Xixin Wu   Assistant Professor, Department of Systems Engineering and Engineering Management, The Chinese University of Hong Kong Room 709B, William M. W. Mong Engineering Building, The Chinese University of Hong Kong, HK SAR, China Email: wuxx [at] se.cuhk.edu.hk Tel: 3943-8243

Biography

I am currently an Assistant Professor at the Department of Systems Engineering and Engineering Management, CUHK. Before joining CUHK, I worked as a Research Associate in the Speech Group of the Machine Intelligence Laboratory, Engineering Department of University of Cambridge, supervised by Prof. Mark Gales and Dr. Kate Knill. I obtain my Ph.D. degree from CUHK, supervised by Prof. Helen Meng, and my M.S. degree from Tsinghua University, supervised by Prof. Zhiyong Wu. My research interests include speech synthesis and recognition, affective computing, and neural network uncertainty.

*I am looking for self-motivated students, research assistants and postdocs with research interests in speech and language processing. Please don't hesitate to contact me if you are interested in our research directions.

Selected Publications (Google Scholar)

Journal publications

• Inferring Speaking Styles for Conversational Speech Synthesis by Learning Contextual Dependencies,
  Jingbei Li, Weihao Wu, Yi Meng, Luwen Zhang, Qiao Tian, Yuping Wang, Xixin Wu, Zhiyong Wu, Helen Meng
  IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2025
  
• Hiformer: Sequence Modeling Networks with Hierarchical Attention Mechanisms,
  Xixin Wu, Hui Lu, Kun Li, Zhiyong Wu, Xunying Liu, Helen Meng
  IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
  
• Estimating the Uncertainty in Emotion Class Labels With Utterance-Specific Dirichlet Priors,
  Wen Wu, Chao Zhang, Xixin Wu, Philip C. Woodland
  IEEE Transactions on Affective Computing, 2022
  
• Exemplar-based Emotive Speech Synthesis,
  Xixin Wu, Yuewen Cao, Hui Lu, Songxiang Liu, Shiyin Kang, Zhiyong Wu, Xunying Liu, Helen Meng
  IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, 2021
  
• Speech Emoftion Recognition Using Sequential Capsule Networks,
  Xixin Wu, Yuewen Cao, Hui Lu, Songxiang Liu, Disong Wang, Zhiyong Wu, Xunying Liu, Helen Meng
  IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, 2021
  
• Any-to-Many Voice Conversion With Location-Relative Sequence-to-Sequence Modeling,
  Songxiang Liu, Yuewen Cao, Disong Wang, Xixin Wu,Xunying Liu, Helen Meng
  IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, 2021
  
• Intonation Classification for L2 English Speech Using Multi-distribution Deep Neural Networks,
  Kun Li, Xixin Wu and Helen Meng
  Computer Speech & Language, 2016
  

Conference publications

• UniSRM: A Unified Speech Reward Model for Reasoning-Based Fine-grained Assessment,
  Yuanyuan Wang, Dongchao Yang, Yayue Deng, Zhiyong Wu, Yiwen Guo, Helen Meng, Xixin Wu
  in Proc. ACL'26
  
• Gogo: Group-wise Granularity-ordered Codec for Stable and Efficient Speech Generation,
  Weidong Chen, Helen Meng, Xixin Wu
  in Proc. ICLR'26
  
• DualSpeechLM: Towards Unified Speech Understanding and Generation via Dual Speech Token Modeling with Large Language Models,
  Yuanyuan Wang, Dongchao Yang, Yiwen Shao, Hangting Chen, Jiankun Zhao, Zhiyong Wu, Helen Meng, Xixin Wu
  in Proc. AAAI'26
  
• DrawSpeech: Expressive Speech Synthesis Using Prosodic Sketches as Control Conditions,
  Weidong Chen, Shan Yang, Guangzhi Li, Xixin Wu
  in Proc. ICASSP'25
  
• AudioComposer: Towards Fine-grained Audio Generation with Natural Language Descriptions,
  Yuanyuan Wang, Hangting Chen, Dongchao Yang, Zhiyong Wu, Xixin Wu
  in Proc. ICASSP'25
  
• Defending Unauthorized Voice Cloning with Watermark-Aware Codecs,
  Jiankun Zhao, Lingwei Meng, Chengxi Deng, Helen Meng, Xixin Wu
  in Proc. INTERSPEECH'25
  
• UniSep: Universal Target Audio Separation with Language Models at Scale,
  Yuanyuan Wang, Hangting Chen, Dongchao Yang, Weiqin Li, Dan Luo, Guangzhi Li, Shan Yang, Zhiyong Wu, Helen Meng, Xixin Wu
  in Proc. ICME'25
  
• UniAudio 1.5: Large Language Model-Driven Audio Codec is A Few-Shot Audio Task Learner,
  Dongchao Yang, Haohan Guo, Yuanyuan Wang, Rongjie Huang, Xiang Li, Xu Tan, Xixin Wu, Helen Meng
  in Proc. NeurIPS'24
  
• UniAudio: Towards Universal Audio Generation with Large Language Models,
  Dongchao Yang, Jinchuan Tian, Xu Tan, Rongjie Huang, Songxiang Liu, Haohan Guo, Xuankai Chang, Jiatong Shi, Sheng Zhao, Jiang Bian, Zhou Zhao, Xixin Wu, Helen Meng
  in Proc. ICML'24
  
• A Hierarchical Regression Chain Framework for Affective Vocal Burst Recognition,
  Jinchao Li, Xixin Wu*, Kaitao Song, Dongsheng Li, Xunying Liu, Helen Meng
  in Proc. ICASSP'23, (ACII 2022 Affective Vocal Burst (AV-B) Recognition competition 1st place, *corresponding author)
  
• Inferring Speaking Styles from Multi-modal Conversational Context by Multi-scale Relational Graph Convolutional Networks,
  Jingbei Li, Yi Meng, Xixin Wu*, Zhiyong Wu*, Jia Jia, Helen Meng, Qiao Tian, Yuping Wang, Yuxuan Wang
  in Proc. ACM MM'22, (*corresponding author)
  
• Neural Architecture Search for Speech Emotion Recognition,
  Xixin Wu, Shoukang Hu, Zhiyong Wu, Xunying Liu, Helen Meng
  in Proc. ICASSP'22
  
• Ensemble Approaches for Uncertainty in Spoken Language Assessment,
  Xixin Wu, Kate M. Knill, Mark J.F. Gales, Andrey Malinin
  in Proc. Interspeech'20
  
• Speech Emotion Recognition Using Capsule Networks,
  Xixin Wu, Songxiang Liu, Yuewen Cao, Xu Li, Jianwei Yu, Dongyang Dai, Xi Ma, Shoukang Hu, Zhiyong Wu, Xunying Liu, Helen Meng
  in Proc. ICASSP'19
  
• Rapid Style Adaptation Using Residual Error Embedding for Expressive Speech Synthesis, [demo]
  Xixin Wu, Yuewen Cao, Mu Wang, Songxiang Liu, Shiyin Kang, Zhiyong Wu, Xunying Liu, Dan Su, Dong Yu and Helen Meng
  in Proc. Interspeech'18
  
• Feature Based Adaptation for Speaking Style Synthesis,
  Xixin Wu, Lifa Sun, Shiyin Kang, Songxiang Liu, Zhiyong Wu, Xunying Liu, Helen Meng
  in Proc. ICASSP'18
  

Activities & Service

• Publicity chair, ISCSLP 2022, SLT 2024
• Tutorial chair, ISCSLP 2024
• Area chair, COLING 2022, 2024
• Tutorial (together with Prof. Longbiao Wang), Odessey 2022
• Invited talk, The SpeechHome Conference on Speech Technology 2021
• Tutorial, Seoul International Conference on Speech Sciences (SICSS), Seoul, Korea 2019

Honors & Awards

• First place in two tasks of ACII 2022 Affective Vocal Bursts (AV-B) Recognition Competition
• First place in ACL 2022 Doc2Dial Shared Task
• Second place in ICASSP 2022 Multi-party Multi-channel Meeting Transcription Challenge (M2Met)
• Best paper award, IEEE Robio 2022
• Champion, HKSTP SciTech Challenge 2021
• First place in INTERSPEECH 2020 Challenge: Automatic Speech Recognition for Non-Native Children's Speech

Patents

• Grounded Dialogue Generation with Cross-encoding Re-ranker, Grounding Span Prediction and Passage Dropout,
  Helen Meng, Xixin Wu, Kun Li, Tianhua Zhang, Liping Tang, Junan Li, Hongyuan Lu,
  U.S. Provisional Patent No. 17900806, filed on 31 August, 2022
  
• Voice synthesis method, model training method and device, and computer equipment,
  Xixin Wu, Mu Wang, Shiyin Kang, Dan Su, Dong Yu,
  Pub. No. US 2020/0380949 A1, pub. date 3 December, 2020
  
• Acoustic-Graphemic Model and Acoustic- Graphemic-Phonemic Model for Computer-Aid Pronunciation Training and Speech Processing,
  Kun Li, Lifa Sun, Xixin Wu and Helen Meng,
  U.S. Provisional Patent No. 62/413,939, filed on 31 October 2016