Master’s student in Computer Science

Johns Hopkins University


I am Yi-Te Hsu, an M.S. Computer Science student at Johns Hopkins University. Currently, I am working on speech emotion conversion and detection with Prof. Archana Venkataraman. I was a machine learning engineer intern at Apple Inc. this summer, working on the machine translation projects.

Before coming to JHU, I conducted research on speech and signal processing with Dr. Yu Tsao at Academia Sinica. I mainly focused on model compression and acceleration for deep neural networks. I also collaborated with Prof. Frank Rudzicz at University of Toronto (UofT) and Vector Institute as a visiting researcher. I worked on healthcare projects for detecting pathological voice and identifying Alzheimer’s disease.

My interests include speech processing, natural language processing, AI, and ML. I am excited about applying these techniques to solve real-world problems.

I am looking for 2021 full-time opportunities!


  • Speech and Voice technology
  • Natural Language Processing
  • Bio-acoustic Signal Processing
  • Healthcare
  • Machine Learning


  • M.S. in Computer Science, 2019-

    Johns Hopkins University

  • Visiting Student Researcher, 2018

    University of Toronto

  • BSc in Electrical Engineering, 2017

    National Taiwan University

Recent Publications

Efficient Inference For Neural Machine Translation. accepted to SustaiNLP Workshop at EMNLP 2020, 2020.


IA-NET: Acceleration and Compression of Speech Enhancement using Integer-adder Deep Neural Network. accepted to Annual Conference of the International Speech Communication Association (INTERSPEECH 2019), 2019.


Detecting dementia in Mandarin Chinese using transfer learning from a parallel corpus. accepted to Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2019), 2018.


Robustness against the channel effect in pathological voice detection. accepted to Machine Learning for Health Workshop at NIPS 2018, 2018.


A study on speech enhancement using exponent-only floating point quantized neural network (EOFP-QNN). accepted to IEEE Spoken Language Technology conference (IEEE SLT), 2018.




Machine Learning Engineer Intern

Apple Inc.

Jun 2020 – Aug 2020 California
  • Investigated the state-of-the-art model efficiency techniques for deep neural networks.
  • Proposed a machine translation architecture for faster inference. Achieved up to 109% speedup and reduced the number of parameters by 25% while maintaining the same translation quality in terms of BLEU.

Research Intern

Vector Institute; University of Toronto

Sep 2018 – Dec 2018 Toronto
  • Healthcare Project – Robust pathological voice detection system: Used bidirectional LSTM to develop an early detection system. Combined with an unsupervised domain adaptation method to solve the channel mismatch of different devices.
  • NLP Project – Detection of Alzheimer’s disease: Proposed a method to transfer Mandarin features to English ones with the corpus of a picture description task. Combined algorithms from different languages to achieves multi-language application.

Research Assistant

Academia Sinica

Feb 2018 – Jul 2019 Taipei
  • Quantization on deep neural network: Proposed a novel exponent-only floating-point quantized neural network (EOFP-QNN) to quantize the model. Achieved a 4x compression rate.
  • Acceleration on a compressed DNN: Proposed IA-Net, which can simultaneously compress the model size and accelerate the inference process by replacing the floating-point multiplier with an integer adder without performance degradation.
  • Bio-signal processing: Cooperated with the otolaryngologist and speech therapist at Far Eastern Memorial Hospital in Taiwan. Combined deep learning and machine learning techniques to develop the disease classification system by the patients’ voice. Taught electronic engineering and researched semiconductor physics.

Data Scientist Intern


Jun 2016 – Feb 2017 Taipei
  • Applied ML techniques and statistic model to extract core information from different types of IoT data. Applied unsupervised machine learning methods to detect anomalies with unlabeled data
  • Predicted the office space occupancy rate for 85% accuracy with the time-series data for 6 months from real-time sensors.
  • Product deployment: Deploy the machine learning models to real products. Cooperate with collaborates efficiently. Gained solid experience in machine learning and engineering.


COVID-19 News Website

A COVID-19 news website containing the latest and the most important local information.


Analyze the data and provide insight for the business.

Facebook Likes Estimator

Facebook Likes Estimator for Major News Publishers’ Pages.

how social media influence human emotion

An analysis of how social media influence human emotion.


An intelligent movie recommendation system.

Speaker Identification

Using FFT and signal processing techniques to identify speakers.