About
I work on speech and audio processing, especially methods for detecting manipulated or synthetic speech. My research spans speech deepfake detection, including speaker-specific analysis, time-frequency consistency, and detection in real-time communication settings. I am also interested in how detection methods can be evaluated across languages and in-the-wild conditions.
Beyond deepfake detection, I study speech editing detection and content localization, as well as audiovisual speech recognition under video-conferencing conditions. These topics connect to broader interests in digital media forensics, biometric identification and security, and music and audio processing. I am affiliated with Wuhan University. My work aims to understand the challenges posed by varied audio conditions and to develop approaches and evaluation resources for reliable speech and audio analysis.
Research interests
- Speech Deepfake DetectionI study approaches for detecting synthetic and manipulated speech across speakers, settings, and communication channels.
- Speech and Audio ProcessingI am interested in computational methods for analyzing and understanding speech and audio.
- Audiovisual Speech RecognitionI explore audiovisual speech recognition, including its behavior under video-conferencing conditions.
- Digital Media ForensicsI work on detecting and localizing manipulations in digital audio.
- Biometric Identification and SecurityI am interested in speech-related methods for identification and security.
Selected publications
2026
- International Joint Conference on Artificial Intelligence · 20263
- Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · 20262
- ACM International Conference on Multimedia (ACM MM) · 20261
- Interspeech (USB) · 20260
OpenPrints on CSPaper
No OpenPrints published yet.
Contact
No contact details.