DPChallenge: A Digital Photography Contest You are not logged in. (log in or register
 

DPChallenge Forums >> General Discussion >> People needed for web-based audio experiment
Pages:  
Showing posts 1 - 17 of 17, (reverse)
AuthorThread
09/02/2009 03:36:11 PM · #1
I am a Master's Student at Stellenbosch University doing research on spatial audio and I require subjects to do a web-based auditory experiment. The experiment only requires a browser with Flash installed and stereo headphones (the experiment will not work with loudspeakers). Link to the experiment. The experiment should only take about 10 minutes to do and all the assistance that I can get is helpful :) Feel free to spread the link around.

-Konrad
09/02/2009 06:24:14 PM · #2
I did 5 tests before the OCD kicked in.

Good luck with the research!
09/02/2009 06:58:54 PM · #3
I did all 10 that was cool to do.
09/02/2009 06:59:34 PM · #4
That was fun! Did 12 before I found my mind wandering. The first white noise test scared the crap out of me -- didn't expect that!

I don't know if there is a way you can configure the flash player, but I found that mine would periodically go into "buffering" right in the middle of playback. It might be nice to configure it to wait for the entire sample to load before playback starts if possible.
09/02/2009 07:09:09 PM · #5
Did my meaningful 10. Best side comments:

steam boats challenged the wilderness
Contracted by contact
said to fall among the invaders
stared at the face of the dead man
09/02/2009 09:47:38 PM · #6
I did just three...

SO hard to understand another language on that audio mess!!! I gave up because i think that my tests would distort the analisys...
09/04/2009 07:30:40 AM · #7
Thank you to everyone who has done the experiment so far.

Originally posted by cdrice:

I don't know if there is a way you can configure the flash player, but I found that mine would periodically go into "buffering" right in the middle of playback. It might be nice to configure it to wait for the entire sample to load before playback starts if possible.
The player is supposed to buffer 5 seconds of audio before playback (longer than the sample in most cases) and I will look into why it is not doing this correctly.
09/04/2009 08:25:46 AM · #8
I did 12 but I think you need many more questions at the start to cope with variations due to nationality etc. Being a native English speaker is likely to make a big difference and there may even be a difference between British English speakers and American English speakers.
09/09/2009 06:31:59 PM · #9
Originally posted by paynekj:

I did 12 but I think you need many more questions at the start to cope with variations due to nationality etc. Being a native English speaker is likely to make a big difference and there may even be a difference between British English speakers and American English speakers.

The level of English (or nationality) that a person possesses will make a difference in the results, but they should be consistent across the various presentation modes.
09/09/2009 08:35:25 PM · #10
I gave up after two tests because I just plain cannot make out anything.....it's all run together for me.

I have extraordinary hearing in some instances, but this test hasd proven beyond a doubt that I cannot differentiate background noise! LOL!!!
09/10/2009 03:06:31 AM · #11
Originally posted by NikonJeb:

I gave up after two tests because I just plain cannot make out anything.....it's all run together for me.

I have extraordinary hearing in some instances, but this test hasd proven beyond a doubt that I cannot differentiate background noise! LOL!!!

The presentation mode and number of maskers that you get on each test is randomised and in order to test the thresholds of the various presentation methods I designed the experiment so that some cases are extremely easy and that others are next to impossible. It is highly likely that the two tests you pulled came from the difficult cases.
09/16/2009 08:49:25 AM · #12
I have other experiment up and running. The experiment is similar to the first and the procedure is exactly the same. Headphones mandatory as before. If you have some time to spare please help me in my MScEng research by doing the experiment. For me to get meaningful data you need to do at least 10 runs (which should take about 10 minutes). Some of the test cases are meant to be almost impossible to follow as I wish to test the limits of the encoding methods, so do not feel discouraged if some of the runs are very unintelligible.

Link to the experiment.

Thanks,
Konrad
09/24/2009 08:44:00 AM · #13
bump
09/24/2009 09:18:38 AM · #14
I'll try it on the morning when it's quieter here. ;-Þ
09/24/2009 11:13:00 PM · #15
Yep, the white noisy ones were the hardest.
09/25/2009 02:21:11 AM · #16
Maybe I'm too old. Total and complete mangled audio crap. Nothing resembling the selections given was heard by my ears. Maybe I'll try again with the phones reversed but I'm pretty sure they were aligned properly.
Maybe there were some instructions I didn't see?????

Went back and did 10, only 3 runs did I get most of the set commands. Weird. I hope this isn't the basis of a new voice codec for cell phones. I won't be able to hear a damn thing!!!!

Message edited by author 2009-09-25 02:37:33.
10/07/2009 10:14:47 AM · #17
I have finished setting up all three of the web-based psychoacoustic experiments for my master's research. Please take some time and do some of the experiments. Each experiment should take about 10 minutes. Headphones are required for the experiments, as the auditory processing done does not work with loudspeakers.

The experiments are designed to stress test some algorithms that make multi-speaker conversations easier to follow and understand. The difficulty of each test run is randomly selected, so it it possible you will get cases that are next to impossible, please don't give up too quickly.

Links:
Experiment 1
Experiment 3
Experiment 2
The experiments listed first have the least results so far, so please do then first.

Please pass these links (or a link to a page with them on) around.
Pages:  
Current Server Time: 01/28/2020 06:24:44 PM

Please log in or register to post to the forums.


Home - Challenges - Community - League - Photos - Cameras - Lenses - Learn - Prints! - Help - Terms of Use - Privacy - Top ^
DPChallenge, and website content and design, Copyright © 2001-2020 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Proudly hosted by Sargasso Networks. Current Server Time: 01/28/2020 06:24:44 PM EST.