Artificial Intelligence Can Use Your Voice to Guess What You Look Like with Accuracy

News Artificial Intelligence Voice Featured

Artificial intelligence just gets smarter and smarter. At a certain point it seems like there’s no limit to what it can do. It continues to be more and more useful while at the same time being more and more awe-inspiring.

The most recent advancement in artificial intelligence technology is that it can guess what you look like with accuracy. It’s not perfect, but the results are still stunning with the knowledge it is only based on your voice.

Also read: 15 Interesting AI Experiments You Can Try Online

Speech2Face AI Guesses What You Look Like Based on Your Voice

Researchers at the Massachusetts Institute of Technology created Speech2Face, artificial intelligence that analyzes a short sample of a person’s voice and uses that to reconstruct what the person may look like.

Again, it doesn’t produce a mirror image, but it’s still very close, making Speech2Face somewhat creepy while also being stunning, noting that it uses just a very small sample of a voice.

A paper was published this past week with the MIT team describing its work of training a generative adversarial network to analyze short clips of voices to “match several biometric characteristics of the speaker.” This results in “matching accuracies that are much better than chance,” according to the researchers.

A deep neural network was designed and trained to carry out this work using millions of videos of people speaking on YouTube or the Internet in general. It captured physical identifiers of people based on age, gender, and ethnicity.

News Artificial Intelligence Voice Examples

On the Speech2Face GitHub page, researchers do raise caution as they acknowledge this technology does bring up questions of privacy and discrimination.

“Although this is a purely academic investigation, we feel that it is important to explicitly discuss in the paper a set of ethical considerations due to the potential sensitivity of facial information,” wrote the researchers.

They added that “any further investigation or practical use of this technology will be carefully tested to ensure that the training data is representative of the intended user population.”

The researchers also add the note that their goal isn’t to reconstruct an image of a person based on accuracy — their goal is to recover. identifiable physical features that correlate with the speech sample.

Where Is This Technology Headed?

As amazing as this technology is, it has to be questioned where it’s headed with this work. Like the warnings of the researchers, sensitivity needs to be used when using this technology, as it definitely seems like it can infringe on privacy.

When you speak and don’t visually appear in a recording, it’s with the knowledge that you won’t be seen. With the technology being developed, it brings up morality and legality. Many people would not be comfortable with this.

But technology such as Speech2Face can’t be stopped. Technology continues to develop, sometimes in uncomfortable ways. Does it make you uncomfortable that someone can figure out what you look like just based on your voice? Let us know what you think in the comments.

Image Credit: Speech2Face via Futurism and Public domain

Subscribe to our newsletter!

Our latest tutorials delivered straight to your inbox

Laura Tucker Avatar

Read next

In 2016, archaeologists dated two rings of snapped stalagmites in France’s Bruniquel Cave to 176,500 years ago, evidence that Neanderthals had walked 336 metres into darkness with fire and built architecture deep underground long before modern humans reached Europe
Otto von Bismarck was 74 when Germany adopted the world’s first national old-age social insurance program in 1889, setting the pension age at 70 after years of fighting socialists with bans, laws, and a promise few workers would live long enough to use
When cosmonaut Valeri Polyakov stepped out of his Soyuz capsule in March 1995 after 437 consecutive days aboard Mir, doctors recorded him at several centimetres above his pre-flight height, and his spine had become so unaccustomed to gravity that the recovery team carried him to a chair rather than risk the compression of letting him walk.
When Bell Labs engineer Karl Jansky pointed a rotating antenna at the sky in 1932 looking for sources of transatlantic radio static, he kept picking up a faint hiss that peaked every 23 hours and 56 minutes, and he eventually realized he had become the first human to hear the center of the Milky Way.
When Harvard astronomer Cecilia Payne submitted her 1925 doctoral thesis arguing that the Sun was made almost entirely of hydrogen, the field’s senior figure Henry Norris Russell talked her into adding a line calling the result ‘almost certainly not real,’ and then published the same conclusion himself four years later to widespread acclaim.
When seismic waves from the Chicxulub impact reached what is now North Dakota roughly ten minutes after the asteroid struck, they appear to have triggered a ten-metre standing wave in an inland river that flung fish onto the bank and buried them under glass beads still falling from the sky.
When survivors near Lake Nyos woke on the morning of 22 August 1986, the cattle were dead in the fields, the birds had fallen out of the trees, and 1,746 of their neighbours were lying where they had stood the night before, with no fire, no flood, and no wound to explain it.
In October 2002, a Russian scientist named Dimitri Malashenkov stood up at a space conference in Houston and quietly explained that the dog Laika, whom the Soviet Union had publicly mourned as a heroic week-long orbiter in 1957, had actually died of heat and panic within about five hours of launch.