Published on
Updated
Reading 2 mins.
In China, researchers have developed an artificial intelligence system capable of detecting signs of depression from a person’s voice. Given the magnitude of this mental disorder, this discovery could become an effective tool for assessing a patient’s mental health.
Researchers from Jinhua Advanced Research Institute and Harbin University of Technology in China have developed a deep learning algorithm that can detect depression from a person’s voice. The tool was trained to recognize emotions in human speech by analyzing the different vocal characteristics commonly found in the speech of patients diagnosed with depression.
A detection rate of 87%
The results, published in the journal Mobile Network and Apps, are quite promising. The detection rate is 87% in male patients and 87.5% in female patients. To achieve these results, the participants were questioned by a virtual agent about their mood and their personal life. From this information, the algorithm machine identified verbal and non-verbal indicators of depression.
“First, the voice information is pre-processed, including voice signal pre-emphasis, framing windowing, endpoint detection, noise reduction, etc.“, says the study. Several technologies such as 3D and auditory recognition were used to design this tool. The research team used the DAIC-WOZ, which is a collection of audio recordings and 3D facial expressions from patients diagnosed with depressive disorders and people without depression. A toolbox called OpenSmile, was also used to extract features from speech signals. Commonly used by computer scientists, it has made it possible to extract and classify characteristics of audio clips.
Additional help in diagnosing depression
Ultimately, one can imagine that this algorithm could serve as a tool and additional aid in the diagnosis of depression. It could also inspire other researchers to develop artificial intelligence to detect other psychiatric disorders.
Previous studies had already demonstrated the ability of artificial intelligence to detect depressive disorders. In 2021, a team of researchers from the Institut de stevens in the United States had managed to detect depression from texts published on social networks thanks to AI.
According to the WHO, 5% of adults worldwide suffer from depression. In the long term, depression can become a serious illness leading to suicide.