[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [AUDITORY] Online rhythm production experiments



Hi Henkjan and everybody,

My research group has developed a technology that has solved this problem and allowed us to collect reliable tapping data in an online setup. We’ve successfully collected large tapping datasets this way, and we believe that our method fully addresses the issues mentioned in this thread (low latency and jitter) while also being practical in terms of realistic online data collection. We plan to publish a preprint by the end of the year and therefore make the details of the technology accessible to everyone soon. If you are interested in using the technology earlier, please contact me.

Very best,
Nori Jacoby


Max Planck Group Leader, “Computational Auditory Perception”

Max Planck Institute for Empirical Aesthetics

Grüneburgweg 14, 60322 Frankfurt am Main, Germany

nori.jacoby@xxxxxxxxx +49 69 8300479-820




From: AUDITORY - Research in Auditory Perception <AUDITORY@xxxxxxxxxxxxxxx> on behalf of Samuel Mehr <sam@xxxxxxxxxxxxxxx>
Sent: Monday, October 19, 2020 1:30 PM
To: AUDITORY@xxxxxxxxxxxxxxx
Subject: Re: [AUDITORY] Online rhythm production experiments
 
Hi Henkjan & list, we haven't heard of a solution for this, but have tried a few times to do synchronization experiments to little avail. My sense is that there is quite a bit of variability in latency across physical keyboards, browsers, internet connections, operating systems, etc etc, such that it makes what seems like a simple problem rather hard. I'd love to hear if anyone has a solution.

Sam

--
Samuel Mehr
Department of Psychology
Harvard University
Be a citizen scientist at themusiclab.org!


On Mon, Oct 19, 2020 at 7:12 AM Prof. dr Henkjan Honing <honing@xxxxxx> wrote:

We are looking for some advice from those that have been doing online rhythm production experiments in the last few years. 

Rhythm production experiments that are performed online – like typing on the spacebar 'in sync’ to a fragment of streamed music – is notoriously tricky due to all kinds of delays mostly due to the effects of scan rate, device drivers and OS scheduling at the client-side [1,2]; Delays / jitter in the order of 50-100 ms are not uncommon.

There are all kinds of workarounds available, like offline analysis of audio files or using estimation algorithms [3], but it might well be that some of you have solved some of these issues in a more direct way. For instance, a piece of software at the client-side that handles the timing of both the streaming and recording the taps. 

If you have experience on these issues, we would appreciate if you would email us your tips, solutions, etc.

Henkjan Honing
Fleur Bouwer


.
University of Amsterdam
Faculty of Humanities 
Faculty of Science
.