If you haven’t been listening to pop radio in the past few months, you’ve missed the rise of two seemingly opposing trends. In a medium in which mediocre singing has never been a bar to entry, a lot of pop vocals suddenly sound great.
Better than great: note- and pitch-perfect, as if there’s been an unspoken tightening of standards at record labels or an evolutionary leap in the development of vocal cords. At the other extreme are a few hip-hop singers who also hit their notes but with a precision so exaggerated that on first listen, their songs sound comically artificial, like a chorus of ’50s robots singing Motown.
The force behind both trends is an ingenious plug-in called Auto-Tune, a downloadable studio trick that can take a vocal and instantly nudge it onto the proper note or move it to the correct pitch. It’s like Photoshop for the human voice. Auto-Tune doesn’t make it possible for just anyone to sing like a pro, but used as its creator intended, it can transform a wavering performance into something technically flawless. “Right now, if you listen to pop, everything is in perfect pitch, perfect time and perfect tune,” says producer Rick Rubin. “That’s how ubiquitous Auto-Tune is.” (Download TIME’s Auto-Tune Podcast from iTunes)
Auto-Tune’s inventor is a man named Andy Hildebrand, who worked for years interpreting seismic data for the oil industry. Using a mathematical formula called autocorrelation, Hildebrand would send sound waves into the ground and record their reflections, providing an accurate map of potential drill sites. It’s a technique that saves oil companies lots of money and allowed Hildebrand to retire at 40. He was debating the next chapter of his life at a dinner party when a guest challenged him to invent a box that would allow her to sing in tune. After he tinkered with autocorrelation for a few months, Auto-Tune was born in late 1996.
Almost immediately, studio engineers adopted it as a trade secret to fix flubbed notes, saving them the expense and hassle of having to redo sessions. The first time common ears heard Auto-Tune was on the immensely irritating 1998 Cher hit “Believe.” In the first verse, when Cher sings “I can’t break through” as though she’s standing behind an electric fan, that’s Auto-Tune–but it’s not the way Hildebrand meant it to be used.
The program’s retune speed, which adjusts the singer’s voice, can be set from zero to 400. “If you set it to 10, that means that the output pitch will get halfway to the target pitch in 10 milliseconds,” says Hildebrand. “But if you let that parameter go to zero, it finds the nearest note and changes the output pitch instantaneously”–eliminating the natural transition between notes and making the singer sound jumpy and automated. “I never figured anyone in their right mind would want to do that,” he says.
Like other trends spawned by Cher, the creative abuse of Auto-Tune quickly went out of fashion, although it continued to be an indispensable, if inaudible, part of the engineer’s toolbox. But in 2003, T-Pain (Faheem Najm), a little-known rapper and singer, accidentally stumbled onto the Cher effect while Auto-Tuning some of his vocals. “It just worked for my voice,” says T-Pain in his natural Tallahassee drawl. “And there wasn’t anyone else doing it.”