The Differentiable Auditory Loop (DAL): An ML Framework for Hyper-Personalized Hearing Aids

2026-06-02

AI Summary

This work proposes a Differentiable Auditory Loop (DAL) framework to address the limitations of conventional hearing aids, which rely on fixed-frequency amplification strategies and fail to account for individual differences in neural encoding under complex auditory scenes. DAL uniquely integrates a differentiable human cochlear model (CARFAC) with an end-to-end waveform generation network (SEANet), enabling gradient-based optimization to align hearing aid outputs with the neural activity patterns of normal-hearing listeners. Trained on personalized neural representations and implemented in JAX, the method introduces novel loss functions based on Neural Activity Patterns (NAP) and Stable Auditory Images (SAI). Experimental results demonstrate that DAL significantly outperforms existing baselines in both neural fidelity and speech quality metrics, confirming its innovation and efficacy in personalized hearing assistance.

Abstract

Conventional hearing aids rely on fixed, frequency-dependent amplification and compression to manage reduced sensitivity, which often fails to provide sufficient listening support in complex environments, such as situations with multiple speakers (the ``cocktail party'' problem). To more comprehensively address the underlying encoding dysfunctions of hearing loss, we introduce the Differentiable Auditory Loop (DAL), a new open-source framework for personalized hearing aid design and fitting. Our first implementation of DAL incorporates CARFAC, a differentiable model of human cochlear function, which we ported to JAX, to optimize a deep neural network to match impaired auditory neural activity patterns with a normal-hearing reference. To build a hearing aid with the fine-grained spectro-temporal signal processing required, we adopt SEANet, a waveform-to-waveform fully convolutional UNet generator. We fine-tune the network by comparing the outputs of a CARFAC model fitted to normal hearing with that of a CARFAC model fitted to match each subject's individual hearing impairment. The comparison is done using loss functions derived from the respective CARFAC neural activity pattern (NAP) outputs and stabilized auditory images (SAIs), the latter providing a 2D representation that captures phase-insensitive temporal structure in the auditory nerve output. Through gradient descent, the SEANet model learns to both denoise the input and compensate for the hearing loss modelled by the impaired CARFAC model. Across neural-representation and signal-fidelity metrics, the DAL-optimized SEANet model outperformed the tested master hearing aid (MHA) baselines. The DAL framework provides a practical path toward model-based, machine-learning-driven personalization of hearing aid signal processing. Next steps include hardware deployment to enable real-world clinical testing.

Problem

• hearing aids
• cocktail party problem
• auditory encoding dysfunction
• hyper-personalization
• complex listening environments

Innovation

• Differentiable Auditory Loop
• CARFAC
• personalized hearing aid
• neural activity pattern
• SEANet