[scuttle]

Quote:

try a planar magnetic, maybe the HE500

Hifiman have a rep for uneven build quantity. For science you'd want  headphones to be as repeatable as possible. I'd go for Audeze LCD 3s if the budget is big or Fostex TR50's if it is low - modded (search for the thread with mods and FR graphs) these get very flat and they're $100 a pair.

 

[jerg]

Quote:

Hifiman have a rep for uneven build quantity. For science you'd want  headphones to be as repeatable as possible. I'd go for Audeze LCD 3s if the budget is big or Fostex TR50's if it is low - modded (search for the thread with mods and FR graphs) these get very flat and they're $100 a pair.

I'm not so sure, the HE500 measurements at least in terms of FR are quite consistent. It is indeed not neutral though, there is a slight mid-bass hump, and gentle ridge in mid-treble flanked by two dips.
 
AFAIK LCD3s are MUCH, MUCH less consistent in measurements. It has something to do with their extremely thin diaphragm membranes with which tension is hard to control.

 

[vid]

Looked through the first few pages of hits on Google for the search term "vowel perception headphones", which indeed returned mostly research papers. Most of these papers were content with stating only that they used headphones. A few specified the model: Sennheiser HD 414 SL, Sony DR-S3, and Sony MDR-V6 were used in those cases. In addition, two papers used "Sennheiser headphones".
 
I also came across one specific reference that may or may not be relevant: "Hirahara, T. and Ueda, K. (1990). Investigations of headphones suitable for psychophysical experiments. J. Acoust. Soc. Am. Suppl. 1, 79, S142, FFF1". Since I don't have access to that journal at home, can't say whether it's useful or not. The text that makes the reference (Hirahara & Kato 1992: The effect of F0 on vowel identification [in Speech Perception, Production and Linguistic Structure, p. 95]) does so like this: "all stimuli were presented to subjects through headphones (STAX SR-Lambda Professional [Hirahara and Ueda, 1990])".

 

[Joe Bloggs]

Quote:

@Bluepumpkin
The idea is to be able to drive the auditory system in a predictable manner using synthetic sounds without having to compensate for the frequency response of the headphones. We played a sine sweep and measured the response of the Senns at a simulated eardrum. It was actually quite close to flat, except for the aforementioned bump at 2-4khz. Because of the fragility and difficulty of insertion of IEMs we've kinda ruled those out.

 
What's the resolution of your spectrum analysis?  There should be quite a bit more ups and downs in the measured FR of any circumaural headphones in the treble range.
 
Most headphones are engineered to simulate the FR of loudspeakers in a free field or diffuse field.  The bump at 2-4kHz is one of those traits.  If you want flat at the eardrum you should measure the impulse response of the headphones on the subject at the eardrum with a probe microphone and then modify your test sounds with a corrective digital filter.
 
Will it be a problem for your experiment if the FR of the headphones changes and the frequencies of resonant peaks change every time the headphones are put on and off?  Coz they do.  IEMs have more repeatable acoustics if fitted properly.
 
Care to tell us more about your experiment?  What are those synthetic sounds to be used for, such that they have to be presented at the eardrum with no change in frequency response?