From speech to primate vocalizations: self-supervised deep learning as a comparative approach - Laboratoire d'informatique fondamentale de Marseille Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

From speech to primate vocalizations: self-supervised deep learning as a comparative approach

Résumé

The deep learning revolution partly embodied in transformers architectures and pre-trained self-supervised models opens many perspectives in the study of linguistics and animal communication. By exploring transfer learning approaches for computational bioacoustics applied to primate vocalizations, we delve into the explainability of pre-trained speech models to understand what they can teach us about the origins of language. To examine divergences and similarities between speech and primate vocalizations from a deep learning perspective, our method consists in probing and fine-tuning experiments based on self-supervised acoustic models. By analyzing their ability to process primate vocalizations, we test the effect of models' architectures, pre-training datasets, and task specificities on their transfer learning performance. In doing so, we want to evaluate the validity of deep transfer learning as a scientific tool in the study of the origins of language from a comparative standpoint.
Fichier principal
Vignette du fichier
evolang.pdf (58.97 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04608044 , version 1 (13-06-2024)

Identifiants

  • HAL Id : hal-04608044 , version 1

Citer

Jules Cauzinille, Benoît Favre, Ricard Marxer, Arnaud Rey. From speech to primate vocalizations: self-supervised deep learning as a comparative approach. Evolang 2024, May 2024, Madison, United States. ⟨hal-04608044⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More