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

SPE-5: Speech Enhancement 1: Speech Separation

Session Type: Poster
Time: Tuesday, 8 June, 14:00 - 14:45
Location: Gather.Town
Session Chair: Takuya Yoshioka, Microsoft Research
 
SPE-5.1: DUAL-PATH MODELING FOR LONG RECORDING SPEECH SEPARATION IN MEETINGS
         Chenda Li; Shanghai Jiao Tong University
         Zhuo Chen; Microsoft
         Yi Luo; Columbia University
         Cong Han; Columbia University
         Tianyan Zhou; Microsoft
         Keisuke Kinoshita; NTT Corporation
         Marc Delcroix; NTT Corporation
         Shinji Watanabe; Johns Hopkins University
         Yanmin Qian; Shanghai Jiao Tong University
 
SPE-5.2: TIME-DOMAIN LOSS MODULATION BASED ON OVERLAP RATIO FOR MONAURAL CONVERSATIONAL SPEAKER SEPARATION
         Hassan Taherian; The Ohio State Universtiy
         DeLiang Wang; The Ohio State Universtiy
 
SPE-5.3: CONTINUOUS SPEECH SEPARATION WITH CONFORMER
         Sanyuan Chen; Harbin Institute of Technology
         Yu Wu; Microsoft Corporation
         Zhuo Chen; Microsoft Corporation
         Jian Wu; Microsoft Corporation
         Jinyu Li; Microsoft Corporation
         Takuya Yoshioka; Microsoft Corporation
         Chengyi Wang; Microsoft Corporation
         Shujie Liu; Microsoft Corporation
         Ming Zhou; Microsoft Corporation
 
SPE-5.4: A FLOW-BASED NEURAL NETWORK FOR TIME DOMAIN SPEECH ENHANCEMENT
         Martin Strauss; International Audio Laboratories Erlangen
         Bernd Edler; International Audio Laboratories Erlangen
 
SPE-5.5: SANDGLASSET: A LIGHT MULTI-GRANULARITY SELF-ATTENTIVE NETWORK FOR TIME-DOMAIN SPEECH SEPARATION
         Max W. Y. Lam; Tencent AI Lab
         Jun Wang; Tencent AI Lab
         Dan Su; Tencent AI Lab
         Dong Yu; Tencent AI Lab
 
SPE-5.6: TRANSMASK: A COMPACT AND FAST SPEECH SEPARATION MODEL BASED ON TRANSFORMER
         Zining Zhang; National University of Singapore, PVoice technology
         Bingsheng He; National University of Singapore
         Zhenjie Zhang; PVoice technology