Description
This record archives the source code and a representative subset of the data accompanying the paper: "Passive acoustic identification of social groups in the Hainan gibbon."
The repository contains all scripts required for data preprocessing, feature extraction, model development, training, evaluation, uncertainty quantification and reproduction of the computational workflow described in the above publication.
To facilitate testing and reproducibility while maintaining a manageable Github archive size, only a representative subset of the dataset is included in this repository. This subset is sufficient to verify the installation, execute the analysis pipeline, and reproduce example experiments.
The complete dataset used in the study above is archived separately and distributed through a dedicated Zenodo record:
Full dataset: https://doi.org/10.5281/zenodo.20834429
Researchers wishing to reproduce the full experimental results reported in the publication should download the complete dataset from the dataset archive and follow the instructions provided in the repository documentation.
Reproducibility
This software release has been designed to support transparent and reproducible research. The included sample data enables users to validate the implementation and run the complete workflow, while the separate full dataset archive provides access to all gibbon social group recordings, phrase and segment annotations, and train-validation-test data splits required to reproduce the results presented in the associated publication.
The repository contains all scripts required for data preprocessing, feature extraction, model development, training, evaluation, uncertainty quantification and reproduction of the computational workflow described in the above publication.
To facilitate testing and reproducibility while maintaining a manageable Github archive size, only a representative subset of the dataset is included in this repository. This subset is sufficient to verify the installation, execute the analysis pipeline, and reproduce example experiments.
The complete dataset used in the study above is archived separately and distributed through a dedicated Zenodo record:
Full dataset: https://doi.org/10.5281/zenodo.20834429
Researchers wishing to reproduce the full experimental results reported in the publication should download the complete dataset from the dataset archive and follow the instructions provided in the repository documentation.
Reproducibility
This software release has been designed to support transparent and reproducible research. The included sample data enables users to validate the implementation and run the complete workflow, while the separate full dataset archive provides access to all gibbon social group recordings, phrase and segment annotations, and train-validation-test data splits required to reproduce the results presented in the associated publication.
| Date made available | 24 Jun 2026 |
|---|---|
| Publisher | Zenodo |