Passer à la navigation principale Passer à la recherche Passer au contenu principal

DARKGAN: EXPLOITING KNOWLEDGE DISTILLATION FOR COMPREHENSIBLE AUDIO SYNTHESIS WITH GANS

  • Telecom Paris
  • Sony Computer Science Laboratory

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

9 Citations (Scopus)

Résumé

Generative Adversarial Networks (GANs) have achieved excellent audio synthesis quality in the last years. However, making them operable with semantically meaningful controls remains an open challenge. An obvious approach is to control the GAN by conditioning it on metadata contained in audio datasets. Unfortunately, audio datasets often lack the desired annotations, especially in the musical domain. A way to circumvent this lack of annotations is to generate them, for example, with an automatic audio-tagging system. The output probabilities of such systems (so-called "soft labels") carry rich information about the characteristics of the respective audios and can be used to distill the knowledge from a teacher model into a student model. In this work, we perform knowledge distillation from a large audio tagging system into an adversarial audio synthesizer that we call DarkGAN. Results show that DarkGAN can synthesize musical audio with acceptable quality and exhibits moderate attribute control even with out-of-distribution input conditioning. We release the code and provide audio examples on the accompanying website.

langue originaleAnglais
titreISMIR 2021 - The International Society For Music Information Retrieval Conference, Proceedings
EditeurInternational Society for Music Information Retrieval
Pages484-492
Nombre de pages9
ISBN (Electronique)9781732729902
étatPublié - 1 janv. 2021
Evénement22nd International Society for Music Information Retrieval Conference, ISMIR 2021 - Virtual, Online
Durée: 7 nov. 2021 → 12 nov. 2021

Série de publications

NomProceedings of the 22nd International Conference on Music Information Retrieval, ISMIR 2021

Une conférence

Une conférence22nd International Society for Music Information Retrieval Conference, ISMIR 2021
La villeVirtual, Online
période7/11/21 → 12/11/21

Empreinte digitale

Examiner les sujets de recherche de « DARKGAN: EXPLOITING KNOWLEDGE DISTILLATION FOR COMPREHENSIBLE AUDIO SYNTHESIS WITH GANS ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation