The No.1 Website for Pro Audio
headphone doppler
Old 15th July 2005
  #1
Lives for gear
 
minister's Avatar
headphone doppler

!

did a search and couldn't find this. ... and, not sure where to post it, but i'll start here. feel free to move it.

i experience a small amount of pitch difference when i hear music come out of a monitor than in a headphone. i even notice a slight pitch change as i move the headphones up to my head. it always sounds like 10-25 cents sharper in the headphones. i know a few other people who claim to experience this too.

am i crazy? does anybody know about this?
Old 15th July 2005
  #2
More cowbell!
 
natpub's Avatar
you're not crazy, and that is exactly what it is

crank up your monitors and go stand outside.

the pitches will seem slightly lower

if you then drive by your house, you can hear the pitch shift as you are moving, same as the moving headphones
Old 15th July 2005
  #3
Gear Head
 
Johnny T's Avatar
 

I've been experiencing this for years (on a pair of Sennheiser HD25's).

The pitch is definitely a bit sharp, and as you pull them away from your ears, it gradually drops back closer to normal.

I find that it's more prominent in the low frequencies, which can almost make the bassline sound like it's out of tune with the other instruments.
Old 15th July 2005
  #4
Lives for gear
 
scott petito's Avatar
 

a prime reason I've found for out of tune singing.....especially at high spls'.... my solution quiter phones better pitch

cheers
SP
Old 15th July 2005
  #5
Lives for gear
 
minister's Avatar
Quote:
Originally Posted by scott petito
a prime reason I've found for out of tune singing.....especially at high spls'.... my solution quiter phones better pitch
cool! thanks folks! i thought this was the case... i used to play in one of those rock bands that your kids talk about, and i was a trained singer. but it freaked me out that i was singing flat all the time. i sounded in tune in the head phones. then, sure enough, flat in the monitors. my solution was to pull one ear off fromn the cans. now i compose and produce and do post. most singers i hire have great pitch. some, i have to have them pull an ear off. i'll try monitoring the headphone feed. (as well as how much lows and highs i send them)
Old 15th July 2005
  #6
Lives for gear
 
DrFrankencopter's Avatar
It's not the doppler effect that's causing this as there is no relative velocity between your ears and the headphones.

However, this is a real (not imagined) issue. I've brought it up before on a thread about singers singing out of tune in headphones. It seems to have something to do with SPL...the louder the cans, the worse the problem. Who knows, it might be somthing like your mouth and nose need to be exposed to the same relative pressure changes in order to keep the tympanic membranes 'calibrated', but that's just a wild ass guess.

Cheers,

Kris
Old 15th July 2005
  #7
Lives for gear
 
minister's Avatar
Quote:
Originally Posted by DrFrankencopter
It's not the doppler effect that's causing this as there is no relative velocity between your ears and the headphones.
no, i understand about doppler being relative to movement. it happens when i MOVE the headphoes to my ears. once on my ears, it sounds consistenly sharp.
Old 15th July 2005
  #8
Lives for gear
 
max cooper's Avatar
 

Quote:
Originally Posted by minister
no, i understand about doppler being relative to movement. it happens when i MOVE the headphoes to my ears. once on my ears, it sounds consistenly sharp.

If it was doppler, the pitch would seem to rise as the waves compress (as in a train moving towards you with the horn sounding) but once they were on your ears (and therefore stationary) the pitch would return to it's original value.

The fact that the pitch rises as they approach your ears and the STAYS sharp means it's something else. Another way you can tell if it's doppler or not is to move them towards your ears very slowly. If the pitch rises at the same rate relative to position, it's not doppler. The train's horn doesn't rise if you walk towards a stationary one (well, not much .)

If a train moves towards you fast enough you hear the pitch of the horn rise, but if the train then stops next to you and blows the horn, the pitch is back to normal.

I do hear the pitch effect with headphones. It's another reason I love to use an SM7 and listen to myself through monitors.
Old 15th July 2005
  #9
Gear Guru
 

OK let's skip for now the need for a scientific explanation of why pitch would change on headphones but not on speakers.

My question is this- if wearing the headphones makes everything sound sharp, and if you were getting your voice folded back into the cans - wouldn't that make your voice sound sharp as well as make the track sound sharp? So wouldn't you just adjust your singing to be in tune to the track, and wouldn't both be off by the same amount so wouldn't they both be in tune when you played it back?



by this logic the effect should be less at higher SPLs, (because the vocals in the cans drown out any bone conductions) yet people are reporting it as worse, and people are reporting improvements by taking one side of headphones off.
Old 15th July 2005
  #10
Lives for gear
 
DrFrankencopter's Avatar
Quote:
My question is this- if wearing the headphones makes everything sound sharp, and if you were getting your voice folded back into the cans - wouldn't that make your voice sound sharp as well as make the track sound sharp? So wouldn't you just adjust your singing to be in tune to the track, and wouldn't both be off by the same amount so wouldn't they both be in tune when you played it back?
Good question...had me convinced at first. The singer is in a feedback loop at this point in time, but there will be overshoot. For purposes of illustating an example lets imagine that the mix comes back 1 semitone sharper than it actually is. Say the actual tune requires the singer to sing C, while the singer hears C#. The singer will start to sing a C# to be in key with what (s)he's hearing, but this will be heard back as a D in the headphones...the singer will realize that they are going sharp and flatten the note naturally, but there's a pitch warble owing to the finite reaction time of the singer.

How's that for a theory?

Cheers,

Kris
Old 15th July 2005
  #11
Gear Maniac
 

Has anyone ever experienced a drop in tempo when you go into another room? When I was just starting out, I used fruity loops, and I use it once in a great while for sequencing now. It seems to only happen in fl. When I walk into another room, the tempo seems to drop down. It doesn't happen when I'm listening in cubase, only in fruityloops. Does anyone have a reason for this?
Old 15th July 2005
  #12
Gear Guru
 

Quote:
Originally Posted by DrFrankencopter
.the singer will realize that they are going sharp and flatten the note naturally, but there's a pitch warble owing to the finite reaction time of the singer.

How's that for a theory?
I don't know...
It seems that there might be a warble at the very beginning, but that once you had 'compensated' for this pitch variation it would be stable - assuming that
1) the "headphone doppler" effect actually exists and
2) that it is not constantly slewing up and down somehow

I am more than a little dubious about the whole thing. Thousands of vocalists have recorded with the headphones on at high volumes, and they all didn't go flat or we would have very little music to listen to.

As others have correctly pointed out, the Doppler effect only applies to temporary pitch changes caused by relative motion. Once the headphones are ON... what could be causing it? Perhaps evolution has designed our ears/brains to perceive closer sounds as higher in pitch as an attention-getting device? We do tend to perk up at higher pitched sounds and a nearby Sabre-Toothed Tiger is more alarming than a distant one.

djavid15-
As far as the tempo shifts are concerned, pitch and tempo are just micro and macro of the same thing- so if the effect exists for pitch it should also manifest itself as tempo. When there are sample rate boo-boos, I often notice the tempo shift first (drummer)

If Proximity _does create the perception of sharp (faster) then distance should sound flat (slower).


As far as Fruity Loops sounding slower in the next room and Cubase not, now you are just pulling our chains. tutt
Old 15th July 2005
  #13
Lives for gear
 
DrFrankencopter's Avatar
Quote:
I don't know...It seems that there might be a warble at the very beginning, but that once you had 'compensated' for this pitch variation it would be stable - assuming that
1) the "headphone doppler" effect actually exists and
2) that it is not constantly slewing up and down somehow
Yes, a warble at the beginning of every single note...really just leading to an overwhelming feeling of "I just can't sing naturally with these damn headphones on".

Again, it's not a doppler effect....there's no relative velocity. It's something else. It's a function of physiology and headphone design. I swear it exists.

Cheers,

Kris
Old 15th July 2005
  #14
Gear Guru
 

Quote:
Originally Posted by DrFrankencopter
Yes, a warble at the beginning of every single note.
this does not seem logical to me.
the way I see it:

Some phenomenon makes the pitch seem higher overall.
You sing higher to compensate-
You (maybe) overshoot the pitch, but then
You hear the overshoot, settle down and then _that's _it for the rest of the song .

You've "found" the pitch and the feedback is consistent with what you do- sing a half step up and you hear it a half step up. It would be like putting a capo on your guitar.

why aren't we all detuning the track in our DAWs (or vari-speeding our tape decks) so that singers with headphones can sing in tune? I believe it is because singers with headphones can already sing in tune and so it isn't necessary.

can the fact that some people sing more in tune with one headphone off be attributed purely to SPL? That is the interesting question to me: Is it volume or is it proximity or is it something else- like the "equal pressure" idea? Would a singer such as Max Cooper who likes to sing in front of the monitors, sing more in tune if the monitors presented the exact same level at his eardrums that the headphones did?

Quote:
Originally Posted by DrFrankencopter
Again, it's not a doppler effect....there's no relative velocity. It's something else. It's a function of physiology and headphone design. I swear it exists.

Well I am a drummer, so what do I know about pitch? heh I don't doubt what you are hearing, and you are clearly not alone, so there must be something going on. But what?

I have always noticed that pitch "tolerance" increases with volume. i.e. that the louder a band plays, the more likely they are to be out of tune, but I always assumed it was a um, "sociological" phenomenon.
Old 15th July 2005
  #15
Quote:
Originally Posted by DrFrankencopter

How's that for a theory?

Kris


Bunk.

There is another path to consider. Since the vocal cords are only 3-4 inches from the skull, there is a lot of conduction to the ear directly (the middle ear). That path sounds the same no mater what (shift the pitch 1/4 octave if you like).

Seems that if the vox mic is loud enough in closed cans, the effect should be minimized.



-tINY

Old 15th July 2005
  #16
Gear Maniac
 
Larrchild's Avatar
 

Yeah Tiny's on it. It's the pitch from direct bone-conduction fighting the pitch of the headphone-pressurized tympanic membranes. One ear off, lets em hear their head rattling again. ( ed: the rattling is then in tune with the sound entering their ear, and all is well)
Old 15th July 2005
  #17
Gear Guru
 

so let me try and grasp this-

If headphones are indeed "sounding sharp" and you take one side off to hear your vocal cords in "correct" pitch via bone conduction, won't you own voice appear to be _flat relative to the track?? and wouldn't you hear your voice at two different pitches- one "sharp" through the cans and one "correct" through the air?




If you had a will of steel and perfect pitch you could grit your teeth and sing "correctly" (i.e. flat to the track) or you could compensate by singing higher in which case your voice will come out sharp when played back on the monitors.


I think these contradictions are evidence that the phenomenon is suspect, or that it is an illusion of perception like everything looking red when you first remove the blue sunglasses. When you take off one side of the cans you are simply decreasing the overall amount of track and increasing the overall amount of vocal you hear.

"more me"

I think if you did the same adjustments with the cue send you would also improve the singer's pitch.

does this only happen to singers? what about someone with variable pitch and minimal bone conduction- like a pedal steel guitarist? A theramin player? do they play out of tune with headphones? (OK theremin players are rarely in tune anyway, but my point is the same)

And what about drummers:

to perceive an A=440 as sharp, it might have to be as much as a quarter tone higher (i.e. faster) that would be about 453. In tempo terms, that's like going from 110 BPM to 113 BPM! Any good drummer would feel that for sure. I know a lot of drummers who take one side of the headphones off, but its not because they feel that headphones make the track "faster".

Again, I am not trying to tell anybody what they are hearing, but obviously it has to be an illusion of some kind or we would never be able to overdub anything.
Old 16th July 2005
  #18
Lives for gear
 
DrFrankencopter's Avatar
Check this out:

http://www.audiomedia.com/archive/re...earsanalog.htm

Particularly relevant quote from the text:
Quote:
The listener might actually perceive a slightly lower (or higher) pitch. For a singer, it means that inside his or her head their voice is in tune with the track, but in the control room, the producer is saying... "that’s great, but you were a little flat on that take." It’s virtually impossible for the singer to sing in tune because they are in tune inside his or her head! Even worse, if the producer also has the volume in the control room cranked up, because his or her accuracy is also diminished, he or she might not notice the singer was off pitch.
and

http://www.acoustics.hut.fi/teaching/S-89.320/KA6b.pdf

and

http://hyperphysics.phy-astr.gsu.edu...und/pitch.html

I believe it to be a real effect.

JoeQ, pitch and tempo are different....they are linked by speed of the recording medium, but they are not necessarilly dependant (i.e. you can do a pitch shift while keeping tempo constant).

Regarding the concept of singing closed loop to a pitch shifted source....no you wouldn't just naturally start singing a little flat to compensate, you need that closed loop feedback on every note to know if you're in tune or not. You're a drummer, well imagine playing drums to a headphone feed that is delayed by 200 ms (not in time with the music). You now need to play well ahead of the music in order to make the playback you hear sound right....how's your playing going to be? I bet you won't be 'in the pocket'...just like a singer won't be solidly 'in tune'.

Cheers,

Kris
Old 16th July 2005
  #19
Gear Maniac
 

Quote:
Originally Posted by joeq


As far as Fruity Loops sounding slower in the next room and Cubase not, now you are just pulling our chains. tutt

I know it sounds ridiculous, but It's done it many times. I can listen in cubase and walk into another room, and it still maintains it's tempo, but in fl, it seems to slow down. I know that this doesn't seem logical, but I can hear it. I don't know, maybe I'm just crazy.
Old 16th July 2005
  #20
Gear Guru
 

good stuff. So if Place Theory is correct and higher volumes excite adjacent cillia in the cochlea that would not only explain why pitch perception goes out the window when things get loud, it would also explain why headphones are worse than speakers for this - because they are attached to the head and can vibrate the cochlea directly. This totally makes sense to me.

On the other hand, everybody here says "sharp" and Johnny T specifically reports bass frequencies going sharp, whereas the chart by Karjalainen shows frequencies below 2k perceived as going _flat as SPL goes up

according to the chart, 60db is the perfect level for pitch perception. I will try that on monday's vocal session and see if the singer goes for it. heh



Quote:
JoeQ, pitch and tempo are different....they are linked by speed of the recording medium, but they are not necessarilly dependant (i.e. you can do a pitch shift while keeping tempo constant).
Try doing that on a tape deck. The only reason you can change pitch and tempo separately is because your DAW is interpolating the samples -essentially creating new information. From a physics point of view, they are the same thing. A kick drum twice a second is a disco beat. A kick drum 52 times a second is a G#.

Anyway, I don't think pitch is really changing with the headphones- just the perception of pitch


Quote:

Regarding the concept of singing closed loop to a pitch shifted source....no you wouldn't just naturally start singing a little flat to compensate, you need that closed loop feedback on every note to know if you're in tune or not. You're a drummer, well imagine playing drums to a headphone feed that is delayed by 200 ms (not in time with the music). You now need to play well ahead of the music in order to make the playback you hear sound right....how's your playing going to be? I bet you won't be 'in the pocket'...just like a singer won't be solidly 'in tune'.
Point taken. But 200ms is an unfair analogy. To be 10 cents flat (which at 1kHz doesn't happen until you get to 90db according to Karjalainen's chart) is more like having a 2 or 3 ms delay, which people compensate for all the time. People using native DAWs, guitarists sitting a few feet from their amps. Hell, I have a big thick ride cymbal that takes a couple of milliseconds to react- I literally have to get a little bit 'on top' with my right hand on a fast song.

besides, you are not actually singing to a pitch shifted source (stick the mic from your tuner into the cup of your earphones- is it sharp?) - only your perception of pitch is changed and that should change for everything you hear. The only thing I can think of is if maybe singers can "feel" the pitch of their vocal cords and rely as much on that feedback as the feedback they get from their ears.


This thread makes me wonder how anybody can play music at all. If you are on stage with a band, and it's really Loud and you are several yards away from other sound sources and they from you- the timing and the pitch is going to be All Off. Then of course, the people on the dance floor in front of the band will think you are playing too fast and will stop dancing. The people outside in the parking lot will think you are playing too slow and won't bother to come in. It obviously can't work.

I had better cancel tomorrow night's gig.
Old 16th July 2005
  #21
Gear Guru
 

Quote:
Originally Posted by djavid15
I know it sounds ridiculous, but It's done it many times. I can listen in cubase and walk into another room, and it still maintains it's tempo, but in fl, it seems to slow down. I know that this doesn't seem logical, but I can hear it. I don't know, maybe I'm just crazy.
Crazy? Fruit Loopy? Cuckoo for Cocoa Puffs?

It doesn't sound ridiculous, but it does sound unscientific. Others have reported hearing the tempo slower from outside, so that part of it at least isn't crazy.

I wonder if you did a serious blind test with someone else doing the switching and matched levels and identical audio files and no plug-ins and so on if you would still hear something with one program and not the other.
Old 16th July 2005
  #22
Gear Addict
 
SpiderM69's Avatar
 

This thread's been pretty enlightening. Has anyone tried using open headphones (e.g., Sennheiser 600's) to see if it improves a singer's pitch? Of course, there'll be the issue of bleed, but I'm wondering if it might reduce the effect of conduction through the bones as well as other effects influencing perception of pitch.

As noted above, many singers have sung spot on for years, so maybe it's just another case of learning your instrument, especially in non-optimal environments.
Old 16th July 2005
  #23
Gear Maniac
 

Quote:
Originally Posted by joeq
Crazy? Fruit Loopy? Cuckoo for Cocoa Puffs?

It doesn't sound ridiculous, but it does sound unscientific. Others have reported hearing the tempo slower from outside, so that part of it at least isn't crazy.

I wonder if you did a serious blind test with someone else doing the switching and matched levels and identical audio files and no plug-ins and so on if you would still hear something with one program and not the other.
I've never done a blind test. I never really cared to even mention it, but i figured since we were on a similiar topic........
I should do a blind test and see what happens. I don't know, maybe it's just a coincidence of it always happening in fl, but I can't ever recall hearing it happen in cubase. I formulated a theory before, but then I forgot it, so I guess that doesn't help right? haha!
Old 16th July 2005
  #24
Lives for gear
 
DrFrankencopter's Avatar
I think the important thing to take away is that there is an effect of volume on pitch perception, and that it's pretty variable between people. Try adjusting headphone levels (or switching types) if pitch problems develop.

The effect that we hear (the pitch perception change as volume increases) does not change the tempo. Think of it this way, say you're playing a tune with 240 snare hits in it (on 2 and 4 for a 120 bpm 4 minute tune), the change in pitch perception does not make the song any longer in time...i.e. it's still a 4 minute tune with 240 snare hits, and therefore still at 120bpm, it's just perceived at a higher pitch. Tempo would be defined by the timing between the pulses of the drum, whereas the pitch would be defined by the tuning of the drum itself.

A kick at 52 times a second is not really a pure G#....not any more than playing the E string on a guitar at 52 times a second (which is what Yngwie Malmsteen sounds like to me heh ) is G#.. These are artificial examples, because they involve repetitions of instruments beyond our ability to play them, and beyond their ability to produce their normal sounds. Samplers can do it, but they would end up producing a G# fundamental with awful out of tune harmonics. In the practicall sense, you probably cannot produce more than 5-7 notes per second, which would make a 10Hz fundamental and be imperceptible.

Quote:
Point taken. But 200ms is an unfair analogy. To be 10 cents flat (which at 1kHz doesn't happen until you get to 90db according to Karjalainen's chart) is more like having a 2 or 3 ms delay, which people compensate for all the time. People using native DAWs, guitarists sitting a few feet from their amps. Hell, I have a big thick ride cymbal that takes a couple of milliseconds to react- I literally have to get a little bit 'on top' with my right hand on a fast song.
I don't see how 10cents can be directly compared to a 2 or 3 ms delay. Really, whats the factor is if the singer can sing within 10 cents pitch accuracy (and not all can as I'm sure most here will agree), that they will be off in pitch. If you can't hear a difference, you won't compensate for it.

I don't compensate my guitar playing for how far I am from the amp, but I do if I have a reverse delay effect on my guitar (i.e. perceivable time lag >>20ms). I'd be willing to bet that your ride cymbal has a 'crest' time more on the order of 100+ miliseconds, you can readily perceive the delay and the need to compensate for it. But, you don't compensate for the kick drum being 2 feet further from your ears than the snare by playing it 2ms early.

Cheers,

Kris
Old 16th July 2005
  #25
Gear Guru
 

Quote:
Originally Posted by DrFrankencopter




I don't see how 10cents can be directly compared to a 2 or 3 ms delay. Really, whats the factor is if the singer can sing within 10 cents pitch accuracy (and not all can as I'm sure most here will agree), that they will be off in pitch. If you can't hear a difference, you won't compensate for it.
just a very rough seat-of-the-pants estimation! I would say a few milliseconds off in timing and a 10 cents off in tuning are both in that area of "some people would notice, some people wouldn't"

Quote:

A kick at 52 times a second is not really a pure G#....not any more than playing the E string on a guitar at 52 times a second (which is what Yngwie Malmsteen sounds like to me heh ) is G#.. These are artificial examples, because they involve repetitions of instruments beyond our ability to play them, and beyond their ability to produce their normal sounds. Samplers can do it, but they would end up producing a G# fundamental with awful out of tune harmonics.
I am not saying its pure and I am not saying I can play it. Harmonics or not, crank up your sampler and you will get a G#. Vibration across time is a continum and when it is slow (under 20 Hz) we call it Tempo and when it is faster we call it pitch. A physicist friend of mind pointed out that the ratios we use most commonly for pitch -fourths, fifths etc are also the most popular polyrhythms. an A=440 against an E=660 (untempered!) has a ratio of .666666 etc. which is the same ratio as 2 against 3 - one of the first things I teach my drum students.


Quote:

I don't compensate my guitar playing for how far I am from the amp...

...you don't compensate for the kick drum being 2 feet further from your ears than the snare by playing it 2ms early.
well this is where it gets deep

I believe we _do make these compensations all the time without thinking about it. In fact the maximum speed of nerve impulses is 200 meters/sec (sound is about 340 meters/sec) which means it takes several milliseconds for an impulse from my brain to even reach my foot- more time than it takes to reach my hand. That means I have to send my kick impulse early to make it sound at the same time and I am convinced that sound delay is factored in as well. Otherwise how would we know "when" we hit the kick drum? By "feeling" it from our foot? That's slower than sound!

As someone who teaches drums to young children and the handicapped, I can tell you that making these compensations for music are not always easy or automatic at first. I think that our brains have already learned to do this somewhat by learning to coordinate the timing necessary to walk, pick things up, catch a ball and so on.

OK, so now I not only am paranoid about my ability to lock up with my bandmates, but I am paranoid about my ability to lock up with myself.

definitely have to cancel that gig
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump