Blog
About

3
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Speaker-independent auditory attention decoding without access to clean speech sources

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Our system separates simultaneous voices and compares them with brain waves of a listener to amplify attended speech.

          Abstract

          Speech perception in crowded environments is challenging for hearing-impaired listeners. Assistive hearing devices cannot lower interfering speakers without knowing which speaker the listener is focusing on. One possible solution is auditory attention decoding in which the brainwaves of listeners are compared with sound sources to determine the attended source, which can then be amplified to facilitate hearing. In realistic situations, however, only mixed audio is available. We utilize a novel speech separation algorithm to automatically separate speakers in mixed audio, with no need for the speakers to have prior training. Our results show that auditory attention decoding with automatically separated speakers is as accurate and fast as using clean speech sounds. The proposed method significantly improves the subjective and objective quality of the attended speaker. Our study addresses a major obstacle in actualization of auditory attention decoding that can assist hearing-impaired listeners and reduce listening effort for normal-hearing subjects.

          Related collections

          Most cited references 44

          • Record: found
          • Abstract: found
          • Article: not found

          Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature.

          Precise localization of sulco-gyral structures of the human cerebral cortex is important for the interpretation of morpho-functional data, but requires anatomical expertise and is time consuming because of the brain's geometric complexity. Software developed to automatically identify sulco-gyral structures has improved substantially as a result of techniques providing topologically correct reconstructions permitting inflated views of the human brain. Here we describe a complete parcellation of the cortical surface using standard internationally accepted nomenclature and criteria. This parcellation is available in the FreeSurfer package. First, a computer-assisted hand parcellation classified each vertex as sulcal or gyral, and these were then subparcellated into 74 labels per hemisphere. Twelve datasets were used to develop rules and algorithms (reported here) that produced labels consistent with anatomical rules as well as automated computational parcellation. The final parcellation was used to build an atlas for automatically labeling the whole cerebral cortex. This atlas was used to label an additional 12 datasets, which were found to have good concordance with manual labels. This paper presents a precisely defined method for automatically labeling the cortical surface in standard terminology. Copyright 2010 Elsevier Inc. All rights reserved.
            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Beamforming: a versatile approach to spatial filtering

              Bookmark
              • Record: found
              • Abstract: not found
              • Article: not found

              The expectation-maximization algorithm

               T.K. Moon (1996)
                Bookmark

                Author and article information

                Journal
                Sci Adv
                Sci Adv
                SciAdv
                advances
                Science Advances
                American Association for the Advancement of Science
                2375-2548
                May 2019
                15 May 2019
                : 5
                : 5
                Affiliations
                [1 ]Department of Electrical Engineering, Columbia University, New York, NY, USA.
                [2 ]Zuckerman Mind Brain Behavior Institute, Columbia University, New York, NY, USA.
                [3 ]Department of Neurosurgery, Hofstra-Northwell School of Medicine and Feinstein Institute for Medical Research, Manhasset, New York, NY, USA.
                Author notes
                [*]

                These authors contributed equally to this work.

                []Corresponding author. Email: nima@ 123456ee.columbia.edu
                Article
                aav6134
                10.1126/sciadv.aav6134
                6520028
                Copyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial License 4.0 (CC BY-NC).

                This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.

                Funding
                Funded by: doi http://dx.doi.org/10.13039/100000052, NIH Office of the Director;
                Funded by: doi http://dx.doi.org/10.13039/100000179, NSF Office of the Director;
                Categories
                Research Article
                Research Articles
                SciAdv r-articles
                Applied Sciences and Engineering
                Neuroscience
                Neuroscience
                Custom metadata
                Eunice Diego

                Comments

                Comment on this article