These data are under MITI license, no change, no valorisation with usage only with the citations of their joint paper and grants : "Arctic diel and circadian acoustic pattern of Orcas, Fin, and Humpback whales revealed by deep learning from two months of continuous recordings". Ecological Informatics - Under review. Data collector: Guiderdoni Julie Researchers: Girardet Justine Glotin Hervé Sarano Véronique Poupard Marion Chavin Stéphane Data recorded by the antenna designed by H. Glotin during missions in Arctic 2022-2023, with the help of ValhallaB and M. Poupard, Jean-Marc Prevot, co-granted by the AI national Chair on Bioacoustics, ADSIL, ANR-20-CHIA-0014, by AID DGA ANR (glotin@univ-tln.fr). This dataset is a sample of the YOLO detection made in Seglvik that supported the findings published in the journal article. The species_detection_sample.csv is a sample of the detection data information on which presence and detection rate measurments were based. A soundfile is provided for each detection. To find the detections corresponding to specific sound files, one should select on the voc_ID column. The file .ipynb is the code used to generate the figures and perform statistical analyzes. Copyright CIAN, Centre International d'Acoustique Naturelle, SMIoT, Université de Toulon, Toulon, France https://cian.lis-lab.fr