# General

This folder contains all the data used for the project. To use it for training the AE and classification net this folder should be copied inside the `freqbinauto-encoder` folder.

# Structure
* The `training_data` folder contains 70% of all defective and intact samples each.  
* The `validation_data` folder contains 15% of all defective and intact samples each.  
* The `test_data` folder contains 15% of all defective and intact samples each.  
* The `noise_data` folder contains different noise files which were used for the augmentation.  
* The `raw_measurement_data` folder contains the .wav files as they were exported from the soundbook.  
* The `unaugmented_data` folder contains the clipped .wav files from each sample.

**NOTE:** Each subfolder has its own `README` file with more detailed information. If there are still questions left feel free to write an email to: finn.meding@stud.tu-darmstadt.de

# License

This dataset is licensed under a [Creative Commons Attribution 4.0 International License (CC BY 4.0)](https://creativecommons.org/licenses/by/4.0/). 

You are free to share and adapt the material for any purpose, even commercially, as long as you give appropriate credit, provide a link to the license, and indicate if changes were made.

If you use this dataset in your research, please cite our paper:
> Feldmann, R., Meding, F., & Melz, T. (2026). Audio Signal-Based Delamination Detection for Wind Turbine Rotor Blades Using an Autoencoder. *Technical University of Darmstadt*.

**BibTeX:**
```bibtex
@article{feldmann2026,
  title={Audio Signal-Based Delamination Detection for Wind Turbine Rotor Blades Using an Autoencoder},
  author={Feldmann, Robert and Meding, Finn and Melz, Tobias},
  year={2026},
  institution={Technical University of Darmstadt, System Reliability, Adaptive Structures and Machine Acoustics SAM}
}