Electroencephalography-Based Auditory Attention Decoding: Toward Neurosteered Hearing Devices

People suffering from hearing impairment often have difficulties participating in conversations so-called cocktail party scenarios where multiple individuals are simultaneously talking. Although advanced algorithms exist to suppress background noise these situations, a device also needs information about which speaker user actually aims attend to. The voice of the correct (attended) can then be enhanced through this information, and all other speakers treated as noise. Recent neuroscientific advances shown that it is possible determine focus auditory attention noninvasive neurorecording techniques, such electroencephalography (EEG). Based on insights, multitude decoding (AAD) has been proposed, could, combined with appropriate separation miniaturized EEG sensors, lead neurosteered devices. In article, we provide broad review statistically grounded comparative study EEG-based AAD address main signal processing challenges field.