OT: Yanny or Laurel?

[Andy Burns]

Mostly I heard Yaurel, but depending on how I tilted my head I could sometimes hear Laurel, but never Yanny.

[Brian Gaff]

I cannot hear that at all, it definitely starts with a y. I guess you could load it into a sound editor and boost the top end to see when it sounds like that to you. On the other hand try removing the top for the reverse.This is not a new idea you know, it has been around for years, though not just one word its why some people with hearing loss cannot understand some people with bassy voices. Brian

[T i m]

;-)

Cheers, T i m

[T i m]

Nope, still Laurel all the way for me. ;-(

Cheers, T i m

[Grumps]

Predominantly Yanny, but I can also hear Yaurel. Have never heard the L of Laurel.

55 with mild tinnitus (as a lot of responders seem to have).

[John Rumm]

How about on this version with pitch shifting:

[Dave W]

I could only hear yanny with slider at centre. There was a very narrow range on the slider less than 1 division to right of centre, where I could hear yanny and laurel both together. I am 74 but still able to hear up to 15KHz.

[Max Demian]

Still "yelly" when the bloke says we should hear yanny.

Anyway, the original recording *was* laurel, so all the yannys just have cloth ears. Or are Greek.

[T i m]

Yup, in that example I hear both easily (and no ambiguity). ;-)

So, that also suggested that the core recording was of 'Laurel' and the Yanny overlaid ... so does that mean the Laurel camp weren't so easily distracted (but could still actually 'hear' the overlaid stuff).

Cheers, T i m

[T i m]

Ok ...

Yes, they do don't they?

I wonder if there is any tie between d-i-y'ers and the potential exposure to noisy tools ... or possibly the likelihood of being 'practical' so a greater chance may be involved in machines and engines etc?

Cheers, T i m

[Martin Brown]

I couldn't hear the LF laurel at all without actually low pass filtering it even after being told that was what was there. The hard HF Y sound at the start influences how the rest of the word is interpreted so much.

Strictly I heard "Yammy" rather than Yanny.

Its a curious example of an auditory illusion.

I suspect most of the information is in the sub 4kHz band but it is most curious how the brain latches on to one or other interpretation and for the most part won't let go even when you are told of the other.

[Reentrant]

Also worth looking for "McGurk Effect" demos on Youtube.

The sound is a constant sort of "Vuh" but what you hear depends on whether the speaker is making a B or F shape with his lips. Even when you know that, the brain insists on hearing what the eyes see.

[John Rumm]

It probably suggests that the brain is highly tuned to use the higher frequency content to aid intelligibility when its available. So words with a crisp treble included, tend to be resolved more easily. Its one of the reasons that intelligibility drops on a bandwidth limited telephony channel.

A more muffled recording (or a full spectrum recording where the listener has attenuation of some of the higher frequency content due to their hearing) will give fewer cues to aid decoding.

So if you hear the "extra", your brain is lead to use that extra information, and will follow that path.

In this case the additional higher frequency cues have been added to aid decoding - but have been deliberately designed to aid decoding of a completely different word from the base recording.

[John Rumm]

I would suggest its the other way around; the recording has added higher frequency content designed to misdirect the brains interpretation. The "cloth ears", don't hear the higher frequency stuff and so don't get misdirected.

Much like if you print a sign with half the information rendered in red on green - some colour blind viewers may not see that information at all.

[John Rumm]

Indeed - and illusion is the right word - since it has been deliberately constructed such that components of the recording come from different base sounds.

Yup, almost certainly - hence why the original 300Hz to 3400Hz bandwidth was selected for a telephone channel.

However when the bandwidth increases, we certainly find it easier to understand what is being said.

Indeed, human hearing and processing is astonishingly sophisticated and complex - and many aspects of it are not well understood as yet (especially the way we do 3D processing of sound). So perhaps not surprising that one can engineer sounds that lead it down various rabbit holes.

[T i m]

That makes sense. ;-)

Cheers, T i m