Ground truth for DCASE2025Task4EvaluationDataset
收藏Zenodo2025-08-05 更新2026-05-26 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.16736215
下载链接
链接失效反馈官方服务：
资源简介：
This dataset provides the ground truth data for the DCASE2025Task4EvaluationDataset. It contains the ground truth separation sounds for the target sound events in the 2290 mixtures of evaluation set.
This dataset is intended to be used in conjunction with the following datasets and the script:
Training and validation data: DCASE2025Task4Dataset: The Dataset for Spatial Semantic Segmentation of Sound Scenes
Evaluation data: DCASE2025Task4EvaluationDataset : The Evaluation Dataset for Spatial Semantic Segmentation of Sound Scenes
Script for evaluation: nttrd-cslabs/dcase2025_SpatialSemanticSegmentation_evaluator
Other resources:
Description page of DCASE2025 task4: Spatial Semantic Segmentation of Sound Scenes - DCASE
Baseline System Github page: nttcslab/dcase2025_task4_baseline
arXiv paper describing this dataset: [2506.10676] Description and Discussion on DCASE 2025 Challenge Task 4: Spatial Semantic Segmentation of Sound Scenes
提供机构：
Zenodo创建时间：
2025-08-05
