Edison Thomaz

Publications Teaching Code + Data Vita Prospective Students

Recent News

Very happy and thankful that my promotion to Associate Professor has been approved by UT Austin and is now official, starting in Fall 2025.
Aug 2025

Dawei completed his doctorate by defending his PhD thesis titled "Towards Practical Acoustic Sensing for Human Activity Recognition on Commercial Wearable Devices". His work was featured at IMWUT, ISWC, MobileHCI and other top venues.
Mar 2024

We have been exploring how LLMs might have an impact in activity recognition. Sloke has conducted studies centered on how they might change data annotation. For more details, check the paper, which will be presented at the CHASE 2024 conference. The paper is titled "Leveraging Large Language Models to Annotate Activities of Daily Living Captured with Egocentric Vision".
Feb 2024

Ted Walls, Kathleen Melanson (both at the University of Rhode Island) and I have been granted funding from NIH to explore new techniques for dietary monitoring using a novel wearable device. We thank NIH for their support, and are looking forward to advancing this research.
Jan 2024

After many years as Associate Editor of IMWUT, I have been invited to join the board as Editor. Additionally, I was elected chair of the UbiComp Steering Committee. Very happy to continue contributing to my academic community in these new roles.
Nov 2023

The paper titled "Automated Face-To-Face Conversation Detection on a Commodity Smartwatch with Acoustic Sensing" was accepted to IMWUT, and was presented at Ubicomp 2023 this month. Dawei led the work with key contributions by Alice.
Oct 2023

Gautham Gudur is the newest graduate student to join the Human Signals Lab. He is starting in Fall 2023 and is interested in exploring the fields of machine learning and federated learning in the context of activity recognition. Welcome to the lab.
Aug 2023

Rebecca successfully defended her PhD thesis titled "Towards Lifelong and Long-Term Sensor-based Human Activity Recognition". Great body of work focused on machine learning for human activity recognition applications.
Jul 2023

Dawei's research on foreground speech analysis has been accepted to the ICASSP 2023 Ambient AI Workshop and got a best paper award. The paper is titled "A Dataset for Foreground Speech Analysis with Smartwatches in Everyday Home Environments". Congrats Dawei.
Apr 2023

Super congrats to Dawei, who received the Meta Research PhD Fellowship this year on the topic of Augmented Reality Audio. Dawei is developing a very nice body of work in this field.
Apr 2023

Very excited and grateful to have received the NSF CAREER award. Looking forward to developing new methods and wearables to advance dietary monitoring over the next several years.
Mar 2023

Xuewen, co-advised with Kaya de Barbaro, successfully defended her PhD thesis titled "Leveraging Pervasive Data to Study and Support Mother-Infant Dyads in the Wild". Congratulations Dr. Yao!
Mar 2023

Dawei's research at Meta in Summer 2022 led to a paper titled "Dynamic Speech Endpoint Detection with Regression Targets". This work has been accepted to ICASSP 2023 and will be presented at the conference in Greece this year.
Feb 2023

Happy to announce that I will be a technical program co-chair for Ubicomp 2023, to be held in Cancún (Mexico). My co-chairs will be my colleagues Prof. Silvia Santini and Prof. Judy Kay.
Nov 2022

Happy to report that Xuewen's paper titled "Understanding Postpartum Parents' Experiences via Two Digital Platforms" has been accepted to the Proceedings of the ACM on Human-Computer Interaction. Excellent work Xuewen.
Nov 2022

Rebecca's work on continual learning for activity recognition titled "Lifelong Adaptive Machine Learning for Sensor-based Human Activity" has been accepted to Sensors. This is an area of research that we are very excited about.
Sept 2022

The conference was held in Atlanta and I served as the PC Chair for the co-located ISWC conference. Rebecca presented her paper on ADL recognition with a smartwatch and Daiwei presented his AudioIMUT paper as well, which was recognized as a best-paper runner-up.
Sept 2022

The team is growing. Yufei, Sloke and Alice are starting in the group this semester to kickoff new research projects and support on-going ones. They join Dawei, Rebecca, Priyanka, Xuewen, and our great research support team of undergraduate students. Welcome to the lab.
Aug 2022

We are thrilled that we have been awarded an NIH R01 grant to study and develop new digital biomarkers and analytics for cognitive impairment. Our UT Austin-based team includes faculty from the Dell Medical School and the Dept. of Speech, Language, and Hearing Sciences.
Mar 2022

Our paper describing the potential of multimodal sensing with a smartwatch to recognize various activities of daily living has been accepted for publication at IMWUT. Congrats to Rebecca and Sarnab, who put a lot work into this research in pandemic times.
Apr 2022

Abduallah Mohamed, who recently defended his dissertation and is now at Meta, turned a class project in the Human Signals course about dealing with highly unlabeled data in activity recognition applications into a nice paper for the CoMoRea PerCom workshop.
Mar 2022

Congrats to Xuewen for getting her paper on infant crying detection in real-world environments accepted to ICASSP 2022. This research involved compiling and annotating 780 hours of real-world audio data.
Feb 2022

I have been appointed as 2021-2022 Fellow of the Jack Kilby/Texas Instruments Endowed Faculty Fellowship in Computer Engineering. Very excited and grateful, thank you for your support UT ECE.
Sept 2021

Keum San successfully defended his PhD thesis titled "Small-Scale Wireless Sensors for Automated Dietary Monitoring" and becomes the first graduate of the Human Signals Lab. Major milestone.
May 2021

Rebecca's paper on acoustic activity recognition with conversational assistants was accepted to IMWUT and presented at Ubicomp 2021. This work shows that it is possible to leverage gaps in voice interactions to learn about a person's context and activities.
Mar 2021

The new UT Austin-based Institute for Foundations of Machine Learning (IFML) has awarded us a grant to work on adaptive and continual learning for activity recognition applications. We are excited to extend our research work in this direction.
Feb 2021

In collaboration with colleagues at Penn State and Stanford, we are kicking off a 5-year project to explore the use of wearables to prevent kidney stones. Thanks to the National Institute of Diabetes and Digestive and Kidney Diseases for supporting this effort.
Jan 2021

We had the pleasure of hosting Professor Cecilia Mascolo from the University of Cambridge at the ECE Colloquia Seminar. Prof. Mascolo went over her recent work on the use of mobile audio data for health diagnostics.
Dec 2020