Sound event localization and detection (SELDnet) results

Description

This package is part of the work - Joint Measurement of Localization and Detection of Sound Events presented in WASPAA 2019. This package consists of results from the SELDnet method for joint sound event localization and detection. The results corresponding to different training states of 5, 25 and 75 epochs are provided here.  These results are for the four cross-validation splits of the TAU Spatial Sound Events 2019 - Microphone Array dataset. The sound events in this dataset consist of stationary point sources from multiple sound classes each associated with a temporal onset and offset time, and DOA coordinate represented using azimuth and elevation angle. This TAU Spatial Sound Events 2019 - Microphone Array dataset is part of the DCASE 2019 Sound Event Localization and Detection Task and can be downloaded here. Each of the results folders consists of 400 files, corresponding to the results of the individual recordings of the TAU Spatial Sound Events 2019 - Microphone Array dataset. This data collection received funding from the European Research Council, grant agreement 637422 EVERYSOUND. Download instructions The three files,  mic_dev_5, mic_dev_25, and mic_dev_75, correspond to SELDnet results at 5, 25 and 75 epochs. Download the zip files and use your favorite compression tool to unzip these split zip files.

Resources

Name Format Description Link
0 http://data.europa.eu/88u/dataset/oai-zenodo-org-3354709
0 http://data.europa.eu/88u/dataset/oai-zenodo-org-3354709
0 http://data.europa.eu/88u/dataset/oai-zenodo-org-3354709
0 http://data.europa.eu/88u/dataset/oai-zenodo-org-3354709
0 http://data.europa.eu/88u/dataset/oai-zenodo-org-3354709

Tags

  • localization-and-detection
  • sound-source-localization
  • doa-estimation
  • sound-event-detection
  • sound-event-localization-and-detection

Topics

Categories