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

Abstract

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.

Analysis and acoustic event classification of environmental data collected in Sons al Balcó Project

“…A total of 365 videos were obtained during the 2020 campaign and another 237 during the subsequent 2021 one. The sounds in these videos were manually annotated using a hierarchical taxonomy [12]. The resulting compilations are complex polyphonic datasets with several sounds overlapping.…”

Predicting the Perceptual Rating of a Soundscape Using Artificial Intelligence

“…The data used to develop the automatic tool will come from the research project Sons al Balcó, which was a citizen science project that gathered more than 600 videos together with a perceptual questionnaire in different spots of Catalonia [2] during 2020 and 2021. In previous works [1], a neural network that detected acoustic events was trained using the sound from these videos. However, this network was not able to determine the quality of the soundscape of the video.…”

“…Acoustic event detection and classification is a research topic that the scientific community has widely explored in the last years [1]. Even though the field is still under development, automatic AI tools to detect acoustic events have become more available.…”