Description of the repository: The goal of our paper (https://doi.org/10.3389/fneur.2022.934650) was to segment the acoustic radiation (AR), one of the most important white matter fiber bundles of the hearing system. This repository contains the segmentations masks of the AR we created from 105 subjects of the Human Connectome Project (HCP) young adult dataset (https://www.humanconnectome.org/study/hcp-young-adult). These subjects are exactly the same used by Wasserthal et al. (2018) https://doi.org/10.1016/j.neuroimage.2018.07.070. In the file "data_training.tar.gz", one directory was created per HCP subject. Every directory contains the file "bundle_masks_AR.nii.gz" that contains the binary masks for the left and right AR. In our paper, we used these masks to train TractSeg. The file best_weights_ep110.npz contains the weights after training TractSeg that can be used in inference for targeting the AR. For using these weights on new data, one can use TractSeg with the option "--exp_name best_weights_ep110.npz". Please read the documentation of TractSeg and our paper for more information. If you use the training data or the pre-trained network, please cite our publication: Malin Siegbahn, Cecilia Engmér Berglin, Rodrigo Moreno. Automatic segmentation of the core of the acoustic radiation in humans. Frontiers in Neurology (2022) 13:934650. doi: 10.3389/fneur.2022.934650