The internet is flipping out over the Yanny or Laurel thing. Here’s the deal: Which ever you’re hearing is right, and it’s all based on a little audio science—and the quality of your headphones.
Yanny or Laurel is an internet meme started by Cloe Feldman with a single post on Twitter. She shared short low-quality audio clip and asked followers which name they hear. Some hear “Yanny,” and some hear “Laurel.”
What do you hear?! Yanny or Laurel pic.twitter.com/jvHhCbMc8I
— Cloe Feldman (@CloeCouture) May 15, 2018
Spoiler: The audio recording is someone saying “Laurel.” That doesn’t, however, mean if you hear “Yanny” you’re wrong.
What you hear is based on the quality of your earbuds or speaker. If the audio is compressed and the sound is pushed to the high end your brain can interpret the word as “Yanny” instead of “Laurel.”
So really, this is a test of how good your earbuds or speakers are. If you hear “Yanny” you’re probably using low quality Bluetooth earbuds or speakers. Sorry to have to break it to you like that.
Speaking of perception; fire up the color accurate big monitor here – your iOS toy won’t help here – and take this simple color test. Not to brag because I’m fading fast today, but a few years ago I took this test to get an ad agency gig (I still don’t know why) but it had like 10x the level of samples – insane!!! And, I had like two swatches out of order out of 100. Still, even here it shows how hard it is to be objective even after a couple seconds of staring….
http://www.colormunki.com/game/huetest_kiosk 🌈
Yeah, I can input a G note and apply a filter and it comes out a step up to A, or any other note. Same thing here, what he INPUT has zero relevance. Clearly this points out a paradox not unlike the blue dress white dress absurdity which you could put the Photoshop eyedropper on and end the RGB argument right there but still people will interpret things in their own way. The old – “how do you know my “green” is your “green” paradox. Also, many men are colorblind to the greens – which is 1/3 of ALL the color input humans have in the RGB space so again it points to each person’s interpretation based on their brain at least for color. Same thing here – the first “sound” in the formant would be either an “L” or “Y” sound – and that ‘attack’ can be read with both the waveform and the frequency analyzer plug ins in Pro tools and this is data from the file – not pedestrian iOS outputs thru old stereo equipment – I’m talking 192khz 32 bit sampling through dedicated audio cards – then, take the 2nd syllable as the 2nd most important sound in the clip (“nee” v “rel” or in my case “ri”-ish) and visually you can match them up closer to “yarry” than “laurel” any day. Having said that – aurally I can make the crappy 8 bit sample SOUND like anything with simple envelope filters so I think somebody is merely having some “fun” here. 🎹🎼🎧🤡
This from the person who recorded the audio
https://abcnews.go.com/GMA/Culture/laurel-yanny-debate-creator-speaks/story?id=55211906
Formants is what they call the initial “envelope” of sounds aka attack-sustain & release of human voice. There are guitar FX pedals that can “talk” by adjusting the formants and of course the legendary awesome hilarious “Miku” pedal that sings in Japanese as you play guitar riffs….
I mess with audio for decades – use Pro Tools now (even Garageband for quick-n-dirty pieces) and in my PT system thru pro Sennheiser HD cans and M-Audio BX 5a studio reference monitors I hear a crappy scooped EQ saying “yarry” – so go finger. Now to make it interesting I dragged the clip into PT and compared the waveform to me saying ALL sorts of things including the suspected sounds and “yarry” with a “megaphone” plug-in on 50% and THAT vocal looked MOST like the original web waveform. Jus’ sayin’. Yarry, out
“Laurel,” no question. Speakerwise: 2.1, Polk 12″ powered sub + L/R Polk bookshelf speakers, fed from an old JVC Class A Amp from MacBook Pro’s headphone jack. “Laurel,” also from Bose BT SoundSports on iPhone. “Laurel,” also from Apple wired earpods. “Laurel,” also from cheap-ass BT hearing protector-cans I got to listen using power equipment. “Laurel,” from Laptop built in speaker. “Laurel,” from monitor speakers via HDMI from Win10 desktop. That’s a pretty broad spectrum of price and quality. Might be less “Cheapness” than audio engineering tailored specifically to deficits in specific classes of drivers. Perhaps the audio equivalent of how LED headlights flicker on some videocamera settings more than others.
I’ve been trying to find out the answer to the “yanni” or “laurel” phenomena but I’m not convinced the whole answer is the quality of the speakers.
I only hear “yanni” and I’m listening through $200 Audioengine speakers hooked up to my Mac.
I think there’s more to it.
The key being low quality