{"type":"ScholarlyArticle","title":"Optimizing Spectrogram Resolution and Training Strategies for Real-Time Killer Whale Call Type Classification","authors":[{"name":"Vladislav Naumov","slug":"vladislav-naumov","given":"Vladislav","family":"Naumov","affiliation":""},{"name":"Iaroslav Sheipak","slug":"iaroslav-sheipak","given":"Iaroslav","family":"Sheipak","affiliation":""},{"name":"Yuriy Ivanov","slug":"yuriy-ivanov","given":"Yuriy","family":"Ivanov","affiliation":""},{"name":"Ilya Makarov","slug":"ilya-makarov","given":"Ilya","family":"Makarov","affiliation":""}],"abstract":"Automated identification of killer whale call types from continuous acoustic recordings is essential for scalable population monitoring, yet existing general-purpose frameworks such as ANIMAL-SPOT suffer from suboptimal spectrogram resolution and lack strategies tailored to the spectral-temporal characteristics of killer whale vocalizations. We identify two key limitations of the ANIMAL-SPOT framework: (1) no possibility to choose different architecture of CNN backbone, and (2) the absence of regularization and class-balancing techniques limits generalization across call types with varying abundance. To address these issues, we propose a framework that combines optimized STFT parameters (FFT size 1024, hop length 172) with label smoothing and targeted oversampling, evaluated across three CNN backbones and five segment lengths. On a dataset of 12 killer whale vocalization classes from Avacha Gulf, Russia, our best configuration (ResNet-18 with 1200 ms segments) achieves 97.1% segment-level accuracy, compared to 96.2% for the ANIMAL-SPOT baseline — a relative error reduction of 23.7%. We further present an interactive demonstration system for uploading recordings and obtaining time-resolved call-type predictions, enabling rapid analysis of passive acoustic monitoring data.","doi":"10.24963/ijcai.2026/982","url":"https://ijcai.org/proceedings/2026/982","pdf":"https://ijcai.org/proceedings/2026/0982.pdf","markdown":"https://ijcai.org/proceedings/2026/982.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/982","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Demo Track","track":"Demo Track","number":982,"pages":{"first":8471,"last":8474},"keywords":["AI: Humans and AI","AI: Machine Learning","AI: Multidisciplinary Topics and Applications"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}