SPE-76.4
OMNI-SPARSITY DNN: FAST SPARSITY OPTIMIZATION FOR ON-DEVICE STREAMING E2E ASR VIA SUPERNET
Haichuan Yang, Yuan Shangguan, Dilin Wang, Meng Li, Pierce Chuang, Xiaohui Zhang, Ganesh Venkatesh, Ozlem Kalinli, Vikas Chandra, Facebook, United States of America
Session:
Resource Constrained Speech Modeling
Track:
Speech and Language Processing
Location:
Gather Area C
Presentation Time:
Thu, 12 May, 23:00 - 23:45 China Time (UTC +8)
Thu, 12 May, 15:00 - 15:45 UTC
Thu, 12 May, 15:00 - 15:45 UTC
Session Chair:
Shi-Xiong Zhang, Tencent
Session SPE-76
SPE-76.1: END-TO-END LOW RESOURCE KEYWORD SPOTTING THROUGH CHARACTER RECOGNITION AND BEAM-SEARCH RE-SCORING
Ephrem Tibebe Mekonnen, University of Trento, Italy; Alessio Brutti, Daniele Falavigna, Fondazione Bruno Kessler, Italy
SPE-76.2: CURRICULUM OPTIMIZATION FOR LOW-RESOURCE SPEECH RECOGNITION
Anastasia Kuznetsova, Francis Tyers, Indiana University Bloomington, United States of America; Anurag Kumar, Jennifer Drexler Fox, Rev.com, United States of America
SPE-76.3: EXPLORING EFFECTIVE DATA UTILIZATION FOR LOW-RESOURCE SPEECH RECOGNITION
Zhikai Zhou, Wei Wang, Wangyou Zhang, Yanmin Qian, Shanghai Jiao Tong University, China
SPE-76.4: OMNI-SPARSITY DNN: FAST SPARSITY OPTIMIZATION FOR ON-DEVICE STREAMING E2E ASR VIA SUPERNET
Haichuan Yang, Yuan Shangguan, Dilin Wang, Meng Li, Pierce Chuang, Xiaohui Zhang, Ganesh Venkatesh, Ozlem Kalinli, Vikas Chandra, Facebook, United States of America
SPE-76.5: ANALYZING THE ROBUSTNESS OF UNSUPERVISED SPEECH RECOGNITION
Guan-Ting Lin, Chan-Jan Hsu, Da-Rong Liu, Hung-Yi Lee, National Taiwan University, Taiwan; Yu Tsao, Academia Sinica, Taiwan
SPE-76.6: INTERPRETING INTERMEDIATE CONVOLUTIONAL LAYERS IN UNSUPERVISED ACOUSTIC WORD CLASSIFICATION
Gašper Beguš, Alan Zhou, University of California, Berkeley, United States of America