[May 15, 2025]  🔊 vTAD Challenge 2025 is now open!
Explore how humans perceive voice timbre through comparative attribute detection. Join us in building explainable and perceptually grounded spoken language systems.
📅 Challenge runs from May to October 2025, with a special session at NCMMSC 2025 in Zhenjiang, China.
👉 Learn more and register: https://vtad2025-challenge.github.io/
[Nov 1, 2023] Â Welcome to Join Dr. Lee's Research Team @ PolyU, Hong Kong
I'm seeking highly motivated Ph.D. candidates to join my research team and engage in advanced studies within the fields of paralinguistic AI, speech information processing, speech deepfake detection, and voice privacy and security. If you're an exceptional student interested in these areas, please feel free to reach out for further discussion via email at kong-aik.lee@polyu.edu.hk.
[Sep 22, 2023] Our Paper on Speech Disentanglement Learning Accepted at NeurIPS 2023! The paper, titled "Disentangling Voice and Content with Self-Supervision for Speaker Recognition," was a collaborative effort with Tianchi Liu, Qiongqiong Wang, and Haihzou Li.Â
[July 1, 2022] ASVspoof5 is now calling for spoofed speech data contributors. Check it out here.Â
[Aug 31, 2021] PyTorch implementation of Serialized Multi-Layer Multi-Head Attention for Neural Speaker Embedding is now available. Thanks, Hongning Zhu for making this possible. Click here to access.Â
[April, 2021] We have just launched the ASVSpoof 2021, the fourth edition of the ASVspoof initiative. Click here for more details.
[May 19, 2020] CSL Special Issue on "Two Decades of Speaker Recognition Evaluation - Are We There Yet?" is now available. Thanks to Omid Sadjadi, Haizhou Li, and Douglass Reynolds who are my co-editors, and authors who have contributed to the special issue.Â
https://www.sciencedirect.com/journal/computer-speech-and-language/special-issue/10V76T5RRL2Â