Improving Primate Sounds Classification using Binary Presorting for Deep Learning
This work addresses the challenge of improving classification accuracy for wildlife conservation efforts, though it is incremental as it builds on existing CNN and data-augmentation techniques.
The paper tackled the problem of classifying primate sounds from noisy and weakly labeled audio datasets by introducing a binary pre-sorting method on MEL spectrograms, resulting in significantly higher Accuracy and UAR scores compared to baseline models on the ComparE 2021 dataset.
In the field of wildlife observation and conservation, approaches involving machine learning on audio recordings are becoming increasingly popular. Unfortunately, available datasets from this field of research are often not optimal learning material; Samples can be weakly labeled, of different lengths or come with a poor signal-to-noise ratio. In this work, we introduce a generalized approach that first relabels subsegments of MEL spectrogram representations, to achieve higher performances on the actual multi-class classification tasks. For both the binary pre-sorting and the classification, we make use of convolutional neural networks (CNN) and various data-augmentation techniques. We showcase the results of this approach on the challenging \textit{ComparE 2021} dataset, with the task of classifying between different primate species sounds, and report significantly higher Accuracy and UAR scores in contrast to comparatively equipped model baselines.