Making Weak Supervision Interactive: Exploring Transfer from Sound Libraries to Passive Acoustic Monitoring Data

Mammadli, Novruz; Saghir, Rida; Ali, Kanwar Ammar; Doddanawar, Prathmesh; S. Gouvêa, Thiago; Sonntag, Daniel

doi:10.24963/ijcai.2026/979

Making Weak Supervision Interactive: Exploring Transfer from Sound Libraries to Passive Acoustic Monitoring Data

Novruz Mammadli, Rida Saghir, Kanwar Ammar Ali, Prathmesh Doddanawar, Thiago S. Gouvêa, Daniel Sonntag

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Demo Track. Pages 8458-8461.

https://doi.org/10.24963/ijcai.2026/979

Keywords

AI: Humans and AI
AI: Multidisciplinary Topics and Applications
AI: Machine Learning

Abstract

Passive Acoustic Monitoring (PAM), an increasingly popular method for wildlife monitoring, generates large volumes of data whose analysis depends on instance-level annotations that are costly to obtain. Archival sound collections provide weak labels that lack temporal localisation. In prior work, we demonstrated that Multiple Instance Learning (MIL) can extract approximate event locations from weakly labelled PAM data, suggesting it may be applied to sound collection data.

This demo operationalizes that approach within an interactive workflow that connects weakly annotated sound collections to downstream PAM deployment. The system supports configurable MIL-based localisation, lightweight interactive refinement, and transfer to an independent PAM dataset.

We carried out a preliminary evaluation with an actual sound library from a museum collection and a benchmark PAM dataset. Results confirm that weakly annotated sound collections can serve as a viable training signal for downstream PAM detection and illustrate differences between alternative MIL instantiations under real transfer conditions. (Video available at https://cst.dfki.de/projects-weak-supervision-demo)

Novruz Mammadli, Rida Saghir, Kanwar Ammar Ali, Prathmesh Doddanawar, Thiago S. Gouvêa, and Daniel Sonntag. 2026. Making Weak Supervision Interactive: Exploring Transfer from Sound Libraries to Passive Acoustic Monitoring Data. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 8458–8461. https://doi.org/10.24963/ijcai.2026/979

@inproceedings{ijcai2026p979,
  title     = {Making Weak Supervision Interactive: Exploring Transfer from Sound Libraries to Passive Acoustic Monitoring Data},
  author    = {Mammadli, Novruz and Saghir, Rida and Ali, Kanwar Ammar and Doddanawar, Prathmesh and S. Gouvêa, Thiago and Sonntag, Daniel},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {8458--8461},
  year      = {2026},
  month     = {8},
  note      = {Demo Track},
  doi       = {10.24963/ijcai.2026/979},
  url       = {https://doi.org/10.24963/ijcai.2026/979},
}

TY  - CPAPER
TI  - Making Weak Supervision Interactive: Exploring Transfer from Sound Libraries to Passive Acoustic Monitoring Data
AU  - Mammadli, Novruz
AU  - Saghir, Rida
AU  - Ali, Kanwar Ammar
AU  - Doddanawar, Prathmesh
AU  - S. Gouvêa, Thiago
AU  - Sonntag, Daniel
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 8458
EP  - 8461
DO  - 10.24963/ijcai.2026/979
UR  - https://doi.org/10.24963/ijcai.2026/979
ER  -