Our oceans are full of sound, and the sounds made by fish contribute significantly to underwater soundscapes. In a recent paper entitled “Applications of machine learning to identify and characterize the sounds produced by fish” (Barroso et al., 2023), Viviane Barroso, underwater acoustics expert at Diatom, and Fabio Contrera, AI and Bioacoustics Lead at Diatom, examined how artificial intelligence can assist in the detection and classification of fish sounds, thereby advancing our understanding of marine ecosystems.
Sound is now recognised as an essential ocean variable, offering valuable insights into communication, behaviour, spawning and biodiversity. However, one of the biggest challenges in applying AI is the scarcity of validated sound data for individual species. This study reviews recent advances in machine learning and deep learning to highlight the opportunities and challenges ahead and point to key directions for future research.
This work reflects our commitment to integrating marine biodiversity into practical solutions, in line with our mission to support businesses, investors, and governments in integrating marine biodiversity into their projects and becoming nature-positive in the oceans.