IC data from Drakopoulos et al., 2026

This data repository contains the inferior colliculus (IC) recordings and simulations that were used to evaluate the ICNet and AidNet models in Drakopoulos et al., 2026. Example Jupyter notebooks are included that can be used to replicate the main evaluation results of the paper from the provided datasets. An example AidNet model is also available via https://doi.org/10.5281/zenodo.18407090 or https://github.com/fotisdr/AidNet_example along with a simple usage guide and sound samples. Drakopoulos, F., Pellatt, L., Sabesan, S., Xia, Y., Gong, T., Fragner, A., & Lesica, N. A. (2026). Optimal hearing aid design through restoration of the neural code. bioRxiv 2026.02.02.703273; doi: 10.64898/2026.02.02.703273 ICNet evaluation The ICNet_evaluation folder includes an example Jupyter notebook that can be used to replicate the results of the ICNet model evaluation (Fig. 1c). The folder also includes a pre-compiled Python version of the notebook, as well as the ICNet_sound_data and ICNet_MUA_data matfiles that can be accessed on MATLAB or Python and contain the data corresponding to the 4 evaluation sounds. Each sound segment is included in the ICNet_sound_data matfile, is 30 s in duration and is calibrated with a 2e-5 Pa reference: speech_in_quiet: A speech segment from the UCL SCRIBE dataset consisting of sentences spoken by a male talker and presented at 60 dB SPL. speech_in_noise: A speech segment from the UCL SCRIBE dataset consisting of sentences spoken by a female talker and presented at 85 dB SPL. The speech segment was mixed in hallway noise from the Microsoft Scalable Noisy Speech dataset at 0 dB SNR. moving_ripples: Dynamic moving ripples with frequency sinusoids between 4.7 kHz and 10.8 kHz presented at 85 dB SPL. music: Three seconds from each of 10 mixed pop songs from the musdb18 dataset presented at 75 dB SPL. fs: The sampling rate of the sounds (24414.0625 Hz). The ICNet_MUA_data matfile contains the real and simulated IC activity elicited by each of the 4 sounds on two successive trials. The data is structured in 4 fields (true_trial1, true_trial2, model_trial1 and model_trial2) that contain the respective normal-hearing (NH) and hearing-impaired (HI) responses to the 4 sounds. Each response corresponds to a matrix of size [animals x time bins x channels] that contains multi-unit activity (MUA) across time bins and channels for each of the 10 NH and 10 HI animals. Each time element of the matrix indicates the spike count in one 1.3 ms bin (exactly 32 samples of the corresponding audio; MUA sampling rate fs = 24414.0625 / 32 = 762.9395 Hz). AidNet in vivo evaluation The AidNet_invivo_evaluation folder includes an example Jupyter notebook that can be used to replicate the results of the AidNet in vivo evaluation (Fig. 4c and Supplementary Fig. 6b). The folder also includes a pre-compiled Python version of the notebook, as well as the AidNet_invivo_sound_data and AidNet_invivo_MUA_data matfiles that can be accessed on MATLAB or Python and contain the data corresponding to the 4 evaluation sounds. Each sound is provided in the AidNet_invivo_sound_data matfile before and after processing with each sound strategy, with the corresponding data structured as follows: unprocessed: The 4 sounds speech_in_quiet, speech_in_noise, moving_ripples and music are provided without any processing (repeated from the matfile ICNet_sound_data). nalrp: The 4 sounds are provided for each animal of the in vivo evaluation after processing with the NAL-RP linear amplification strategy. Each data structure is a matrix of size [animals x time] that contains the processed sound for each of the 3 HI animals. nalnl2: The 4 sounds are provided after processing with the NAL-NL2 compressive amplification strategy following the same format. aidnet: The 4 sounds are provided after processing with the respective AidNet model following the same format. The AidNet_invivo_MUA_data matfile contains the real and simulated IC activity elicited by each of the unprocessed and processed sounds. The data are provided for each of the sound processing strategies explained above and are structured in 2 fields (model and true) that contain the respective normal-hearing (NH) and hearing-impaired (HI) responses to each sound (after averaging across repeated presentations and after MCA alignment). Each response corresponds to a matrix of size [animals x time bins x channels] that contains multi-unit spike means across time bins and channels for each of the 9 NH and 3 HI animals. AidNet in silico evaluation The AidNet_insilico_evaluation folder includes an example Jupyter notebook that can be used to replicate the results of the AidNet in silico evaluation (Fig. 4d). The folder also includes a pre-compiled Python version of the notebook, as well as the AidNet_insilico_sound_data and AidNet_insilico_MUA_data matfiles that can be accessed on MATLAB or Python and contain the data corresponding to the 4 evaluation sounds. Each sound is provided in the AidNet_insilico_sound_data matfile before and after processing with each sound strategy, with the corresponding data structured as follows: unprocessed: The 4 sounds speech_in_quiet, speech_in_noise, moving_ripples and music are provided without any processing. nalrp: The 4 sounds are provided for each animal of the in silico evaluation after processing with the NAL-RP linear amplification strategy. Each data structure is a matrix of size [animals x time] that contains the processed sound for each of the 14 HI animals. nalnl2: The 4 sounds are provided after processing with the NAL-NL2 compressive amplification strategy following the same format. aidnet: The 4 sounds are provided after processing with the respective AidNet model following the same format. aidnetswap: The 4 sounds are provided after processing with the swapped AidNet model following the same format. The AidNet_insilico_MUA_data matfile contains the simulated IC activity elicited for all NH and HI animals by each of the unprocessed and processed sounds. The model data are provided for each of the sound processing strategies explained above and contain the expectation over counts of the responses of each respective NH and HI ICNet model after MCA alignment. Each response corresponds to a matrix of size [animals x time bins x channels] that contains multi-unit spike means across time bins and channels for each of the 9 NH and 14 HI animals. How to run the evaluation notebooks To run the provided Jupyter notebooks, it is necessary to install the Python packages included in requirements.txt. We recommend installing conda and executing the following commands: conda create --name AidNet python=3.11conda activate AidNetpip install -r requirements.txt The first two commands are used to create (and activate) a new conda environment named AidNet with Python v3.11. If using only Python (without conda), running the last command is sufficient to install all necessary packages. The notebooks can then be executed on Jupyter lab by running jupyter lab &, selecting the corresponding notebook and running all cells. For questions, please reach out to one of the corresponding authors: Fotios Drakopoulos: [e-mail] Nicholas A Lesica: [e-mail] This work was supported by UK MRC UKRI3206, EPSRC EP/W004275/1, BBSRC BB/Y008758/1 and MRC MR/W019787/1.