Alberta researchers use AI to detect depression in voices

Alberta research is using the timbre in a person's voice to track depression and mood over time. Caiaimage/Agnieszka Wozniak

Artificial intelligence algorithms can now more accurately detect depressed moods based on the sound of a person’s voice, thanks to research by University of Alberta computer scientists.

The study used previous research that suggests the timbre of a person’s voice contains information about their mood.

READ MORE: Alberta scientists use machine learning and Twitter to better understand our health

PhD student Mashrura Tasnim and professor Eleni Stroulia developed a methodology that combines several machine-learning algorithms to more accurately detect depression using acoustic cues.

The pair sees the technology being applied by individuals and care providers.

READ MORE: University of Alberta students get in touch with well-being, mental health through WellTrack app

“A realistic scenario is to have people use an app that will collect voice samples as they speak naturally,” Stroulia said.

Story continues below advertisement

“The app, running on the user’s phone, will recognize and track indicators of mood, such as depression, over time.

The latest health and medical news emailed to you every Sunday.
Receive the latest medical news and health information delivered to you every Sunday.

Get weekly health news

Receive the latest medical news and health information delivered to you every Sunday.
By providing your email address, you have read and agree to Global News' Terms and Conditions and Privacy Policy.

“Much like you have a step counter on your phone, you could have a depression indicator based on your voice as you use the phone,” she said.

LISTEN BELOW: Dr. Eleni Stroulia joins the Ryan Jespersen Show

Federal government data shows about 11 per cent of Canadian men and 16 per cent of Canadian women will experience a major depression sometime in their life. More than three million youth aged 12 to 19 are at risk of developing depression, according to the Canadian Mental Health Association.

READ MORE: ‘Depression isn’t like a broken bone’: Steps to overcome this common mental illness

Stroulia said this type of research helps establish baseline data on depression.

Story continues below advertisement

“This work, developing more accurate detection in standard benchmark data sets, is the first step,” she said.

READ MORE: Website uses artificial intelligence to generate realistic human faces from scratch

The paper, Detecting Depression from Voice, was presented at the Canadian Conference on Artificial Intelligence.

Watch below (June 25, 2019): Depression doesn’t just affect the mind. Here are some of the physical symptoms of being depressed.

Click to play video: 'Physical symptoms of being depressed'
Physical symptoms of being depressed

Sponsored content