Effective individual identification is essential for monitoring and conserving animal populations. In bottlenose dolphins (Tursiops truncatus), signature whistles serve as reliable acoustic cues for individual recognition. This study explores three complementary approaches to classify and identify these distinctive vocalizations. First, we applied a manual visual classification method based on comparisons with a signature whistle catalogue. Second, we used an unsupervised neural network algorithm (ARTWARP) to automatically group whistle contours by shape similarity. Lastly, we analyzed whistle sequences at the bout level to assess patterns of vocal usage over time. Together, these approaches provide a comprehensive framework for evaluating signature whistle identification performance in free-ranging male bottlenose dolphins. This work lays the foundation for developing scalable and objective tools for passive acoustic monitoring, with potential applications in dolphin conservation and behavioral ecology.
promotor/supervisor feedback
nothing yet
