BCI Award 2018 Nomination: Neural decoding of attentional selection in multi speaker скачать в хорошем качестве

BCI Award 2018 Nomination: Neural decoding of attentional selection in multi speaker

Neural decoding of attentional selection in multi-speaker environments without access to clean sources People who suffer from hearing impairments can find it difficult to follow a conversation in a multi-speaker environment. Current hearing aids do not know who a user is paying attention to, so they cannot suppress competing talkers. Cognitively controlled hearing aids that use auditory attention decoding (AAD) methods are the next step in offering help. This submission proposed a novel framework that combines single-channel deep neural network (DNN) speech separation algorithms with AAD and presented an end-to-end system that 1) receives a single audio channel containing a mixture of speakers along with the listener’s neural signals, 2) automatically separates the speakers, 3) determines the attended speaker, and 4) amplifies that speaker. James O’Sullivan 2, Zhuo Chen 1, Jose Herrero 4, Guy M McKhann 3, Sameer A Sheth 3, Ashesh D. Mehta 4, Nima Mesgarani 1,2,5 1 Department of Electrical Engineering, Columbia University, New York, USA. 2 Mortimer B Zuckerman Mind Brain Behavior Institute, Columbia University, New York, USA. 3 Department of Neurological Surgery, The Neurological Institute, 710 West 168 Street, New York, USA. 4 Department of Neurosurgery, Hofstra-Northwell School of Medicine and Feinstein Institute for Medical Research, Manhasset, NY, USA. Read more: http://www.bci-award.com/2018