| Paper ID | AUD-13.4 |
| Paper Title |
THE BENEFIT OF TEMPORALLY-STRONG LABELS IN AUDIO EVENT CLASSIFICATION |
| Authors |
Shawn Hershey, Daniel P. W. Ellis, Eduardo Fonseca, Aren Jansen, Caroline Liu, R Channing Moore, Manoj Plakal, Google, United States |
| Session | AUD-13: Detection and Classification of Acoustic Scenes and Events 2: Weak supervision |
| Location | Gather.Town |
| Session Time: | Wednesday, 09 June, 15:30 - 16:15 |
| Presentation Time: | Wednesday, 09 June, 15:30 - 16:15 |
| Presentation |
Poster
|
| Topic |
Audio and Acoustic Signal Processing: [AUD-CLAS] Detection and Classification of Acoustic Scenes and Events |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
To reveal the importance of temporal precision in ground truth audio event labels, we collected precise (∼0.1 sec resolution) “strong” labels for a portion of the AudioSet dataset. We devised a temporallystrong evaluation set (including explicit negatives of varying difficulty) and a small strong-labeled training subset of 67k clips (compared to the original dataset’s 1.8M clips labeled at 10 sec resolution). We show that fine-tuning with a mix of weak- and stronglylabeled data can substantially improve classifier performance, even when evaluated using only the original weak labels. For a ResNet50 architecture, d' on the strong evaluation data including explicit negatives improves from 1.13 to 1.41. The new labels are available as an update to AudioSet. |
