A sea mammal expert has developed a groundbreaking tool that can track dolphins by their regional accents. Dr. Julie Oswald, of the University of St Andrews' Scottish Oceans Institute, created the tool known as the Real-time Odontocete Call Classification Algorithm (Rocca) using AI.
Rocca categorizes dolphin calls by species and is tailored to different geographical areas. With around 42 species of dolphins using hundreds of sounds to communicate, this tool enhances understanding of their communication.
Dolphins learn to mimic sounds from a young age, developing unique signature whistles to announce their identity. Traditionally, scientists have used hydrophones—underwater microphones—to study dolphin sounds, but analyzing these can be challenging.
Rocca simplifies this process by applying machine learning to detect subtle acoustic differences, allowing for the identification of dolphin species in specific areas. Dolphins develop region-specific accents, meaning a bottlenose dolphin in the North Sea may produce different sounds compared to one in the Pacific.
This innovation aids conservationists in accurately identifying dolphin species and assessing the impact of human activities like fishing and sonar disturbances. Dr. Oswald is expanding Rocca's applications to other species and ecosystems, including regions in West Africa and the Mediterranean.
Rocca was recognized as one of the winners of the EarthRanger 2024 Conservation Technology Award, which honors innovative solutions that make measurable conservation impacts.