AI tool can now detect if people have cold from the tone of their voice; here's how
2 min read . Updated: 12 Apr 2023, 04:53 PM IST
Premium- Researchers from the Sardar Vallabhbhai National Institute of Technology in Surat analysed the voice patterns of 630 people. Of the 630 people analysed, 111 of them genuinely had cold.
With the wave of Artificial Intelligence (AI) coming in, people are already at the risk of losing their jobs. Now a new researcher from India have developed an AI tool to detect if someone has a cold just by analyzing their voice tone. This could identify if an employee is taking a fake sick leave in the name of cold.
Researchers from the Sardar Vallabhbhai National Institute of Technology in Surat analysed the voice patterns of 630 people. Of the 630 people analysed, 111 of them genuinely had cold. As reported by the Economist, the study made use of the harmonics or a vocal rhythm detected in human speech to analyse if they had cold or not.
As per the report, during the study the researchers told the volunteers to count from 1 to 40 and later asked them to describe as what they during the weekend. After this, they were also asked to recite Aesop's fable 'The North Wind and the Sun'. As per Business Insider, this fable was previously proved popular for phonetics researchers. As per the report, the study was able to find out if people had cold about 70 percent of the time.
As per study which is published in science journal ScienceDirect, “The proposed features efficiently capture the spectral difference between cold and non-cold speech and can be utilised to automatically diagnose common cold and related disorders." The purpose of the study was to identify cold without the need for costly and time-consuming visits to a doctor. However, employers could also use this to catch employee who pretend to be sick.
This is not the first time AI has been used for medical purposes. In another study, a method for finding high-affinity antibody therapeutics that is based on artificial intelligence (AI) was created by researchers at the University of California San Diego School of Medicine. The study published in Nature Communications, researchers used the approach to identify a new antibody that binds a major cancer target 17-fold tighter than an existing antibody drug. The authors say the pipeline could accelerate the discovery of novel drugs against cancer and other diseases such as COVID-19 and rheumatoid arthritis. Some research group was also able to use AI to detect Parkinson's through the breathing patterns of the patients. Currently, work is currently underway to create anxiety and depression prediction models using AI tools, a researchers said. Researchers at the University of São Paulo (USP) in Brazil said that preliminary findings from the model suggested the possibility of detecting the likelihood of a person developing depression based solely on their social media friends and followers. The findings are published in the journal Language Resources and Evaluation.