS Spatial Sound Events 2020
收藏资源简介:
TAU-NIGENS Spatial Sound Events 2020数据集由芬兰坦佩雷大学的音频与语音处理研究组创建,旨在为DCASE 2020挑战赛中的声音事件定位与检测(SELD)任务提供训练和测试数据。该数据集包含714个声音事件样本,分布在14个类别中,如警报、婴儿哭声、撞击声等。数据集通过真实的房间脉冲响应(RIRs)捕捉,模拟了多种声学条件和动态条件,如移动声源。创建过程中,使用了真实的房间脉冲响应和环境噪声,以增强数据集的真实性。该数据集适用于机器听觉领域的多种应用,如声学监控、机器人导航和智能人机交互,旨在解决声音事件的时空特性描述问题。
The TAU-NIGENS Spatial Sound Events 2020 dataset was developed by the Audio and Speech Processing Research Group at Tampere University, Finland, to provide training and test data for the Sound Event Localization and Detection (SELD) task of the DCASE 2020 Challenge. This dataset comprises 714 sound event samples spanning 14 categories, including alarms, baby cries, impact sounds, and more. Captured using real room impulse responses (RIRs), the dataset simulates a wide range of acoustic and dynamic conditions, such as moving sound sources. Real room impulse responses and environmental noise were incorporated during the dataset's development to enhance its realism. This dataset supports various applications in the machine hearing domain, including acoustic surveillance, robot navigation, and intelligent human-computer interaction, and aims to tackle the challenge of describing the spatiotemporal properties of sound events.

- 1A Dataset of Reverberant Spatial Sound Scenes with Moving Sources for Sound Event Localization and Detection坦佩雷大学,芬兰 · 2020年



