Realistic Gramophone Noise Synthesis Using a Diffusion Model

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

91 Downloads (Pure)

Abstract

This paper introduces a novel data-driven strategy for synthesizing gramophone noise audio textures. A diffusion probabilistic model is applied to generate highly realistic quasiperiodic noises. The proposed model is designed to generate samples of length equal to one disk revolution, but a method to generate plausible periodic variations between revolutions is also proposed. A guided approach is also applied as a conditioning method, where an audio signal generated with manually-tuned signal processing is refined via reverse diffusion to improve realism. The method has been evaluated in a subjective listening test, in which the participants were often unable to recognize the synthesized signals from the real ones. The synthetic noises produced with the best proposed unconditional method are statistically indistinguishable from real noise recordings. This work shows the potential of diffusion models for highly realistic audio synthesis tasks.
Original languageEnglish
Title of host publicationProceedings of the 25th International Conference on Digital Audio Effects (DAFx20in22)
EditorsGianpaolo Evangelista, Nicki Holighaus
Place of PublicationVienna, Austria
PublisherUniversität für Musik und darstellende Kunst Wien
Pages240-247
Number of pages8
Edition2022
ISBN (Print)978-3-200-08599-2
Publication statusPublished - 6 Sept 2022
MoE publication typeA4 Conference publication
EventInternational Conference on Digital Audio Effects - University of Music and Performing Arts Vienna, Vienna, Austria
Duration: 7 Sept 20229 Sept 2022
Conference number: 25
https://dafx2020.mdw.ac.at/DAFx20in22/
https://dafx2020.mdw.ac.at/DAFx20in22/index.html

Publication series

NameProceedings of the International Conference on Digital Audio Effects
ISSN (Print)2413-6700
ISSN (Electronic)2413-6689

Conference

ConferenceInternational Conference on Digital Audio Effects
Abbreviated titleDAFx
Country/TerritoryAustria
CityVienna
Period07/09/202209/09/2022
Internet address

Keywords

  • Audio Effects
  • Machine learning
  • Deep learning

Fingerprint

Dive into the research topics of 'Realistic Gramophone Noise Synthesis Using a Diffusion Model'. Together they form a unique fingerprint.

Cite this