Hello, I just wanted to say that, even though it can be simpler using sox, gstreamer can natively modify the playback speed using "seek events", and it also can alter the sound pitch by using an element surprisingly called "pitch".
Using gstreamer for the whole process may be trickier, but we at least keep everything into the same "realm", so to speak. Using external libraries probably means using temporary files for intermediate conversions. The suitability of one method or another depends highly on what exactly we try to accomplish. For instance, if we want to change the speed in real time, I don't see how we are going to link external libraries to gstreamer, except by writing a gstreamer plugin, with all the implications it has. I'm wondering how harder than modifying the source files would be modifying the player to allow changing the speed of a video, but of course this is a completely different discussion. Best regards Rubén 2012/3/22 Rajat Jain <[email protected]> > Hello Rüdiger, > > I understand what you want to say, but I am skeptic about the need of such > a tool in the player. The whole point of developing the application is to > make sure that the students don't have to watch the complete length > original videos, and if that is the case, the viewers don't know where they > have to point there playback time to (unless told by another viewer who has > watched the video). Generally we use the time bar to move back and forth in > video we have not seen earlier. > However, we can think of integrating some sort of mechanism which will > tell us the exact time of current frame in the original video, if the > student is somehow unable to understand the concept and wishes to see the > same topic explained in original video. > > Regards > > > On 22 March 2012 13:25, Ruediger Rolf <[email protected]> wrote: > >> Hi Chris, >> >> I just want to add an additional level of complexity to the discussion. >> If we want to add this feature to our engage player we have an interesting >> problem to keep consistency. The videos with a different playback speed >> have a different lenght from the original video. If I select a different >> speed I have to normalize all navigation elements. >> An example: if the video would be at 0:45 minutes in real time and I >> decide too speed the video up now to 1.5x I would have to jump to minute 30 >> in the faster video. If I click on the segment/chapter that would be at 1h >> I will need to jump to 0:40 in the video. >> So there needs to be some JavaScript hacking for the player too, to make >> sure that that everything acts like expected here. >> >> Rüdiger >> >> Am 21.03.2012 22:30, schrieb Christopher Brooks: >> >> Hi Rajat, >>> >>> I have been going through the project ideas of the *openCast*, and I >>>> came across the idea of *Variable Playback Speed**. *However I have >>>> some doubts regarding the project. Such as how are we going to reduce >>>> the time of the playback. Are we going to increase the speed of the >>>> playback (which I don't believe is the idea), or are we going to >>>> remove some part of video from the original file ? Can anyone help >>>> me. Or does anyone know how to contact the mentor. >>>> >>> There are two ways to do this, either increase the speed of playback, >>> or compress playback by removing uninteresting parts of the video >>> (called "video skims"). Either of these methods are interesting to me >>> as a feature, but require a different background. >>> >>> Increasing the speed of playback would require investigating gstreamer, >>> specifically how gstreamer can read and write to file descriptors in >>> order to hook it up with the SOX project to pitch shift audio so it >>> doesn't sound too "chipmonky". Fairly minimal java code would need to >>> be written for this integration, and the majority of code would need to >>> be gstreamer related with a little UI work to show alternate videos. >>> >>> Creating video skims is much more involved, but a much less naive >>> approach to speeding up playback. You can think of video skims as >>> similar to lossy compression for video, where duplicate frames or >>> frames of video that aren't important are removed. e.g. when an >>> instructor has written something on the board and hasn't said anything >>> because he is waiting for students to transcribe it, you can delete >>> some seconds video. You could imagine some parameters to creating skims >>> to determine how aggressive they might be (e.g. ignore one video feed >>> over another, ignore audio, etc.). >>> >>> Of course, both approaches could be used together as well, to create a >>> lossy high speed playback. >>> >>> I'm willing to mentor this project, but it does require a fair amount >>> of self study and self interest. It's not by any means a trivial >>> project, and would probably require hanging out in #gstreamer and >>> asking alot of questions. A background in gstreamer would be great. >>> Feel free to drop me a question if you have any more thoughts, >>> >>> Chris >>> >> >> >> -- >> >> ______________________________**__________________ >> Rüdiger Rolf, M.A. >> Universität Osnabrück - Zentrum virtUOS >> Heger-Tor-Wall 12, 49069 Osnabrück >> Telefon: (0541) 969-6511 - Fax: (0541) 969-16511 >> E-Mail: [email protected] >> Internet: www.virtuos.uni-osnabrueck.de >> >> >> ______________________________**_________________ >> Community mailing list >> [email protected] >> http://lists.opencastproject.**org/mailman/listinfo/community<http://lists.opencastproject.org/mailman/listinfo/community> >> >> >> To unsubscribe please email >> community-unsubscribe@**opencastproject.org<[email protected]> >> ______________________________**_________________ >> > > > _______________________________________________ > Community mailing list > [email protected] > http://lists.opencastproject.org/mailman/listinfo/community > > > To unsubscribe please email > [email protected] > _______________________________________________ >
_______________________________________________ Community mailing list [email protected] http://lists.opencastproject.org/mailman/listinfo/community To unsubscribe please email [email protected] _______________________________________________
