Re: [WnCC] Remote collaboration, virtual presence, etc

[Prashant Sohani]

Haha.. interesting problem here. I'm afraid, however, that trying to realtime sync music is not exempt from delay, and no workaround is possible. (Do this: Call up a friend, and try to sing a song together on phone. No matter how minimal the latency, it's enough to ruin the experience :/ )

So.. to me the question really seems to be: how less can we get this latency to be, hardware-wise? Not to forget: this is not simply a Counter Strike positional update packet; it's a full-blown audio stream.

On 23 February 2013 05:07, Vishal Patwa <vish...@gmail.com> wrote:

Guys I am working on a project related to remote collaboration. Looking for new ways to collaborate virtually for composing music/ jamming together preferably in realtime. If you guys have some ideas please do share them here. I am looking for ideas that enrich the present experience of collaboration or move beyond skype/NINJAM kind of thing. 

Vishal Patwa

Interaction Design

IDC 

--
--
The website for the club is http://stab-iitb.org/wncc
To post to this group, send email to wncc...@googlegroups.com
 
---
You received this message because you are subscribed to the Google Groups "Web and Coding Club IIT Bombay" group.
To unsubscribe from this group and stop receiving emails from it, send an email to wncc_iitb+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Prashant Sohani]

But having accepted the theoretical limitations, let's say we go for the NINJAM solution: to extend the latency to a fixed interval.

In that, one possible improvement could be to allow the interval to be driven by any one voice (likely to be the drumbeat). That way, even if the song demands sudden tempo changes etc, it'll all continue to remain in sync after the one time measure of mismatch.

(Here I'm assuming that some beat detection algorithm is already being used by NINJAM, so that it can adaptively track the drum voice. If not, there's another area for improvement)

Another: Extracting audio pitch from layered music may be a fair challenge on its own; but now with the data already separated as voices (and possibly also specified by voice types as extra metadata from the users), one may possibly look into the potential for automated transcription.. (even might make use of the assumption that there is some correlation between the music being played by the separate voices)

[Twishmay Shankar]

You've got to check this out. This product has been making some noises lately. 

http://www.businessinsider.com/chromatik-ipad-music-app-investors-2012-12

Best Regards, 

Twishmay Shankar

Mobile: +91 97697 47393