Jui-Yang Hsu

Master Student

Speech Processing Lab



I’m a 2nd year Master student in Speech Processing & Machine Learning Laboratory at NTU EECS, supervised by Prof. Hung-Yi Lee


  • Meta/ Transfer Learning
  • Speech Processing
  • Natural Language Processing


  • MS in Computer Science, 2018 - Present

    National Taiwan University

  • BSE in Electrical Engineering, 2013 - 2018

    National Taiwan University



Master Student

National Taiwan University (NTU)

Oct 2018 – Present Taipei, Taiwan
  • Graduate Researcher in Speech Processing & Machine Learning Lab
  • Researched on low-resource speech recognition, focusing on improving the system with gradient-based meta learning and transfer learning

As the network adminstrator in Lab

  • Managed the slurm-based computation cluster (10 nodes, over 20 GPUs)
  • Migrated netdata to replace the original unstable monitor system to support real-time resource monitoring for users
  • Developed health check and notification mechanism to drain problematic node and notify to the public platform automatically

NLP Research Intern

Apple Inc.

Jun 2018 – Sep 2016 Cupertino, CA, USA
  • Researched on deep generative model to develop algorithm improving keyboard experience of users
  • The research results will be published in Apple Machine Learning Journal after feature released

Exchange Student

Kungliga Tekniska högskolan (KTH)

Aug 2017 – Jun 2018 Stockholm, Sweden
Exchanged to the department of Comuter Science & Communication (CSC)

Speech Research Intern

Delta Research Center

Jul 2016 – Aug 2016 Taipei, Taiwan
  • Researched on end-to-end speech recognition based on CTC
  • Reduced 3% CER on the corpus held by DRC, comparable to the original system
  • Migrated acoustic modeling to Tensorflow for faster development
    (building the interface between Kaldi & Tensorflow)

Undergraduate Student

National Taiwan University (NTU)

Sep 2013 – Jun 2018 Taipei, Taiwan
  • Undergraduate Researcher in Speech Processing & Machine Learning Lab
  • Proposed the hierarchical attention-based model for the TOEFL Listening Comprehension Test by machine
  • Researched on unsupervised audio embeddings
  • Teaching Assistant for Machine Learning → designed assignment