Keng-hao Chang

About me

I am a final year Ph.D. student in Computer Science Division, EECS Department, at University of California, Berkeley. I'm also a member of Berkeley Institute of Design. My advisor is Prof. John Canny. I'm also an alumnus of Users Group at IBM Research Almaden, Everyday Sensing and Perception project group at Intel Research Seattle, and Applied Machine Learning and Statistics Group at Microsoft Research Redmond. My research interests span Human Computer Interaction and Ubiquitous Computing, focusing on healthcare IT.

My thesis project is to build mental health monitor via the human voice. Human voice has a wealth of affective information, as well as indicators of early stage mental illness (a.k.a. prodrome, psychomotor dysfunction). Coupled with the pervasiveness and the built-in microphones of mobile phones, the human voice becomes the most penetrating and unobtrusive means to monitor mental health. We started this work by learning psychiatrists' diagnostic knowledge and planned to apply the knowledge to build an effective mental health monitor. I have been working on projects using voice to detect emotion and abnormal speaking styles, and to differentiate depressed patients from healthy subjects. We are in the process of building a machine learning model that predicts the mental health level (via regression) from the voice signals. At the same time, I'm implementing a mental health monitor library that can run efficiently on mobile phones, turning mobile phones as "cognitive phones". As a side project, I'm also working on biogsignals to detect psychological stress.

I am originally from Taiwan. I gradated at 2006 from Deopartment of Computer Science and Information Engineering at National Taiwan University in Taipei, Taiwan. I worked with Ubicomp Lab and my adviosr was Prof. Hao Chu. In my master program, I was focusing on user-centric and intelligent artifacts which bridge physical to digital. The major project is dietary-aware dining table, which monitors nutrition and calories of food intake by tracking the food transfer interactions into personal containers during courses of a meal. This project is envisioned to give just-in-time persuasive feedbacks, changing our behaviors to eat healthier.



Keng-hao Chang, Ph.D. student
Department of Electrical Engineering and Computer Sciences, and
Berkeley Institute of Design
University of California, Berkeley

Email: kenghao <@> eecs <dot> berkeley <dot> edu

Phone: +1 (510) 642-1268
Fax: +1 (510) 642-1450

Mailing address:
Keng-hao Chang
266 Sutardja Hall, MC 1764
Attn: Keng-hao Chang, Rm 360 HMMB
Berkeley, CA 94720-1764

Current projects

Mental health monitor through voice analysis

Using biosignals to detect psychological stress

AMMON: Affective and Mental health MONitor on mobile phones







11/1/2011: Our paper "AMMON: A Speech Analysis Library for Analyzing Affect, Stress, and Mental Health on Mobile Phones" is awarded as the best paper in PhoneSense 2011.

5/27/2011: Our paper "Improved Classification of Speaking Styles for Mental Health Monitoring using Phoneme Dynamics" is accepted to Interspeech 2011. This work uses information in phoneme sequences to discriminate abnormal speaking styles, which is based on our knowledge transfer work described in the AnalyisThis paper.

5/5/2011: I'm heading to Vancouver, BC to present my work as a poster, "AnalyzeThis: Unobtrusive Mental Health Monitoring by Voice". It's a work of knowledge transfer from psychiatrists so that we can use the domain knowledge to build a mental health monitor.

5/19/2010: Our paper "A Practical Examination of Multimodal Feedback and Guidance Signals for Mobile Touchscreen Keyboards" is accepted by MobileHCI 2010, Industrial Case Studies. This paper is about my internship work at Microsoft Research in Summer 2009.

4/15/2010: My thesis project is awarded by 2010 CITRIS Seed Funding Awards.

11/16/2009: I passed my qualifying exam (thesis proposal) and my thesis topic is: M-Health, monitoring menthal health through voice analysis.

5/28/2009: I'll be a summer intern in Microsoft Research from June to August in Redmond, WA.

2/15/2009: Our paper "Inferring Identity using Accelerometers in Television Remote Controls" is accepted by Pervasive 2009, which will be hold in Nara, Japan. The paper is about my internship work at Intel Research Seattle in summer 2008.

11/27/2007: There are followup media coverage for the project about tracking free weight exercises: in India Times: "Technology to monitor progress in gym" and in The Daily Californian on Nov 27 (a UC Berkeley student newspaper): "For weight lifters: a glove for fitness".

11/16/2007: The project about tracking free weight exercises is now reported by Discovery News: "A Personal Trainer, in the Palm of Your Hand".

09/26/2007: Our paper "tracking free weight exercises" was presented on Ubicomp 2007, at Innsbruck Vienna.

Updated: Nov 1, 2011 . (C) 2005-2011 | Keng-hao Chang