Vowel Sound Synthesis from Electroencephalography during Listening and Recalling. (22nd February 2021)

• Record Type: Journal Article
• Authors: Akashi, Wataru; Kambara, Hiroyuki; Ogata, Yousuke; Koike, Yasuharu; Minati, Ludovico; Yoshimura, Natsue
• Is Part Of: Advanced intelligent systems. Volume 3:Number 2(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-02-22
• Subjects: Artificial intelligence -- Periodicals; Robotics -- Periodicals; Control theory -- Periodicals; 006.3
• DOI: 10.1002/aisy.202170023
• Languages: English
• ISSNs: 2640-4567

Abstract

Abstract : Sound Synthesis In article number 2000164, Natsue Yoshimura and co‐workers succeed to synthesize vowel speech sounds from electroencephalography recorded during listening and recalling. The sounds were sufficiently natural with recognition rates of >85%, and the brain areas extracted by convolutional neural network (CNN) were physiologically reasonable areas. The cover image represents sound signals inside and outside of the brain and a deformed structure of the CNN.