Skip to main content
zenodorestricted

The Hume Vocal Burst Competition Dataset (H-VB) | Raw Data [A-VB: updated 03.01.22]

<p>This package includes the raw data for a subset of The Hume Vocal Burst Database (H-VB).</p> <p>This dataset&nbsp;contains&nbsp;<strong>59,201&nbsp;audio recordings of vocal bursts from&nbsp;1,702&nbsp;speakers</strong>, from 4 cultures, the&nbsp;<strong>U.S.A, South Africa, China,&nbsp;</strong>and<strong>&nbsp;Venezuela</strong>, ranging in age from&nbsp;<strong>20&nbsp;to 39.5 years old</strong>.</p> <p>The total duration of data in this version of&nbsp;<strong>H-VB&nbsp;is 36 Hours&nbsp;</strong>&nbsp;(Mean: 02.23 &nbsp;sec). A subset of 10 emotional intensity labels is also available. Data has been partitioned into equal training, validation, and test sets, and the test set is blind.&nbsp;&nbsp;</p> <p><strong>Emotion Labels:&nbsp;</strong><br> <em>Awe,&nbsp;Excitement,&nbsp;Amusement,&nbsp;Awkwardness,&nbsp;Fear,&nbsp;Horror,&nbsp;Distress,&nbsp;Triumph,&nbsp;Sadness,&nbsp;Surprise, Valence, Arousal</em></p> <p>For further questions about the data contact:&nbsp;<strong>competitions@hume.ai</strong></p>

ShareScore

12/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship
4
Harmonization
4
Access
0
Reuse readiness
0
Engagement
4