Invariance learning for a music indexing robust to sound modifications 51:59

  • Saison 2023-2024 - None - None > Réunion du Groupement de Recherche "Information Signal Image viSion" : Traitement du signal pour l'audio et l'écoute artificielle - Musique
  • Nov. 16, 2023
  • Ircam, Paris
Participants
  • Rémi Mignot (conférencier)

Music indexing allows the finding of music excerpts among a large music catalog and the detection of duplicates. With the rise of social media, it is more and more important for music owners to detect misuse and illegal use of their music. The main difficulty of this task is the detection of music excerpts when they have been strongly modified, intentionnally or not. To deal with this issue, the presented method is based on an audio representation relevant to the music content and robust to some sound modifications. Then, using a data augmentation approach, a discriminant transformation is learnt to improve the robustness of the compact representation. Finally, a hash function is derived to allow a fast searching with a large catalog together with a robustness to bit corruption.

Réunion du Groupement de Recherche "Information Signal Image viSion" : Traitement du signal pour l'audio et l'écoute artificielle - Musique

From the same archive

GDR ISIS, Méthodes et modèles en traitement de signal, Introduction 05:24

Hybrid deep learning for music analysis and synthesis 53:30

Invariance learning for a music indexing robust to sound modifications 51:59

Basic Pitch: A lightweight model for multi-pitch, note and pitch bend estimations in polyphonic music 43:52

Labeling a Large Music Catalog 01:04:50