Forest Sound Classification Dataset: FSC22

Abstract

The study of environmental sound classification (ESC) has become popular over the years due to the intricate nature of environmental sounds and the evolution of deep learning (DL) techniques. Forest ESC is one use case of ESC, which has been widely experimented with recently to identify illegal activities inside a forest. However, at present, there is a limitation of public datasets specific to all the possible sounds in a forest environment. Most of the existing experiments have been done using generic environment sound datasets such as ESC-50, U8K, and FSD50K. Importantly, in DL-based sound classification, the lack of quality data can cause misguided information, and the predictions obtained remain questionable. Hence, there is a requirement for a well-defined benchmark forest environment sound dataset. This paper proposes FSC22, which fills the gap of a benchmark dataset for forest environmental sound classification. It includes 2025 sound clips under 27 acoustic classes, which contain possible sounds in a forest environment. We discuss the procedure of dataset preparation and validate it through different baseline sound classification models. Additionally, it provides an analysis of the new dataset compared to other available datasets. Therefore, this dataset can be used by researchers and developers who are working on forest observatory tasks.

A Comparative Study of Preprocessing and Model Compression Techniques in Deep Learning for Forest Sound Classification

“…The Forest Sound Classification dataset (FSC22) [ 19 ] comprises 2025 labeled sound clips in a forest environment. Each audio clip is standardized to a length of 5 s, sampled at a rate of 44.1 kHz, and stored in the WAV file format.…”

“…Henceforth, we conduct a comparative analysis of seven CNNs, namely ACDNet, AlexNet, ResNet-50, DenseNet-121, Inception-v3, MobileNet-v3-small, and EfficientNet-v2-B0 to exhibit the state-of-the-art [ 18 ]. The workflow involves the utilization of the FSC22 dataset [ 19 ], which is a dataset specifically created for forest sound data, subjected to preprocessing, followed by successive stages of data augmentation and feature extraction. Subsequently, the CNN models undergo training with k-fold cross-validation.…”

ESC-NAS: Environment Sound Classification Using Hardware-Aware Neural Architecture Search for the Edge

“…The refined dataset consisted of 1950 audio clips related to forest environments distributed across 26 classes, with each class comprising 75 audio clips. Each audio clip is 5 s long and sampled at a 44.1 kHz sampling rate [ 39 ].…”

Sound classification with time-frequency features in forest environment

“…However, these two classes of datasets cannot reflect the real forest acoustic environment with good quality. Recently, a forest sound classification dataset (FSC22) was built [7] containing five classes of sounds that possibly exist in a forest environment.…”