Artificial intelligence could soon diagnose illness based on the sound of your voice

There is a lot of information offered by voices. Researchers are working on an app to help diagnose illnesses. The National Institute of Health is funding a project to collect voice data and develop an artificial intelligence that could diagnose people based on their speech. Dr. Yael Bensoussan is the director of the University of South Florida's Health Voice Center and leads the study.

If you close your eyes and listen to the patient's voice, can you tell if they have a problem? Bensoussan said so. That's where we got all our information. Parkinson's disease can be caused by someone who speaks slowly. lurring is a symptom of a stroke Depression can be diagnosed by scientists. The team will collect the voices of people with neurological disorders, voice disorders, mood disorders, respiratory disorders, and children with speech delays.

The goal of the project is to create large-scale health care databases for precision medicine. Bensoussan says that they lacked large open source databases. Every institution has their own database. To allow researchers from other generations to use the data was important.

This isn't the first time artificial intelligence has been used to study human voices, but this is the first time data will be collected on this level. Everyone was doing the same thing but at a smaller level. We need to build a network and do something together.

An app that helps bridge access to rural or underserved communities is the ultimate goal. If you change your voice, you could be advised to seek medical attention.

Since the artificial intelligence can only get as good as the database it's learning from, researchers have to begin by amassing data. The goal is to collect 30,000 voices by the end of the four year period.

"If we can only collect data in our acoustic laboratories and people have to come to an academic institution to do that, then it kind of defeats the purpose," Bensoussan said.

There are a few obstacles It's not clear if researchers can share voices under the law that regulates medical privacy.

Yael Bensoussan wants you to give your voice to the project. The voice belongs to someone. How can we use it? Researchers are allowed to use it. Is it possible that it could be commercial?

Other health data can be separated from a patient's identity and used for research. A team of bioethicists will explore ethical and legal questions because every institution has different rules for sharing things.

Here are three voice samples that can be used.

There is a credit to SpeechVive in the video.

The license for the Perceptual Voice Qualities database can be found here. The audio was not changed.