Shievanie Sabesan

Optimal hearing aid design through restoration of the neural code

This paper develops a deep learning-based hearing aid that restores neural coding by simulating normal auditory processing in impaired brains, outperforming existing methods in distortion correction and phoneme recognition.

Modeling neural coding in the auditory midbrain with high resolution and accuracy

The paper introduces ICNet, a deep learning model that accurately simulates neural dynamics in the auditory midbrain across various sounds, including speech.

Musical complexity governs a tradeoff between reliability and dimensionality in the neural code

The paper analyzes neural responses to complex music, finding a tradeoff between reliability and dimensionality in the neural code as musical complexity increases.

Distorted neural signal dynamics create hypersensitivity to background noise after hearing loss

The paper investigates how hearing loss distorts neural signal dynamics, leading to hypersensitivity to background noise, using brain recordings and deep neural networks.

Temporal masking and rollover in the neural code for speech with and without hearing loss

This study investigates how hearing loss affects neural encoding of speech, finding that temporal masking and rollover phenomena impact speech discrimination in both normal and hearing-impaired animals.