A Linguist Explains Why 'Laurel' Sounds Like 'Yanny' | Hacker News Reader

A Linguist Explains Why 'Laurel' Sounds Like 'Yanny'

theatlantic.com

42 pointssethbannon63 comments

The NYT has an excellent interactive demo that lets you listen between both "yanny" and "laurel". There is a slider of how much "yanny" and how much "laurel" you want to hear. I was not able to hear the other word until I played around with this demo.

Interestingly, after playing with the demo long enough, I could hear both words spoken at the same time.

https://www.nytimes.com/interactive/2018/05/16/upshot/audio-...

I had an odd experience where I would get hysteresis that depended on how fast I adjust the slider. Did anyone else get this?

E.g. if I start on the 'yanny' side and very slowly move it down, I hear 'yanny' all the way down to the 'laurel' side.

If I pause for a few seconds, I finally start hearing 'laurel', and can hear all the way to the other end too. However, moving the slider a bit more breaks the illusion.

Indeed, I had a similar experience. Perception is strange.

I had a similar experience too, but I'm not positive whether it's based entirely on perception (which is certainly possible) or some kind of bug in the demo. It would be nice to save individual Ogg files at particular points on the continuum and try more of a blind test to see how much influence there is from what we've just heard before. (I guess I could generate those myself easily enough with sox or something...)

While I was using the slider on my computer, I kept a laurel/yanny video playing on my phone

=> I think I can confirm there is no bug in the demo.

Wow, I experimented with it more and it's a matter of context, just as you and other people in this thread have described. That's really striking. Thanks for confirming that it's not a bug.

This is maybe an even more extreme phenomenon (from the same subreddit that was apparently involved in making the Laurel/Yanny thing go viral this past week):

https://www.reddit.com/r/blackmagicfuckery/comments/8jxzee/y...

With this video, I find that I consistently hear whichever phrase I'm thinking of at the moment! (In this case either "brainstorm" or "green needle".)

Edit: Also, if anyone in this thread likes this stuff then you might really enjoy

https://www.wnycstudios.org/story/91513-behaves-so-strangely

if you've never heard it before (or even if you have).

I had a similar experience, and for a few minutes I thought the demo was buggy. The original sound sounds like "yanni" to me, but if I moved it to "laurel" and back to the middle, the middle sound would sound like "laurel" as well. At first, the only way I could get it to stop astounding like "laurel" was to pause it for a few seconds. After this happened a few times I realized it wasn't a bug, but that I was somehow subconsciously adjusting which party of the sound I was paying attending to, and then I was able to shift my attention to whichever I wanted.

Same here! For the first time after 24 hrs I hear Yanny, I slide all the way back past 50% and it stays Yanny until much lower. Crazy stuff.

Exactly!

And even weirder I can now choose which to hear! I went back to the original recording off twitter or wherever and I can now force myself to hear either at that pitch.

Before I couldn't hear Yanny at ALL.

Very very cool.

Something is weird with this demo. I was hearing yanny the first time, slided towards laurel, which I heard, and sliding back in the yanny direction, I can't hear anything else than laurel anymore, even at the rightmost yanny position.

However, if I listen to the samples at https://twitter.com/xxv/status/996462632998711297 I can hear yanny.

It's nothing weird with the demo, it's humans that are weird.

Its hysteresis. Also there are 2 different frequencies. Our ears tune for particular ones depending on the history.

That's quite fascinating. I found there was quite a lot of hysteresis. Initially I heard "laurel"; I moved the slider to the right slowly, and eventually it changed to "yanny". Then I moved it slowly back to the left, and it took about 1 1/3 ticks before it changed back to "laurel". Finally I found a point where I could consciously choose which one I heard, which was interesting.

I never hear yanny. I do however hear a drunk guy slurring Gary, which sounds like yarry, though.

Sounds like "yelly" to me.

It makes sense that we'd occasionally find things like this, if human perception does work somewhat similarly to convolutional neural nets. Everyone trains their own feature detectors, and almost everyone ends up with something that works pretty well. (People who don't end up getting called things like "faceblind" or "tonedeaf" when their features don't work for a particular thing.) So it'd make sense that there are edge cases where two common approaches get different results. You could even argue these count as "adversarial examples".

odd, I never hear Yanny. I just hear an incredibly slow Laurel.

I've played a little with this demo and still can only hear "yanny". I've started to doubt if I know how "laurel" should be pronounced (English is not my native language).

It sounds like Lowrall

This demo is awesome! I only heard “yanny” or “gary”, and for the life of me could NOT hear “laurel” after a dozen tries.

But after using the demo, I can now ONLY hear “laurel”.

If you have the ability to apply audio effects on your speakers in real-time, pick one of the Heavy Bass presets and the clip instantly switches to Laurel.

The Yanny clip is in the high register, and the Laurel clip (from dictionary.com) is in the low register, and they've just been overlaid into a single mono clip.

The article is wrong (edit: it might be more complicated; see discussion below). The reason you can hear both is because someone took the high frequencies of "yanny" and combined them with the low frequencies of "laurel". If you apply appropriate high-pass and low-pass filters, you will consistently hear yanny and laurel respectively (assuming your speakers/headphones are capable of reproducing all the required frequencies and your ears are capable of hearing them). Essentially, it's the audio equivalent of the Einstein/Monroe illusion [1]. So this audio illusion doesn't imply that yanny sounds like laurel any more than the visual illusion implies that Einstein looks like Marilyn Monroe. It might be incidentally true that yanny and laurel do in fact sound similar (and this may in turn make the illusion more convincing), but you could probably do the same trick with any two words that have similar cadences, no matter how different they sound.

[1]: http://www.123opticalillusions.com/pages/albert-einstein-mar...

The audio was derived entirely from a recording of a man saying "laurel" (can be heard here: [1]). There's no recording of him saying "yanny" because it isn't a word. The high frequencies of "laurel" do in fact sound like "yanny".

[1] https://www.vocabulary.com/dictionary/laurel

Well, if someone doctored a recording of "laurel" to make it sound like "yanny" and then mixed it back into the original recording of laurel (or did something mathematically equivalent), then maybe both the article and my comment are correct.

Mixing two audio files is contrary to what the people who posted this to instagram & twitter stated they did.

> The reason you can hear both is because someone took the high frequencies of "yanny" and combined them with the low frequencies of "laurel".

No, I very much doubt this is a doctored recording. Wired did an interview with the person(s) who first discovered it [0]. They played the pronunciation of "laurel" from vocabulary.com [1] on their computer and re-recorded the audio to their phone, then posted it to instagram. The audio sounds distorted because it was played through laptop speakers and recorded again with a microphone.

The "laurel/yanny" effect is also reproducible with the audio from the original source at [1]. You don't need to doctor the audio to hear it.

[0] https://www.wired.com/story/yanny-and-laurel-true-history/

[1] https://www.vocabulary.com/dictionary/laurel

Nope, you are completely wrong. You're spreading desinformation like it's fact.

I ran the original sample from vocabulary.com through the Ableton Equalizer with a low-cut and high-cut respectively and can't hear anything then a clear "Laurel" every time.

another example of a similar phenomenon:

https://www.reddit.com/r/blackmagicfuckery/comments/8jxzee/y...

I wonder how similar that actually is, because I can't actually hear yanny when I hear laurel if I think yanny, but there I definitely hear whichever i think about.

Jesus. My brain is being all messed with today.

Thanks for posting, to me that was a much more potent illusion. I can hear whatever combination of brain/green and needle/storm I’m thinking of.

It’s interesting how both this and the yanny/laurel example are recordings of low quality speakers. I guess the added harmonics from the distortion in the playback causes the sounds to be more ambiguous than they would otherwise be.

I can't hear "needle", no matter how much I concentrate on it.

The is a clear S in there, also I can clearly hear it's two syllables - where do you guys hear the third?

the timing is different - the S becomes the EE in needle

I find that I have to use good headphones to get repeatable results. Otherwise I can hear the same clip differently from one repeat to the next, which is really weird.

Using Audacity and my Sennheisers, a notch filter from about 500 to 1000 Hz reliably changes laurel to yanny in the clip that's being shared on Twitter. There is an intermediate setting where I can hear both at once.

The original source from https://www.vocabulary.com/dictionary/laurel stays laurel no matter what I do.

I first heard this clip on my headphones (some cheap SONY at the office). It was definitely "Laurel". Then while driving home I hear the bit on BBC about this and when it plays over the radio a hear "yanny".

The original source sounds like it has much better quality than the ambiguous samples I'd heard so far.

one of my favorite audio illusions involves a sentence that has been distorted https://soundcloud.com/whyy-the-pulse/an-audio-illusion

That is a striking illusion!

Does anyone know how to recreate this illusion with other sentences?

There must be something wrong with my head. It sounds like a computer distorted sound every single time I hear it (except of course when they announce they are playing the clean version).

Apparently, here is the original source.

https://www.vocabulary.com/dictionary/laurel

Thanks. That's the first recording that allowed me to hear Laurel.

I'm a touch creeped out. Initially I could only hear Yanny, then I listened to the varied pitches started with -30, and ending on +40. On the first pass, everything through +30 sounded like Yanny, with a distinct shift to Laurel at +40. Walking it backward, I'm no longer able to hear Yanny at all.

Another auditory illusion is the Shepard tone, which has overlapping sine waves that sound like the tone is descending or ascending without end. Wikipedia has a sample:

https://en.wikipedia.org/wiki/Shepard_tone

MyNoise.net also has that, along with some other interesting audio effects.

https://mynoise.net/NoiseMachines/shepardAudioIllusionToneGe...

What I find just as interesting is how this "went viral". Nearly every publication has a story on it. How did they all decide to publish - at nearly the same time? There has been no lead up here, just instantaneous.

I think it pretty much comes down to social media. All the people that are publishing initially saw it at about the same time and it seemed like something that would get the clicks.

slow news day

In the same way a 'just released' book is instantly #1 in the bestseller lists, and 'just released' music tops the charts almost immediately - paid promotion. Almost all 'viral sensations' from 'influencers' have teams of paid promoters behind them to get what they post in front of as many eyeballs as possible, as quickly as possible.

Top Tier Social Media 'celebrities' can directly influence the sales of almost anything if they promote or demote it enough, so there's a lot of money and effort moving around in the background to make all that happen.

The sad thing is, that in desperation to keep the hip kids clicking or viewing ads on their sites, all the news outlets just lap up this crap on a daily basis. It's a vicious circle.

Why would there be "lead up" for a viral phenomenon?

There seems to be a lot of discussion and effort around a method to allow people who hear it one way to hear it the other: pitch shifting, change volume, different speakers, etc. I found a simple one, that worked with a basic laptop speaker and middle-aged ears.

If you hear "Yanny", put your hands over your ears. The tighter your hold them down, the more high frequencies are filtered out and you will hear "Laurel".

You can modulate that to hear both. Might work the other way too. Cup your hands and aim them at the speakers so you get more high frequency. Someone give it a try, eh?

I agree that muting or muffling allows the presumably original recording (of "Laurel") to come through better. I was hearing a nasally high-pitched "yanny" with a very faint low frequency monotone "laurel" until I heard it on a smaller speaker from across the room, when it was suddenly a normal speaking voice saying "Laurel".

Surprised there wasn't more discussion here about it that I could find. Seems like this was possibly found in a study of adversarial examples for audio codecs/speech processing systems?

Nope, it's just a crappy recording:

https://www.wired.com/story/yanny-and-laurel-true-history

I am another one of those cases who hear "laurel" for most of the spectrum and something like "yelly" at the extreme right side.

After hearing "yelly" at the rightmost position of the slider I still hear "yelly" as it slides to the left, becoming "laurel" at about one third of the way. After that, as I slide back to the right, I do not hear "yelly" until the rightmost position.

It's strange to think that this going viral might be the widest reaching implication of the entire life of the person who originally posted this.

(A story on the origin of this, if you're interested: https://www.wired.com/story/yanny-and-laurel-true-history/ )

On the Laptop (Macbook Pro) speakers at the midpoint I consistently hear Yanny. Upon reading a comment below, I decided to try earphones, in which case, my midpoint became Laurel.

Update: The weirdest thing is that now on my speakers also, at the midpoint I have become a Laurel person..

I initially heard yanny a couple of times and then laurel. Then I was able to switch a few times before laurel won out ever after with a yanny whisper.

With the NYT slider I was able to adjust my yanny boundary leftward through some training.

I've heard all possible permutations (Laurel, Yanny, neither, and both) depending on the putative source. All on the same machine. Compression and transcoding might be a bigger variable than psycho-acoustics or age.

So this is the auditory version of blue/yellow dress.

Obviously, yes, since the phenomena is called the new "dress".

I disagree. That was an example of people making different assumptions and being very stubborn. This is actually hitting deep flaws in audio processing and it's exceptionally difficult to see/hear the alternative. You can't just say "oh, the lighting seems like X but might be Y", it can be impossible to hear the sounds that make up the other version.

Sorry, but the people who see white/gold aren't just being "stubborn". It is difficult for them to see the alternative.

You're reading bias that isn't there. Both were equally stubborn!

I hear "Yarry" ¯\_(ツ)_/¯

Reply on news.ycombinator.com