EUVIMEDEUVIMED entdecken
Crossref

Spectrogram-derived graphs and inductive learning for multi-label avian vocalization detection in field recordings

Noumida A iD, Rajeev Rajan

The Journal of the Acoustical Society of America · 2026

DOI: 10.1121/10.0043592 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, 2026. Verfasst von Noumida A und 1 weiteren Autor:innen. Beteiligte Einrichtung: College of Engineering Trivandrum, APJ Abdul Kalam Technological University, Thiruvananthapuram, Kerala, India. Hauptthemen laut MEDLINE: Vocalization, Animal, Birds, Deep Learning, Signal Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 60 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a methodology that employs inductive spatial geometric deep learning networks to detect multiple avian vocalizations from field recordings. Initially, a graph is constructed from the Mel-spectrogram of each audio file using a trained deep convolutional neural network (Deep CNN). The extracted features are used to build a node-feature graph, which is then processed by two spatial inductive graph-based models: graph sample and aggregation (GraphSAGE) and the graph attention network (GAT), for multi-label classification. To enhance the robustness and generalization of the Deep CNN, SpecAugment is applied to generate additional Mel-spectrograms via data augmentation. The proposed framework is evaluated on the Xeno-canto bird sound database and compared against state-of-the-art methods. The results demonstrate that the proposed inductive spatial graph-based approach outperforms existing techniques, achieving macro F1-scores of 0.90 with GraphSAGE and 0.92 with GAT. We further replaced Deep CNN with AudioProtoPNet-20 and evaluated GAT on the Xeno-canto dataset, obtaining a macro F1-score of 0.93.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Vocalization, AnimalBirdsDeep LearningSignal Processing, Computer-AssistedAnimalsSound SpectrographyConvolutional Neural NetworksAcousticsGraph Neural Networks

Autor:innen und Institutionen

  • College of Engineering Trivandrum, APJ Abdul Kalam Technological University, Thiruvananthapuram, Kerala, India.
  • Government Engineering College Idukki, APJ Abdul Kalam Technological University, Thiruvananthapuram, Kerala, India.

ORCID-Profile: Noumida A

Zitieren

APA

A, N., & Rajan, R. (2026). Spectrogram-derived graphs and inductive learning for multi-label avian vocalization detection in field recordings. The Journal of the Acoustical Society of America. https://doi.org/10.1121/10.0043592

Vancouver

A N, Rajan R. Spectrogram-derived graphs and inductive learning for multi-label avian vocalization detection in field recordings. The Journal of the Acoustical Society of America. 2026. doi:10.1121/10.0043592

BibTeX
@article{A2026,
  author = {A, Noumida and Rajan, Rajeev},
  title = {Spectrogram-derived graphs and inductive learning for multi-label avian vocalization detection in field recordings},
  journal = {The Journal of the Acoustical Society of America},
  year = {2026},
  doi = {10.1121/10.0043592}
}
Kostenlos starten →