The Michigan Engineer News Center

Emily Mower Provost named Toyota Faculty Scholar

Her work uses machine learning to measure mood, emotion, and other aspects of human behavior for purposes of providing early or real-time interventions for people in managing their health. | Medium Read
EnlargeEmily Mower Provost
IMAGE:  Prof. Emily Mower Provost

Prof. Emily Mower Provost has been named a Toyota Faculty Scholar, effective May 1, a title awarded to a small group of the most accomplished assistant and associate professors in the College of Engineering to support their teaching and research activities.

Mower Provost’s research is centered on human-centered speech and video processing and speech-based assistive technology. The goals of her research are motivated by the complexities of human behavior.

Much of her recent work has been focused on the use of machine learning to measure mood, emotion, and other aspects of human behavior for purposes of providing early or real-time interventions for people in managing their health. 

Early in her career at Michigan, Mower Provost collaborated with others including researchers at Michigan Medicine on a project to create a smartphone app that would monitor subtle qualities of a person’s voice during everyday phone conversations in order to detect early signs of mood changes in people with bipolar disorder. An additional goal of the project was to identify a biological marker to prioritize bipolar disorder care to those who need it most urgently to stabilize their moods. This work is ongoing.

In 2015, Mower Provost received the Oscar Stern Award for Depression Research from the U-M Depression Center to further support her investigations into computational methodologies for use in differentiating emotion perception patterns of healthy controls and individuals from those suffering from Major Depressive Disorder or Bipolar Disorder. This is leading to the development of new measures of severity for individuals with mood disorder.

In 2017, she received an NSF CAREER grant for her research project, “Automatic Speech-Based Longitudinal Emotion and Mood Recognition for Mental Health Monitoring and Treatment.” Under this grant, Mower Provost is investigating new approaches in speech-based mood monitoring by taking advantage of the link between speech, emotion, and mood. In addition to expanding on her earlier work, this allowed her to identify new directions and links between the fields of emotion recognition and assistive technology. For example, this would allow clinicians to estimate the risk of suicidal ideation using patterns in emotion expression.

In 2018, Mower Provost leveraged her work in speech recognition to begin development of the first automated system that uses speech analysis to detect Huntington’s disease. To do so, her team partnered with clinicians at Michigan Medicine to create their own system designed to transcribe the unique speech patterns of patients with Huntington’s disease. They then used the system’s output to create a set of measures that can predict the disease.

Most recently, Mower Provost and her collaborators have worked to address the fact that tasks such as emotional recognition store a significant amount of demographic and identifying data about the audio data being analyzed, which carries significant privacy concerns. They have demonstrated that instead of storing raw data, storing representations of the data can remove this identifying information and doesn’t negatively impact performance on the primary emotion recognition task.

Prior to joining the faculty at Michigan, Mower Provost received her PhD in Electrical Engineering from the University of Southern California (USC), Los Angeles, CA in 2010. She was awarded a National Science Foundation Graduate Research Fellowship, the Herbert Kunzel Engineering Fellowship from USC, an Intel Research Fellowship, and the Achievement Rewards For College Scientists (ARCS) Award. She is currently a member of Tau-Beta-Pi, Eta-Kappa-Nu, and a member of ACM, IEEE, and ISCA.

Emily Mower Provost
Portrait of Steve Crang


Steve Crang
CSE Marketing and Communications Manager

Michigan Engineering

(734) 763-9996

3832 Beyster Bldg

The electrons absorb laser light and set up “momentum combs” (the hills) spanning the energy valleys within the material (the red line). When the electrons have an energy allowed by the quantum mechanical structure of the material—and also touch the edge of the valley—they emit light. This is why some teeth of the combs are bright and some are dark. By measuring the emitted light and precisely locating its source, the research mapped out the energy valleys in a 2D crystal of tungsten diselenide. Credit: Markus Borsch, Quantum Science Theory Lab, University of Michigan.

Mapping quantum structures with light to unlock their capabilities

Rather than installing new “2D” semiconductors in devices to see what they can do, this new method puts them through their paces with lasers and light detectors. | Medium Read