Hearing Loss
Hearing loss research focuses on improving communication and quality of life for individuals with impaired hearing. Current efforts utilize machine learning, particularly deep neural networks and various dimensionality reduction techniques, to predict hearing thresholds from neuroimaging data, enhance speech intelligibility in noisy environments, and personalize audio processing for music and speech. These advancements aim to create more effective hearing aids and assistive listening technologies, ultimately improving speech recognition and overall listening experience for a broader population, including those with mild to moderate hearing loss.
Papers
Speech foundation models on intelligibility prediction for hearing-impaired listeners
Santiago Cuervo, Ricard Marxer
Non-Intrusive Speech Intelligibility Prediction for Hearing-Impaired Users using Intermediate ASR Features and Human Memory Models
Rhiannon Mogridge, George Close, Robert Sutherland, Thomas Hain, Jon Barker, Stefan Goetze, Anton Ragni