Research Interests

Deep learning

  • Developing deep learning algorithms for various signal processing, including speech and brain signal.

Speech processing

  • Text-to-speech (TTS) synthesis
    • Developing neural vocoder using domain-specific knowledge.
    • Developing TTS system based on deep learning algorithms used in natural language processing (NLP).
  • Speech enhancement using deep learning
    • Developing noise-robust speech enhancement method based on neural vocoders.
    • Developing neural networks based on domain-specific knowledge for speech enhancement to reduce model complexity.
  • Audio-video speech enhancement (AVSE)
    • Developing speech enhancement method using features extracted from video.

Brain Computer Interface

  • Analysis for brain signal
    • Developing deep learning based feature extraction algorithms for electroencephalograms (EEG) to improve performance of BCI system.
  • Motor imagery classification
    • Developing deep learning/machine learning algorithms for improving classification accuracy.
  • Sleep stage classification
    • Developing neural networks for sleep stage classification.
    • Developing neural encoder based on attention mechanism to capture features represented on EEG waves.