SPE-45.4
MIXED PRECISION DNN QUANTIZATION FOR OVERLAPPED SPEECH SEPARATION AND RECOGNITION
Junhao Xu, Xunying Liu, Helen Mei-Ling Meng, The Chinese University of Hong Kong, China; Jianwei Yu, Tencent AI Lab, China
Session:
Speech Enhancement: Speech Extraction and Audio-visual Enhancement
Track:
Speech and Language Processing
Location:
Gather Area B
Presentation Time:
Tue, 10 May, 23:00 - 23:45 China Time (UTC +8)
Tue, 10 May, 15:00 - 15:45 UTC
Tue, 10 May, 15:00 - 15:45 UTC
Session Chair:
Jun Du, University of Science and Technology of China
Session SPE-45
SPE-45.1: L-SpEx: Localized Target Speaker Extraction
Meng Ge, Longbiao Wang, Jianwu Dang, Tianjin University, China; Chenglin Xu, Kuaishou Technology, China; Eng Siong Chng, Nanyang Technological University, Singapore; Haizhou Li, National University of Singapore, Singapore
SPE-45.3: DPCCN: DENSELY-CONNECTED PYRAMID COMPLEX CONVOLUTIONAL NETWORK FOR ROBUST SPEECH SEPARATION AND EXTRACTION
Jiangyu Han, Yanhua Long, Shanghai Normal University, China; Lukas Burget, Jan Cernocky, Brno University of Technology, Czechia
SPE-45.4: MIXED PRECISION DNN QUANTIZATION FOR OVERLAPPED SPEECH SEPARATION AND RECOGNITION
Junhao Xu, Xunying Liu, Helen Mei-Ling Meng, The Chinese University of Hong Kong, China; Jianwei Yu, Tencent AI Lab, China
SPE-45.5: THE IMPACT OF REMOVING HEAD MOVEMENTS ON AUDIO-VISUAL SPEECH ENHANCEMENT
Zhiqi Kang, Radu Horaud, Xavier Alameda-Pineda, Inria Grenoble Rhône-Alpes & Univ. Grenoble Alpes, France, France; Mostafa Sadeghi, Inria Nancy Grand-Est, France, France; Jacob Donley, Anurag Kumar, Facebook Reality Labs Research, Redmond WA, USA, United States of America
SPE-45.6: VSEGAN: VISUAL SPEECH ENHANCEMENT GENERATIVE ADVERSARIAL NETWORK
Xinmeng Xu, Trinity College Dublin, China; Yang Wang, Dongxiang Xu, Yiyuan Peng, Cong Zhang, Jie Jia, Binbin Chen, Vivo Communication Technology Co. Ltd., China