Realistic Gramophone Noise Synthesis Using a Diffusion Model

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

82 Lataukset (Pure)

Abstrakti

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.
AlkuperäiskieliEnglanti
OtsikkoProceedings of the 25th International Conference on Digital Audio Effects (DAFx20in22)
ToimittajatGianpaolo Evangelista, Nicki Holighaus
JulkaisupaikkaVienna, Austria
KustantajaUniversität für Musik und darstellende Kunst Wien
Sivut240-247
Sivumäärä8
Painos2022
ISBN (painettu)978-3-200-08599-2
TilaJulkaistu - 6 syysk. 2022
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational Conference on Digital Audio Effects - University of Music and Performing Arts Vienna, Vienna, Itävalta
Kesto: 7 syysk. 20229 syysk. 2022
Konferenssinumero: 25
https://dafx2020.mdw.ac.at/DAFx20in22/
https://dafx2020.mdw.ac.at/DAFx20in22/index.html

Julkaisusarja

NimiProceedings of the International Conference on Digital Audio Effects
ISSN (painettu)2413-6700
ISSN (elektroninen)2413-6689

Conference

ConferenceInternational Conference on Digital Audio Effects
LyhennettäDAFx
Maa/AlueItävalta
KaupunkiVienna
Ajanjakso07/09/202209/09/2022
www-osoite

Sormenjälki

Sukella tutkimusaiheisiin 'Realistic Gramophone Noise Synthesis Using a Diffusion Model'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä