3MAS: a multitask, multilabel, multidataset semi-supervised audio segmentation model - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

3MAS: a multitask, multilabel, multidataset semi-supervised audio segmentation model

Résumé

When processing audio data, multiple challenges arise, one of them being the diversity of information present in the audio signal. Various audio segmentation subtasks appeared includ- ing voice activity detection (VAD), overlapped speech detection (OSD), music or noise detection. These tasks are often com- pleted by separate models trained on different datasets, thus in- creasing computational costs and limiting the usage to specific datasets. We first show that a multiclass VAD and OSD model outperforms state of the art models. Then, we propose 3MAS, a novel deep learning-based audio segmentation model capable of handling multiple datasets, and assessing multiple simulta- neously as a multilabel segmentation problem. 3MAS provides similar performances as specialized models with a similar archi- tecture and can be trained using partial and unbalanced annota- tions on different datasets. 3MAS is a gain in computational time, and opens new opportunities to include new labels.
Fichier principal
Vignette du fichier
segmentation_odyssey2024-v2.pdf (303.29 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04591532 , version 1 (28-05-2024)

Licence

Identifiants

  • HAL Id : hal-04591532 , version 1

Citer

Martin Lebourdais, Pablo Gimeno, Théo Mariotte, Marie Tahon, Alfonso Ortega, et al.. 3MAS: a multitask, multilabel, multidataset semi-supervised audio segmentation model. Speaker and Language Recognition Workshop - Odyssey, Jun 2024, Québec (CA), Canada. ⟨hal-04591532⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More