SPE-1.4
ALL-NEURAL BEAMFORMER FOR CONTINUOUS SPEECH SEPARATION
Zhuohuang Zhang, Indiana University Bloomington, United States of America; Takuya Yoshioka, Naoyuki Kanda, Zhuo Chen, Xiaofei Wang, Dongmei Wang, Sefik Emre Eskimez, Microsoft, United States of America
Session:
Speech Separation
Track:
Speech and Language Processing
Location:
Gather Area B
Presentation Time:
Sun, 8 May, 20:00 - 20:45 China Time (UTC +8)
Sun, 8 May, 12:00 - 12:45 UTC
Sun, 8 May, 12:00 - 12:45 UTC
Session Chair:
Reinhold Haeb-Umbach, Paderborn University
Session SPE-1
SPE-1.1: CONTINUOUS SPEECH SEPARATION WITH RECURRENT SELECTIVE ATTENTION NETWORK
Yixuan Zhang, The Ohio State University, United States of America; Zhuo Chen, Jian Wu, Takuya Yoshioka, Peidong Wang, Zhong Meng, Jinyu Li, Microsoft, United States of America
SPE-1.2: SA-SDR: A NOVEL LOSS FUNCTION FOR SEPARATION OF MEETING STYLE DATA
Thilo von Neumann, Christoph Boeddeker, Reinhold Haeb-Umbach, Paderborn University, Germany; Keisuke Kinoshita, Marc Delcroix, NTT Corporation, Japan
SPE-1.3: VARARRAY: ARRAY-GEOMETRY-AGNOSTIC CONTINUOUS SPEECH SEPARATION
Takuya Yoshioka, Xiaofei Wang, Dongmei Wang, Min Tang, Zirun Zhu, Zhuo Chen, Naoyuki Kanda, Microsoft, United States of America
SPE-1.4: ALL-NEURAL BEAMFORMER FOR CONTINUOUS SPEECH SEPARATION
Zhuohuang Zhang, Indiana University Bloomington, United States of America; Takuya Yoshioka, Naoyuki Kanda, Zhuo Chen, Xiaofei Wang, Dongmei Wang, Sefik Emre Eskimez, Microsoft, United States of America
SPE-1.5: SAGRNN: SELF-ATTENTIVE GATED RNN FOR BINAURAL SPEAKER SEPARATION WITH INTERAURAL CUE PRESERVATION
Ke Tan, Buye Xu, Anurag Kumar, Eliya Nachmani, Yossi Adi, Meta Platforms, Inc., United States of America
SPE-1.6: MINING HARD SAMPLES LOCALLY AND GLOBALLY FOR IMPROVED SPEECH SEPARATION
Kai Wang, Yizhou Peng, Hao Huang, Ying Hu, Xinjiang University, China; Sheng Li, National Institute of Information and Communications Technology, China