En
EN
人才队伍
首页 >> 人才队伍 >> 科研人员

陈婧

  • 研究员、博士生导师
  • 北京大学智能学院
    北京大学国家生物医学成像科学中心（兼）
  • 个人简介
    主要从事三个科研方向的研究：1）听觉注意解码：人脑在复杂的听觉场景中受选择性注意的调控，从而实现对目标声源的增强和对干扰声源的抑制。然而，现有助听设备无法对佩戴者感兴趣的听觉注意客体进行有效识别。听觉注意解码，即通过大脑的神经反应(EEG、MEG、ECoG等)解码出听者注意的客体。目前主流方法是从神经信号中，使用线性或非线性（神经网络）方法重构各种听觉线索（如包络、onset、基频等等），亦有研究使用端到端的神经网络直接解码注意客体。2）语义解码：语义概念是人类思维过程和理解世界的基础，研究人员发现特定的语义概念有着特定的神经活动模式。基于此，人们开始探索语义概念解码，即从神经信号中识别出受试者正在思考或关注的语义概念。在该问题下，研究人员以音频、图像、文本等为刺激，引导受试者思考或关注特定语义概念，并记录该过程下受试者的神经信号（fMRI，MEG，EEG等）。在解码过程中，利用信号处理、机器学习等多种手段从神经信号中提取特征，构建从神经信号到语义概念的解码器，实现对语义概念的解码。3）智能助听：听力损失是重要的公共健康问题之一，如果得不到有效干预，那么将会造成造成患者的沟通障碍，影响其日常生活与交流。给予恰当的听力康复服务能够使得绝大部分听力损失者的听力得到补偿，从而改善其言语能力并提高生活质量。然而，传统的听力补偿技术无法为患者提供个性化的听力补偿且验配过程耗时耗力。因此，我们旨在将人工智能应用技术与听力学进行结合，对患者的听力状况进行精细化的建模，并为其提供个性化的听力补偿与自适应验配服务。
  • 所授课程
    《脑与认知科学》 （必修，2学分）
    《人工智能引论》（必修，4学分）
  • 获奖及荣誉
    "Newton International Fellowship" by Royal Society, UK (英国皇家学会"牛顿国际学者奖" , 2009）
  • 个人履历
    2020.8至今 北京大学，智能学院，研究员
    2013.02-2020.7 北京大学，信息科学技术学院，“百人计划”研究员
    2009.05-2012.09 英国剑桥大学，实验心理学系，博士后 (合作导师：Brian C. J. Moore 教授)
    2002.09-2009.01 北京大学，信息科学技术学院，博士 (信号与信息处理，导师：迟惠生 教授)
    1997.09-2001.07 哈尔滨工程大学，电子工程系，学士 (通信工程)
  • 承担项目
    1. 科技部2030重大项目（2021ZD0201500），"类脑听觉前端模型与系统研究"，课题负责人（在研）
    2. 自然科学基金项目，“基于听觉注意检测的助听方法研究”，项目负责人（在研）
    3. 索诺瓦控股集团（SONOVA AG，瑞士），“中国听觉及人工智能”，项目负责人（在研）
    4. 北京大学医信交叉项目，“老年性聋的临床分型与干预策略研究”, 合作负责人（已结题）
    5. 国家自然科学基金面上项目，“基于言语可懂度理论的声电双模态刺激方法及模式研究”，项目负责人（已结题）
    6. 索诺瓦(SONOVA)听力技术（上海）有限公司，“基于深度神经网络的语谱变化增强算法”，项目负责人（已结题）
    7. 索诺瓦(SONOVA)听力技术（上海）有限公司，“语音频谱变化的增强对汉语语音可懂度的影响”，项目负责人（已结题）
    8. 国家自然科学基金面上项目，“听力损伤评价方法及计算模型”， 项目负责人（已结题）
    9. 北京大学985项目建设课题(百人计划)，“听觉抗噪声机理及听力康复技术的研究”， 项目负责人（已结题）
    10. Follow-on Funding for Newton Alumni, The Royal Academy of Engineering UK, Principal Investigator (已结题)
  • 代表性论文及论著
    1. Li, X (PhD student), Sun, Y., Wu, X., Chen, J.* (2022). Multi-Speaker Pitch Tracking via Embodied Self-Supervised Learning, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, pp. 8257–8261. Singapore City, Singapore.
    2. Fu, Z. (PhD student), Wang, B., Chen, F., Wu, X., Chen, J.* (2021). Eye-gaze Estimation with HEOG and Neck EMG using Deep Neural Networks, 29th European Signal Processing Conference, EUSIPCO 2021, pp. 1261-1265, Dublin, Ireland.
    3. Fu, Z. (PhD student), Wang, B., Wu, X., Chen, J.* (2021). Auditory Attention Decoding from EEG using Convolutional Recurrent Neural Network, 29th European Signal Processing Conference, EUSIPCO 2021, pp. 970-974, Dublin, Ireland.
    4. Niu Y (PHD Student), Liu Y, Wu X, Chen, J.* (2021). Categorical perception of lexical tones based on acoustic-electric stimulation. JASA Express Letters, 1(8):084405.
    5. Fu, Z. (PhD student), Chen, J.* (2020). Congruent Audiovisual Speech Enhances Cortical Envelope Tracking during Auditory Selective Attention, 21th Annual Conference of the International Speech Communication Association, INTERSPEECH 2020, pp. 116–120, Shanghai, China.
    6. Li, X. (PhD student), Liu, R., Song, T., Wu, X., Chen, J. * (2020). Single-Channel Speech Separation Integrating Pitch Information Based on a Multi Task Learning Framework, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020, pp. 7279-7283, Barcelona, Spain.
    7. Fu, Z. (PhD student), Wu, X., Chen, J.*(2019). Congruent audiovisual speech enhances auditory attention decoding with EEG, Journal of Neural Engineering, 16(6),066033.
    8. Du, Y. (Master student), Shen, Y., Wu, X., Chen, J. *(2019). The effect of speech material on the band importance function for Mandarin Chinese. Journal of the Acoustical Society of America,146(1):445-457.
    9. Fu, Z(PhD student), Yang, H., Chen, F., Wu, X., Chen, J. *(2019) Brainstem encoding of frequency-modulated sweeps is relevant to Mandarin concurrent-vowels identification for normal-hearing and hearing-impaired listeners. Hearing Research, 380:123-136.
    10. Niu, Y. (PHD Student), Chen, F., Chen, J. *(2019) The effect of F0 contour on the intelligibility of Mandarin Chinese for hearing impaired listeners. Journal of the Acoustical Society of America, 146(2):EL85-EL91.
    11. Fu, Z.(PhD student), Wu, X., Chen, J.*(2019) Contribution of spectral and tonal cues to Mandarin concurrent-vowels identification for normal-hearing and hearing-impaired Listeners, in Proceedings of Interspeech 2019, pp. 3083-3087, Graz, Austria, Sep 15-19, 2019.
    12. Li, X. (PhD Student), Wu, X., Chen, J.* (2019) Integrating Spectro-temporal Context into Features Based on Auditory Perception for Classification-based Speech Separation, in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7165-7169, Brighton, United Kingdom, May 12-17, 2019.
    13. Li, X. (PhD Student), Wu, X., Chen, J.* (2019) A Spectral-change-aware Loss Function for DNN-based Speech Separation, in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6870-6874, Brighton, United Kingdom, May 12-17, 2019.
    14. Chen, J.*, Yang, H.Y., Wu, X.H., & Moore, B.C.J. (2018). The effect of F0 contour on the intelligibility of speech in the presence of interfering sounds for Mandarin Chinese. The Journal of the Acoustical Society of America, 143(2), 864-877.
    15. Chen, J.*, Moore, B.C.J., Baer, T., & Wu, X.H., (2018). Individually tailored spectral-change enhancement for the hearing impaired. The Journal of the Acoustical Society of America, 143(2), 1128-1137.
    16. Du, Y.F. (Master student), Shen, Y., Yang, H.Y., Wu, X.H., Chen, J.* (2018). Measuring the band importance function for Mandarin Chinese with a bayesian adaptive procedure. In Proceedings of Interspeech, pp. 961-965, Hyderabad, India, September 2-6, 2018.
    17. Song, M.J. (Master student), Chen, F., Wu, X.H., Chen, J.* (2018). A time-weighted method for predicting the intelligibility of speech in the presence of interfering sounds. In Proceedings of International Conference of Acoustics, Speech and Signal Processing (ICASSP), pp. 5589-5593, Alberta, Canada, April 15-20, 2018.
    18. Chen, J., Huang, Q., & Wu, X.* (2016). Frequency importance function of the speech intelligibility index for Mandarin Chinese. Speech Communication, 83, 94-103.
    19. He, W., Ding, X., Zhang, R., Chen, J.*, Zhang, D., & Wu, X.* (2014). Electrically-Evoked Frequency-Following Response (EFFR) in the Auditory Brainstem of Guinea Pigs. PloS one, 9(9), e106719, 1-10.
    20. Chen, J.*, Baer, T., & Moore, B. C. (2013). Effect of spectral change enhancement for the hearing impaired using parameter values selected with a genetic algorithm. The Journal of the Acoustical Society of America, 133(5), 2910-2920.
    21. Chen, J., Moore, B.C.J.* (2013) “Effect of individually tailored spectral change enhancement on speech intelligibility and quality for hearing-impaired listeners”, in Proceedings of ICASSP 2013, pp. 8643-8647, Vancouver, Canada, May 2013.
    22. Chen, J.*, Baer, T., and Moore, B. C. J. (2012) Effect of enhancement of spectral changes on speech intelligibility and clarity preferences for the hearing impaired. Journal of Acoustic Society of America, 131(4), 2987-2998.
    23. Chen, J., Li, H.H., Li, L., Moore, B.C.J., Wu, X-H.* (2012), “Informational masking of speech produced by speech-like sounds without linguistic content,” Journal of Acoustic Society of America, 131(4), 2914-2926.
    24. Chen, J.*, Baer, T., and Moore, B. C. J. (2010), “Effects of enhancement of spectral changes on speech quality and subjective speech intelligibility,” in Proceedings of Interspeech 2010, pp. 1640-1643, Tokyo, Japan, October 2010.
    25. Wu, X.H.*, Chen, J.*, Yang, Z.G., Huang, Q., Wang, M.Y., Li, L. (2007), “Effect of number of masking talkers on speech-on-speech masking in Chinese,” in Proceedings of Interspeech 2007, pp. 390-393, Antwerp, Belgium, September 2007.