2021 IEEE International Conference on Acoustics, Speech and Signal Processing
6-11 June 2021 • Toronto, Ontario, Canada
Extracting Knowledge from Information
SPE-5: Speech Enhancement 1: Speech Separation |
| Session Type: Poster |
| Time: Tuesday, 8 June, 14:00 - 14:45 |
| Location: Gather.Town |
| Virtual Session: View on Virtual Platform |
| 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 |