Parsing Birdsong with Deep Audio Embeddings
This work addresses a domain-specific problem for conservationists and ecologists by providing incremental improvements in audio analysis for biodiversity monitoring.
The authors tackled the problem of automating bird population monitoring by developing a semi-supervised approach to identify bird calls and noise in audio data, showing improved classification precision and insights into dataset structure.
Monitoring of bird populations has played a vital role in conservation efforts and in understanding biodiversity loss. The automation of this process has been facilitated by both sensing technologies, such as passive acoustic monitoring, and accompanying analytical tools, such as deep learning. However, machine learning models frequently have difficulty generalizing to examples not encountered in the training data. In our work, we present a semi-supervised approach to identify characteristic calls and environmental noise. We utilize several methods to learn a latent representation of audio samples, including a convolutional autoencoder and two pre-trained networks, and group the resulting embeddings for a domain expert to identify cluster labels. We show that our approach can improve classification precision and provide insight into the latent structure of environmental acoustic datasets.