Training Sound Event Detection On A Heterogeneous Dataset - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2020

Training Sound Event Detection On A Heterogeneous Dataset

Résumé

Training a sound event detection algorithm on a heterogeneous dataset including both recorded and synthetic soundscapes that can have various labeling granularity is a non-trivial task that can lead to systems requiring several technical choices. These technical choices are often passed from one system to another without being questioned. We propose to perform a detailed analysis of DCASE 2020 task 4 sound event detection baseline with regards to several aspects such as the type of data used for training, the parameters of the mean-teacher or the transformations applied while generating the synthetic soundscapes. Some of the parameters that are usually used as default are shown to be sub-optimal.
Fichier principal
Vignette du fichier
main_SED.pdf (215.06 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02891665 , version 1 (07-07-2020)
hal-02891665 , version 2 (10-09-2020)

Identifiants

Citer

Nicolas Turpault, Romain Serizel. Training Sound Event Detection On A Heterogeneous Dataset. 2020. ⟨hal-02891665v1⟩
739 Consultations
1019 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More