Ching-Hua Lee

Samsung Research AI Center - Mountain View

profile_pic.jpg

My name is Ching-Hua Lee and I work on a number of subjects in machine learning and signal processing.

My current research focuses on efficient deep learning-based solutions for speech and audio applications, including Single/Multi-Channel Speech Enhancement, Voice Commands, Speech Recognition/Translation and Speech-LLMs. During my Ph.D. at UC San Diego, I worked on adaptive filtering and sparse signal processing with applications to speech and audio.

News

Mar 01, 2025 I am now a Staff Machine Learning Research Engineer at Samsung AI Center - Mountain View.
Feb 01, 2021 I joined Samsung Research America AI Center as a Senior Machine Learning Researcher.
Dec 04, 2020 I defended my dissertation, titled A Family of Sparsity-Promoting Gradient Descent Algorithms Based on Sparse Signal Recovery, on Dec 4, 2020.
Mar 06, 2020 I received the NSF Student Travel Grants for ICASSP 2020.
Jun 25, 2018 I started my internship at Qualcomm Inc.

Selected publications

  1. ICML
    restoregrad.PNG
    RestoreGrad: Signal restoration using conditional denoising diffusion models with jointly learned prior
    Ching-Hua Lee, Chouchang Yang, Jaejin Cho, and 4 more authors
    In International Conference on Machine Learning (ICML), to appear, 2025
  2. ICASSP
    afnet.PNG
    Better exploiting spatial separability in multichannel speech enhancement with an align-and-filter network
    Ching-Hua Lee, Chouchang Yang, Yashas Malur Saidutta, and 3 more authors
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025
  3. NeurIPS
    cifd.PNG
    CIFD: Controlled information flow to enhance knowledge distillation
    Yashas Malur Saidutta, Rakshith Sharma Srinivasa, Jaejin Cho, and 4 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2024
  4. NeurIPS
    cwcl.PNG
    CWCL: Cross-modal transfer with continuously weighted contrastive loss
    Rakshith Sharma Srinivasa, Jaejin Cho, Chouchang Yang, and 4 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2023
  5. Interspeech
    kws_se_spp.PNG
    Robust keyword spotting for noisy environments by leveraging speech enhancement and speech presence probability
    Chouchang Yang, Yashas Malur Saidutta, Rakshith Sharma Srinivasa, and 3 more authors
    In Annual Conference of the International Speech Communication Association (Interspeech), 2023
  6. NeurIPS
    resnest.PNG
    ResNEsts and DenseNEsts: Block-based DNN models with improved representation guarantees
    Kuan-Lin Chen, Ching-Hua Lee, Harinath Garudadri, and 1 more author
    In Advances in Neural Information Processing Systems (NeurIPS), 2021
  7. IEEE/ACM TASLP
    slms_snlms.PNG
    Proportionate adaptive filtering algorithms derived using an iterative reweighting framework
    Ching-Hua Lee, Bhaskar D. Rao, and Harinath Garudadri
    IEEE/ACM Transactions on Audio, Speech, and Language Processing (IEEE/ACM TASLP), 2020
  8. IEEE SPL
    scg.PNG
    A sparse conjugate gradient adaptive filter
    Ching-Hua Lee, Bhaskar D. Rao, and Harinath Garudadri
    IEEE Signal Processing Letters (IEEE SPL), 2020