INPUT WANTED: music tracks and methodology for testing audibility of group delay

[charlielaub]

FINAL PROTOTYPE TESTING FORM

I think the form is now mostly ready to go. I still have to process all of the audio so the form below contains some dummy audio tracks. See if you can spot the differences...

Feel free to give it a try and then give me your feedback.

LINK :
http://audio.claub.net/temp/GroupDelay Test Form.html

 

[voodooless]

Looking much more sensible now! Maybe have it display what track is currently playing?

Also, is the reference now always the unaltered track? That may be fine, just realize that you could also mix this up. Maybe something to think about. For sure evaluating all the data will get more difficult.

 

[charlielaub]

Looking much more sensible now! Maybe have it display what track is currently playing?

Also, is the reference now always the unaltered track? That may be fine, just realize that you could also mix this up. Maybe something to think about. For sure evaluating all the data will get more difficult.

Unfortunately I cannot really make an indicator for the "currently playing" selection. Each audio track will be only a few seconds long. The user has to click the button, and on each button is a label with the option number. The user can go back and forth between the reference track and an option, playing each one as many times as they would like, and then check or uncheck the corresponding box. This can be done one option at a time to keep it simple if you prefer.

Yes, the reference track will always be the "un-altered" audio.

 

[voodooless]

Just a side note: the current tracks are of low-quality: 16-bit 22.05 kHz. I'm sure the final version will feature at least 16-bit 44.1 kHz

One concern may be that if the reference track is also among the options, subjects may be able to detect it due to the fact that it has no delay on the first playback, because it was cached when listening to the reference already.

 

[charlielaub]

Last chance to make some test signal suggestions! Don't all rush in at once...

I have gathered up a few test signals and have done some preliminary listening. I would like to have more signal that is really revealing of group delay peaking (non-linear phase) to add to the group. If you are familiar with these sorts of signals please suggest one. So far I have a square wave and an impulse type signal as synthetic test signals and a few music tracks.