Defining Sound Recording Timing
Hi,
I am preparing a picture naming task. In each trial, I show a dot, then I start recording audio, then I show the picture, then a blank canvas and then I stop recording. As the sound plugins are immediately before and after the two canvases, I would expect the duration of the sound file to be equal to the sum of their two durations (in my case, 1500 ms and 1000 ms respectively = 2500 ms).
Instead, taking a random sample of recordings (n=185), I find that the duration can either be 2380.95 ms (74.59% of the times) or 2494.33 ms (25.41% of the times). I only find these two values.
I've taken a look at previous discussions
[Record Audio and Response time from voicekey input] (http://forum.cogsci.nl/index.php?p=/discussion/1268/solved-record-audio-and-response-time-from-voicekey-input/pl)
and this one [Using "soundrecording" to control the duration of a sketchpad object] (http://forum.cogsci.nl/index.php?p=/discussion/1655/solved-using-soundrecording-to-control-the-duration-of-a-sketchpad-object/p)
and at the main website but I can't figure out how to:
- either have a costant timing
- or lock the recording onto something.
I have checked whether the type of trial (picture and type of pseudorandomization) affect this difference but I haven't found eny effect. I am using Pygame backend.
Would you have any suggestion?
Thank you very much
harold
Comments
(Asking a question is a key strategy for debugging)
I just counted the number of samples per recording. They are either 110000 or 105000 (corresponding to the above durations).
So I guess this problem is similar to this
check your timing and I should choose a duration which is a multiple of a stable sample numbers.
Has any body tried before?
(Thanks for hosting this aloud thinking so far)
harold
Hi Harold,
I think you were right. Did it solve the problem?
Cheers,
Josh