CV

Education

• M.S. in signal and information processing, The Institute of Acoustics of the Chinese Academy of Sciences, 2011-2014
• B.S. in ocean acoustics, Xiamen University, 2007-2011

Work experience

• 2017.07-Now: Tencent, CSIG
  • 2024.11-now: Tencent Cloud Audio Moderation System
    • Audio Security: Audio Event Detection, Speaker Verification, Deepfake Speech
  • 2024.11-2026.03: Tencent Hunyuan / Grading Agent
    • LLM/MLLM: Exam/Essay Grading, Language Coaching, Problem-Solving
  • 2024.01-2024.11: Tencent Cloud AI Digital Human
    • Speech Generation: Zero-shot Voice Conversion, Zero-shot TTS, Speech LLM
  • 2017.08-2023.12: Tencent Internet of Vehicles / Tencent Map / Tencent Dingdang
    • Speech Recognition: Hybrid ASR, E2E ASR, VAD, Self-supervised learning, Multi-accent, Multi-dialect, Multi-lingual
    • Speaker Recognition: A joint project with Rohan Kumar Das from NUS, supervised by Haizhou Li

• 2014.07-2017.07: Qihoo360, AI Institute
  • 360 search / 360 children watch, 360 children robot, 360 camera / palm input
    • Speech recognition: LM training, AM training, Decoder development
    • Sound event detection: Detection of baby crying
    • Speech2Sing: A joint project with A*STAR, supervised by Shuicheng Yan