SLP-L12.1

HYBRID PRUNING: IN-SITU COMPRESSION OF SELF-SUPERVISED SPEECH MODELS FOR SPEAKER VERIFICATION AND ANTI-SPOOFING

Junyi PENG, Brno University of Technology, Czechia; Lin Zhang, Johns Hopkins University, United States of America; Jiangyu Han, Oldrich Plchot, Johan Rohdin, Brno University of Technology, Czechia; Themos Stafylakis, Athens University of Economics and Business, Greece; Shuai Wang, Nanjing University, China; Jan Cernocky, Brno University of Technology, Czechia
Session:
SLP-L12: Speaker Verification: Model Compression Oral
Track:
Speech and Language Processing [SL]
Location:
Room 115
Presentation Time:
Thu, 7 May, 09:00 - 09:20
Presentation
Discussion
Resources
No resources available.
Session SLP-L12
SLP-L12.1: HYBRID PRUNING: IN-SITU COMPRESSION OF SELF-SUPERVISED SPEECH MODELS FOR SPEAKER VERIFICATION AND ANTI-SPOOFING
Junyi PENG, Brno University of Technology, Czechia; Lin Zhang, Johns Hopkins University, United States of America; Jiangyu Han, Oldrich Plchot, Johan Rohdin, Brno University of Technology, Czechia; Themos Stafylakis, Athens University of Economics and Business, Greece; Shuai Wang, Nanjing University, China; Jan Cernocky, Brno University of Technology, Czechia
SLP-L12.2: DISTILLING ATTENTION KNOWLEDGE FOR SPEAKER VERIFICATION
Zezhong Jin, The Hong Kong Polytechnic University, China; Shujie Liu, Microsoft Research Asia, China; Zhe Li, The University of Hong Kong, China; Chong-Xin Gan, Zilong Huang, Man-Wai Mak, Kong Aik Lee, The Hong Kong Polytechnic University, China
SLP-L12.3: CURRICULUM LEARNING WITH CONTRASTIVE LOSS FOR LIGHTWEIGHT SPEAKER VERIFICATION
Jin Li, Man-Wai Mak, The Hong Kong Polytechnic University, Hong Kong; Johan Rohdin, Oldřich Plchot, Brno University of Technology, Czechia
SLP-L12.4: Triage knowledge distillation for speaker verification
Ju-ho Kim, Youngmoon Jung, Joon-Young Yang, Jaeyoung Roh, Chang Woo Han, Hoon-Young Cho, Samsung Research, Korea, Republic of
SLP-L12.5: ENHANCING SPEAKER VERIFICATION WITH W2V-BERT 2.0 AND KNOWLEDGE DISTILLATION GUIDED STRUCTURED PRUNING
Ze Li, Wuhan University, China; Ming Cheng, Ming Li, Duke Kunshan University, China
SLP-L12.6: Cross-Architecture Knowledge Distillation of WavLM for Lightweight Speaker Verification
Jungwoo Heo, Hyun-seo Shin, Chan-yeong Lim, Kyowon Koo, Seung-bin Kim, Jisoo Son, University of Seoul, Korea, Republic of; Kyung Wha Kim, Supreme Prosecutors' Office Republic of Korea, Korea, Republic of; Ha-Jin Yu, University of Seoul, Korea, Republic of
Contacts