Yayun He

Yayun He

Researcher

He Yayun is a senior algorithm engineer with in-depth research in the field of speech algorithms. In particular, he has rich practical experience in the application of voiceprint recognition technology in the financial field. He graduated from the Hong Kong Polytechnic University with a master’s degree in 2015 and joined Hong Kong Computime Co., Ltd. as a software engineer. In 2018, he joined the CITIC Bank Big Data Center as a data mining engineer. In 2021, he joined Ping An (Shenzhen) Co., Ltd. as a senior algorithm engineer.

Interests
  • Artificial Intelligence
  • Speaker Verification

Publications

  1. EMO-RL: Emotion-Rule-Based Reinforcement Learning Enhanced Audio-Language Model for Generalized Speech Emotion Recognition (2025), In EMNLP2025 (CCF-B)
  2. Generalized Audio Deepfake Detection Using Frame-level Latent Information Entropy (2025), In ICME2025 (CCF-B)
  3. Retrieval-Augmented Audio Deepfake Detection, (2024), ‡Co-first Author, In ICMR2024 (CCF-B)
  4. Efficient Multi-Model Fusion with Adversarial Complementary Representation Learning (2024), In IJCNN2024 (CCF-C)
  5. VoiceExtender: Short-utterance Text-independent Speaker Verification with Guided Diffusion Model, (2023), †First Author, In ASRU2023
  6. Feature-Rich Audio Model Inversion for Data-Free Knowledge Distillation Towards General Sound Classification, (2023), ‡Co-first Author, In ICASSP2023 (CCF-B)

中文期刊文章

  1. 基于深度卷积和自注意力机制的端到端地震波降噪方法 (2025), 《大数据》（CCF-T2）
  2. 节奏舞者：基于关键动作转换图和有条件姿态插值网络的3D舞蹈生成方法研究, (2023), †First Author, 《大数据》（CCF-T2）

Events