Frequently Asked Questions
What is the purpose of this study?
Over the past 15 years, with the rapid advance of machine learning and AI technology in healthcare, it has been shown that a person’s voice can be used to detect a wide variety of medical conditions such as heart disease, neuromuscular disorders, psychiatric conditions, and even viral and bacterial infections. We are applying that same technology to the development of a convenient, rapid, readily available screening test that can detect COVID-19 by simply using voice analysis and your mobile phone.
How long will the study take?
3-4 minutes on average, and normally less than 5 minutes.
What happens during the call?
Using your cell phone and just speaking in your normal conversational voice, you will be asked to provide two 30-second vocal samples. In addition, we will ask you some basic information about yourself including your name, your gender, any symptoms of illness you might have, and whether or not you have been previously diagnosed with COVID-19. All of this information will be used to 'train' the computer algorithm on the differences between who have had and not had COVID-19.
Will my recordings be anonymous?
Yes. All recordings are non-identifiable, and no personal information is stored.
Do I need to be diagnosed with COVID-19 to participate?
No. Even if you are healthy your voice recordings will be helpful to the project.
What will you do with my recordings?
Your recordings will be used for scientific research purposes to improve the technology.
Will I be given a positive or negative COVID-19 diagnosis by using the phone app in this study?
No. Although it has been shown that this kind of technology can detect COVID-19 from voice recordings, this initiative is for scientific research only. If you suspect you may have COVID-19 and have been previously undiagnosed, please visit your primary care provider or local medical services for professional advice.
How old do I have to be to contribute?
To give informed consent, you must be 18 years and over to contribute.
How does it work? How can you detect COVID-19 from voice recordings?
Subtle changes to the voice are detected by specialized algorithms.
Where can I go to learn more about vocal analysis technology?
Vocal analysis technology has been the subject of many news reports and press releases over the past 15 years. Several good articles include:
1) “Researchers are exploring ways to use people’s voices to diagnose coronavirus infections, dementia, depression and much more” (Nature Magazine):
2) “Signs of Covid-19 may be hidden in speech signals” (MIT)
3) “New app collects the sounds of COVID-19” (Univ of Cambridge)
5) “Pfizer uses computer application to analyze speech pattern for early Parkinson’s disease detection”
How can machines help?
The technique of listening to the body is actually very difficult for humans to acquire without a lot of training, but machines are much better at it. Artificial intelligence technologies like machine learning can identify features or patterns in a sound that the human ear cannot.
Will this research be useful after COVID-19?
Yes, very much. This technology can easily be applied to a wide variety of medical conditions. It could even help people to better understand their baseline immunological and nutritional health status. Our big vision is for machine learning algorithms to be linked to wearable devices and smartphones so it can automate the diagnosis of disease through sound. While many people can visit a doctor on demand, this is often expensive and requires an appointment. For those who don’t have access to medical care, or who can’t afford the time and/or expense to see a physician regularly, this would be a remarkable way to cheaply, rapidly, and conveniently evaluate and monitor their own health as often as they wish and from anywhere they have phone access.
Where can I go to learn more about COVID-19 in general?
One of the best sources for information about COVID-19 is the CDC. [LINK: https://www.cdc.gov/coronavirus/2019-ncov/cdcresponse/about-COVID-19.html]
You can also track the prevalence of the virus in the USA and around the world using the Johns Hopkins Coronavirus Resource Website. [LINK: https://coronavirus.jhu.edu]