EEG-based detection of the locus of auditory attention with convolutional neural networks

Autor: Lucas Deckers, Neetha Das, Tom Francart, Servaas Vandecappelle, Alexander Bertrand, Amir Hossein Ansari
Rok vydání: 2018
Předmět:
Male
Hearing aid
Computer science
medicine.medical_treatment
Speech recognition
02 engineering and technology
Electroencephalography
hearing aid
Convolutional neural network
neuro-steered auditory prosthesis
0302 clinical medicine
Auditory attention
convolutional neural networks
Attention
EEG
Biology (General)
medicine.diagnostic_test
General Neuroscience
General Medicine
Tools and Resources
Interval (music)
Sound
Speech Perception
Medicine
medicine.symptom
Human
brain waves
QH301-705.5
Hearing loss
Science
0206 medical engineering
Locus (genetics)
General Biochemistry
Genetics and Molecular Biology
brain-computer interfaces
03 medical and health sciences
medicine
Humans
Speech
hearing loss
Brain–computer interface
General Immunology and Microbiology
auditory attention detection
020601 biomedical engineering
Acoustic Stimulation
Neural Networks
Computer
Locus (mathematics)
030217 neurology & neurosurgery
Neuroscience
Zdroj: eLife
eLife, Vol 10 (2021)
DOI: 10.1101/475673
Popis: In a multi-speaker scenario, the human auditory system is able to attend to one particular speaker of interest and ignore the others. It has been demonstrated that it is possible to use electroencephalography (EEG) signals to infer to which speaker someone is attending by relating the neural activity to the speech signals. However, classifying auditory attention within a short time interval remains the main challenge. We present a convolutional neural network-based approach to extract the locus of auditory attention (left/right) without knowledge of the speech envelopes. Our results show that it is possible to decode the locus of attention within 1-2 s, with a median accuracy of around 81%. These results are promising for neuro-steered noise suppression in hearing aids, in particular in scenarios where per-speaker envelopes are unavailable. ispartof: Elife vol:10 ispartof: location:England status: Published online
Databáze: OpenAIRE
Externí odkaz: