A Strongly-Labelled Polyphonic Dataset of Urban Sounds with Spatiotemporal Context

Ooi, Kenneth; Watcharasupat, Karn N.; Peksi, Santi; Karnapi, Furi Andi; Ong, Zhen-Ting; Chua, Danny; Leow, Hui-Wen; Kwok, Li-Long; Ng, Xin-Lei; Loh, Zhen-Ann; Gan, Woon‐Seng

doi:10.48550/arxiv.2111.02006

Cited by 1 publication

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the USM-SED dataset generated 20,000 polyphonic soundscapes mixing sounds from the FSD50K dataset [ 31 ]. On the other hand, some datasets have been compiled through a set of recordings, usually made by a single user or network such as SONYC-UST [ 32 ] consisting of 3068 recordings from an acoustic sensor network deployed in New York City including 23 different sound classes tagged by volunteers or SINGA:PURA [ 33 ] with a total of 18.2 h of audio data recorded through a wireless acoustic sensor network deployed in the city of Singapore. Both SONYC-UST and SINGA:PURA tagged the sound events using a hierarchical taxonomy.…”

Section: Related Workmentioning

confidence: 99%

Analysis and Acoustic Event Classification of Environmental Data Collected in a Citizen Science Project

Bonet-Solà

Vidaña-Vila

Alsina-Pagès

2023

IJERPH

View full text Add to dashboard Cite

Citizen science can serve as a tool to obtain information about changes in the soundscape. One of the challenges of citizen science projects is the processing of data gathered by the citizens, to obtain conclusions. As part of the project Sons al Balcó, authors aim to study the soundscape in Catalonia during the lockdown due to the COVID-19 pandemic and afterwards and design a tool to automatically detect sound events as a first step to assess the quality of the soundscape. This paper details and compares the acoustic samples of the two collecting campaigns of the Sons al Balcó project. While the 2020 campaign obtained 365 videos, the 2021 campaign obtained 237. Later, a convolutional neural network is trained to automatically detect and classify acoustic events even if they occur simultaneously. Event based macro F1-score tops 50% for both campaigns for the most prevalent noise sources. However, results suggest that not all the categories are equally detected: the percentage of prevalence of an event in the dataset and its foregound-to-background ratio play a decisive role.

show abstract

Section: Related Workmentioning

confidence: 99%