BMC to Conduct Voice-Based COVID-19 Testing; How Does It Work?

The voice-based analysis has been conducted in many countries across the world. 

Updated
Coronavirus
3 min read
The voice-based analysis has been conducted in many countries across the world. 
i

As part of its efforts to ramp up testing for COVID-19, the Brihanmumbai Municipal Corporation (BMC) will be assessing an AI-based technology which uses voice samples to detect the coronavirus within seconds.

This technology is developed by an Israel- based company and the test can be conducted using smartphones, tablets or computers.

According to a PTI report, Suresh Kakani, Additional Commissioner, BMC, said, “This technology is in the initial stage and its results need to be cross-checked. Therefore we have decided to conduct this study.”

The voice-based analysis has been conducted in many countries across the world, including the US, France, Italy and Israel.

AI-Based Tech; What’s the Process?

Patients taking the AI-based test will be asked to count numerals from 50 to 70. 'Voice biomarkers' will be used to detect COVID-19 infection, civic-run Nair Hospital’s Dean Dr Neelam Andrade said.

“When a patient develops symptoms of COVID-19, he or she faces breathing problems which affect the amount of air exhaled, thereby interacting with inflamed muscles on its journey to generate voices or speech.”
Dr Andrade

Depending on its variations, an individual can be diagnosed with COVID-19.

Speaking to The Hindustan Times, Dr Andrade said, “There is a voice application which can be installed on phones or laptops. The suspected patient will be asked to count a few numbers before the device, similar to conducting a breath analyser test. The voice sample will automatically get synced with the main server of the provider. Then, through artificial intelligence, the result will be procured within 30 seconds.”

According to an infographic shared by Shiv Sena MLA Aditya Thackerey on Twitter, the process involves:

  • A suspected patient will be asked to speak into a cellphone (or computer) where the voice analysis app has been uploaded. The patient’s vital parameters too will be taken.
  • The voice sample will be analysed by the app, powered by a programme based on artificial intelligence (AI)
  • The sample will be compared with the voice of a person not suffering from COVID-19 to provide an analysis. The app comes preloaded with thousands of voice patterns.

The Pilot Project

The study will be conducted on 1,000 patients admitted at the jumbo COVID-19 facility at Nesco grounds, Goregaon and will span over two-three months.

The app will analyse voices of suspected, positive as well as negative COVID-19 patients, who will be diagnosed based on their vocal biomarkers.

Since the technology is just being explored, a patient testing positive would have to undergo a confirmatory RT-PCR test, considered the ‘gold standard’ for COVID-19 testing.

How Does This Technology Help?

The voice-based test would be non-invasive and would deliver results within 30 seconds, as opposed to an RT-PCR that could take up to 24 hours and a rapid antigen which would take 30 minutes.

Such a testing technique could help identify asymptomatic people at crowded places like malls, theatres, metros etc.

Dr KK Agarwal, a cardiologist and former president of Indian Medical Association (IMA), told HT,

“For centuries, doctors have relied on sound beats of the heart to detect a person through a stethoscope. Now, the same technique can be used to screen suspected Covid-19 patients. This technology may be new but is being widely used across the globe.”
Dr KK Agarwal

“We have seen that people often pop in fever suppressants to avoid getting detected. Later, they spread the virus to more people. But this technique, which is completely non-invasive, can detect infected people easily especially at railway stations and airports,” added Dr Andrade.

However, the risk of false positives in these tests makes a confirmatory RT-PCR test mandatory.

(With inputs from The Hindustan Times and PTI)

(Make sure you don't miss fresh news updates from us. Click here to stay updated)

Published: 
Stay Up On Your Health

Subscribe To Our Daily Newsletter Now.

Join over 120,000 subscribers!