SpeakerBeam: A new deep learning technology for extracting speech of a target speaker based on the speaker's voice characteristics

作     者：Delcroix, Marc Zmolikova, Katerina Kinoshita, Keisuke Araki, Shoko Ogawa, Atsunori Nakatani, Tomohiro 

作者机构：Signal Processing Research Group Media Information Laboratory NTT Communication Science Laboratories Japan Brno University of Technology Czech Republic 

出 版 物：《NTT Technical Review》 (NTT Tech. Rev.)

年 卷 期：2018年第16卷第11期

页      面：19-24页

核心收录：

主　　题：Audition 

摘      要：In a noisy environment such as a cocktail party, humans can focus on listening to a desired speaker, an ability known as selective hearing. Current approaches developed to realize computational selective hearing require knowing the position of the target speaker, which limits their practical usage. This article introduces SpeakerBeam, a deep learning based approach for computational selective hearing based on the characteristics of the target speaker s voice. SpeakerBeam requires only a small amount of speech data from the target speaker to compute his/her voice characteristics. It can then extract the speech of that speaker regardless of his/her position or the number of speakers talking in the background. © 2018 Nippon Telegraph and Telephone *** right reserved.