|
Chia-ying Jackie Lee
email: chiaying at csail.mit.edu
|
About me
I focus on developing machine learning, AI driven systems.
- I co-founded Silo for the mission of making the agricultural supply chain transparent, efficient, and sustainable.
During my time at Silo, I hired, built, and led the entire engineering team to develop the core product from scratch. In addition to leading the technical part of the company, I also participated heavily in product and bussiness decisions. I left the company after it successfully raised series A from Andreessen Horowitz.
- Previously, I worked at Virta
Health as the first data scientist. I applied deep learning and
other machine learning methods to improve healthcare outcome and
efficiency. Virta is a deeply mission-driven company, and that had a
profound impact on how I hope to build Silo.
- I was also part of Kite. I built
several core fuctions of the artificial programming copilot including program
language detection, related code examples, diff suggestions (among others). Kite has a
stellar engineering team.
Previous Life
I've successfully defended my Ph.D. thesis.
I was a graduate student working with Dr. James Glass at the Computer Science and Artificial Intelligence Lab (CSAIL) of MIT. I applied machine learning techniques to let computers discover linguistic knowledge of a language from a large amount of speech data.
I've spent years studying machine learning, statistical inference, speech recognition, natural language processing, and computational cognitive science. I enjoy building probabilistic models to gain useful knowledge from a massive amount of data.
Besides building probabilistic models and deriving inference algorithms, I also have hands-on experiences in deep learning (I implemented a library in C++ for constructing various neural nets, with a focus on recurrent neural nets, during one of my summer internships before any of the current deep learning libraries became available). I have also worked on HMM-based speech synthesis and statistical machine translation.
Last but not least, I genuinely enjoy the process of looking for problems, brainstorming for innovative solutions, and implementing them.
Resume/CV
Here's my CV. [pdf]
Press
Google Schoar
Thesis
Publications
-
Improving Hemoglobin A1c Prediction by Leveraging Participant Similarity
Chia-ying Lee, Amy McKenzie, David Bill
Workshop on Machine Learning for Health, Neural Information Processing Systems (NIPS), 2016.
-
Unsupervised Lexicon Discovery from Acoustic Input
Chia-ying Lee, Timothy O'Donnell, James Glass
Transactions of Association for Computational Linguistics (TACL), 2015.
[bibtex][pdf][Coded in C++][Github]
-
One-shot Learning of Generative Speech Concepts
Brenden Lake*, Chia-ying Lee*, James Glass, Joshua Tenenbaum
Cognitive Science Soceity (CogSci), 2014.
[bibtex][pdf][slides][Coded in C++][Github]
-
Joint Learning of Phonetic Units and Word Pronunciations for ASR
Chia-ying Lee, Yu Zhang and James Glass.
Empirical Methods in Natural Language Processing (EMNLP), 2013.
[bibtex][pdf][slides][Coded in C++][Github]
-
A Nonparametric Bayesian Approach to Acoustic Model Discovery (Highlighted in SLTC Newsletter, August 2012)
Chia-ying Lee and James Glass.
Association for Computational Linguistics (ACL), 2012.
[bibtex][pdf][slides][Coded in C++][Github]
-
A Transcription Task for Crowdsourcing with Automatic Quality Control
Chia-ying Lee and James Glass.
International Speech Communication Association (Interspeech), 2011.
[bibtex][pdf][coded in JavaScript]
-
An Efferent-Inspired Auditory Model Front-End for Speech Recognition
Chia-ying Lee, James Glass and Oded Ghitza.
International Speech Communication Association (Interspeech), 2011.
[bibtex][pdf][slides][coded in Matlab]
-
Collecting Voices from the Cloud
Ian McGraw, Chia-ying Lee, Lee Hetherington, Stephanie Seneff, Jim Glass.
The Language Resources and Evaluation Conference (LREC), 2010.
[bibtex][pdf][coded in JavaScript]
-
Closed-loop Auditory-based Representation for Robust Speech Recognition
Chia-ying Lee.
Master Thesis, CSAIL MIT, May 2010.
[bibtex][pdf][coded in Matlab]
-
Empirical Mode Decomposition Descriptor for Plane Closed Curves
Soo-Chang Pei, Yu-Zhe Hsiao and Chia-Ying Lee.
IEEE International Conference on Multimedia and Expo (ICME), 2009.
[bibtex][pdf][coded in Matlab]
Invited Talks
- A Nonparametric Bayesian Approach to Acoustic Model Discovery
Johns Hopkins University, April 2012.
- A Nonparametric Bayesian Approach to Acoustic Model Discovery
SLS Seminar, CSAIL MIT, May 2012.
- Zero-Resource Speech Pattern and Sub-Word Unit Discovery (Highlighted in SLTC Newsletter, November 2012)
The Speech and Audio in the Northeast (SANE) Conference, October 2012.
- Bayesian Approaches to Acoustic Model and Pronunciation Lexicon Discovery
NL Seminar, Information Science Institute (ISI), University of Southern California (USC), July 2013.
- Unsupervised Bayesian Approach to Training Automatic Speech Recognizers
Natural Language Processing Group, CSAIL, MIT, October 2013.
- Discovering Linguistic Structures from Speech: Models and Applications
International Computer Science Institute, University of California, Berkeley, November 2015.
- Discovering Linguistic Structures from Speech: Models and Applications
SRI International, January 2016.
- Machine Learning in Healthcare
Virta, August 2017.
Professional Services
- Program Committee, EACL 2021
- Program Committee, NAACL 2020
- Program Committee, NAACL 2020
- Program Committee, EMNLP 2018
- Program Committee, Association of Computational Linguistics (ACL) 2018
- Area Chair, Association of Computational Linguistics (ACL) 2017
- Reviewer, IEEE Transactions on Audio, Speech, and Language Processing
- Reviewer, IEEE Signal Processing Letters
- Reviewer, Journal of Neural Computation
- Reviewer, International Journal of Pattern Recognition and Artificial Intelligence
Teaching
Beyond research
I run, I dance and I take photos.
I participated in quite a few student organizations at MIT. To name just a few, I was a member and the president of the CSAIL Student Committee, the co-president of Graduate Women in Course 6 (GW6), and the chief organizer of CSAIL Student Workshop (CSW) 2009.