| pretty_name: DCASE 2016 Task 2 (mirror) | |
| license: other | |
| license_name: research-use | |
| task_categories: | |
| - audio-classification | |
| tags: | |
| - sound-event-detection | |
| - dcase2016 | |
| # DCASE 2016 Task 2 — mirror | |
| Mirror of the DCASE 2016 Task 2 dataset (sound event detection in synthetic | |
| audio), as distributed for the HEAR benchmark. Research use only; credit the | |
| original DCASE organisers. | |
| - Task page: <https://dcase.community/challenge2016/task-sound-event-recognition-in-synthetic-audio> | |
| - Authors: Mesaros, Heittola, Virtanen et al. | |
| - 18 event classes, strong labels, 44.1 kHz (resampled as needed). | |
| No authorship of the recordings is claimed. If the original organisers object, | |
| this mirror will be removed. | |