Location as Supervision for Weakly Supervised Multi-Channel Source Separation of Machine Sounds

Ricardo Falcon-Perez*, Gordon Wichern, Francois G. Germain, Jonathan Le Roux

*Tämän työn vastaava kirjoittaja

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

46 Lataukset (Pure)

Abstrakti

In this work, we are interested in learning a model to separate sources that cannot be recorded in isolation, such as parts of a machine that must run simultaneously in order for the machine to function. We assume the presence of a microphone array and knowledge of the source locations (potentially obtained from schematics or an auxiliary sensor such as a camera). Our method uses the source locations as weak labels for learning to separate the sources, since we cannot obtain the isolated source signals typically used as training targets. We propose a loss function that requires the directional features computed from the separated sources to match the true direction of arrival for each source, and also include a reconstruction loss to ensure all frequencies are taken into account by at least one of the separated sources output by our model. We benchmark the performance of our algorithm using synthetic mixtures created using machine sounds from the DCASE 2021 Task 2 dataset in challenging reverberant conditions. While reaching lower objective scores than a model with access to isolated source signals for training, our proposed weakly-supervised model obtains promising results and applies to industrial scenarios where collecting isolated source signals is prohibitively expensive or impossible.

AlkuperäiskieliEnglanti
OtsikkoProceedings of the 2023 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA 2023
KustantajaIEEE
ISBN (elektroninen)979-8-3503-2372-6
DOI - pysyväislinkit
TilaJulkaistu - 2023
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE Workshop on Applications of Signal Processing to Audio and Acoustics - New Paltz, Yhdysvallat
Kesto: 22 lokak. 202325 lokak. 2023

Julkaisusarja

NimiIEEE Workshop on Applications of Signal Processing to Audio and Acoustics
Vuosikerta2023-October
ISSN (painettu)1931-1168
ISSN (elektroninen)1947-1629

Workshop

WorkshopIEEE Workshop on Applications of Signal Processing to Audio and Acoustics
LyhennettäWASPAA
Maa/AlueYhdysvallat
KaupunkiNew Paltz
Ajanjakso22/10/202325/10/2023

Sormenjälki

Sukella tutkimusaiheisiin 'Location as Supervision for Weakly Supervised Multi-Channel Source Separation of Machine Sounds'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä