The Surprising Origin of the ‘Yanny’ or ‘Laurel’ Clip That’s Tearing the Internet Apart

A viral audio clip has everyone asking each other: “Yanny” or “Laurel”? And while the clip is launching many debates, its backstory is actually quite simple.

The viral “Yanny” or “Laurel” clip — where some listeners hear a deep male voice saying “Laurel,” while others hear “Yanny” — has taken the internet by storm and recalls the fervor over the dress, the 2015 viral sensation where some saw the same dress as either white and gold in color or blue and black.

But where did the clip, which has more than 12.1 million views on Twitter, come from? The original “Yanny” or “Laurel” clip is, of all places, from a vocabulary.com recording.

Mark Tinkler, founder and chief technology officer of vocabulary.com, told TIME the original audio recording comes from an opera singer contracted to record English language words for the website. Vocabulary.com previously worked with about eight professional opera singers to record every word in English over a six month period because they are fluent in International Phonetic Alphabet, Tinkler said.

In somewhat of a disappointment to the many people who heard “Yanny” in the clip, the actual word recorded in the original clip is laurel, defined as a “wreath worn on the head, usually as a symbol for victory.”

But that doesn’t mean the viral sensation didn’t stump Tinkler, who said he’s heard both “Yanny” and “Laurel” after listening to the original clip on his website. “It’s just as much of a fierce debate in my office as it is anywhere else,” he said.

However, it wasn’t the vocabulary.com clip went itself that went viral this week. What most people are listening to is a clip from Roland Szabo, a high school student in Georgia. Szabo, 18, tells TIME he saw a video of the audio clip on a friend’s Instagram story. The friend had gotten the audio recording from a classmate who had looked up the world “laurel” on vocabulary.com, only to find that the pronunciation stumped multiple people. After no one could agree on what they were hearing, Szabo put the clip of the poll on Reddit “thinking it was going to get lost through millions of posts.”

“It blew up,” Szabo said. ” I am blown away by the fact that this became such a big deal.”

The poll really kicked off when Cloe Feldman, a YouTube vlogger, posted the clip on Twitter.

Audiologists say people are divided on what they hear because of several cognitive processes that the brain uses to give meaning to sound. Dr. Kevin Franck, the audiology director at Massachusetts Eye and Ear, told TIME that the audio clip in this instance forces the brain to decide quickly about what word it hears.

According to Tinkler, the changes come depending on what kind of technology a person is using to listen to the clip. When he heard the clip from someone else’s computer, he heard “yanny.” But when he tried the clip again back at his desk, he heard “laurel.”

“It just depends on the audio and the equipment that you’re using,” he said.