2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

Technical Program

AUD-1: Audio and Speech Source Separation 1: Speech Separation

Session Type: Poster
Time: Tuesday, 8 June, 13:00 - 13:45
Location: Gather.Town
Session Chair: Zhuo Chen, Microsoft Corporation
 
AUD-1.1: RETHINKING THE SEPARATION LAYERS IN SPEECH SEPARATION NETWORKS
         Yi Luo; Columbia University
         Zhuo Chen; Microsoft Corporation
         Cong Han; Columbia University
         Chenda Li; Shanghai Jiao Tong University
         Tianyan Zhou; Microsoft Corporation
         Nima Mesgarani; Columbia University
 
AUD-1.2: ON PERMUTATION INVARIANT TRAINING FOR SPEECH SOURCE SEPARATION
         Xiaoyu Liu; Dolby Laboratories
         Jordi Pons; Dolby Laboratories
 
AUD-1.3: COUNT AND SEPARATE: INCORPORATING SPEAKER COUNTING FOR CONTINUOUS SPEAKER SEPARATION
         Zhong-Qiu Wang; The Ohio State University
         DeLiang Wang; The Ohio State University
 
AUD-1.4: ULTRA-LIGHTWEIGHT SPEECH SEPARATION VIA GROUP COMMUNICATION
         Yi Luo; Columbia University
         Cong Han; Columbia University
         Nima Mesgarani; Columbia University
 
AUD-1.5: ATTENTION IS ALL YOU NEED IN SPEECH SEPARATION
         Cem Subakan; Mila
         Mirco Ravanelli; Mila
         Samuele Cornell; Università Politecnica delle Marche (IT)
         Mirko Bronzi; Mila
         Jianyuan Zhong; University of Rochester
 
AUD-1.6: MULTICHANNEL OVERLAPPING SPEAKER SEGMENTATION USING MULTIPLE HYPOTHESIS TRACKING OF ACOUSTIC AND SPATIAL FEATURES
         Aidan Hogg; Imperial College London
         Christine Evers; University of Southampton
         Patrick A. Naylor; Imperial College London