Hi, To further the discussion from a technique standpoint, there are quite a few ways to accomplish remote voice tracking.
One system that I have worked with uses a web interface (for the client/user-end) and on the server end it'll send the client information from a requested block of time in a requested log. When you want to voice track you pick the VT you want to record in the log, it'll compress the last 20 (or thereabouts) seconds of the track prior to the voice track and the first 20 seconds of the track after the voice track. The web client plays the tail, allows you to record your voice track, and plays the start of the next track. You then get to adjust where your voice track starts / stops and listen to it, then when you hit "Save" it'll compress what you recorded and upload it (along with the start/stop data) to the server, which then puts that VT into the log. Repeat for the next VT. Essentially this is what Rivendell's current voice tracker is doing, the main difference is that it sends the log and audio data back and forth across the web using compressed audio (for transfer speed). Given Rivendell's current Web API, a bunch of the server-end functionality needed to do something like this is already present. We can import and export audio via the web API. In the Export we can specify the start / end point that we want to export, so it should be possible to only export the length of audio we wish to play back at the client end for the purpose of voicetracking. We can manipulate carts, cuts, trim audio, and list logs and such. On the server / web api end the main items that I see that we're missing are the ability to insert and delete a voicetrack from a log, and the ability to set / change the segue points for how the voicetrack sounds once it is recorded. It could also be helpful to have more log-based commands in the web API (the ability to move stuff around in the log, add / remove tracks from a log, etc) but for a bare-bones remote voicetracker advanced log editing would not be absolutely necessary to get things working. Of course the other stuff that we're missing is the code to pull it all together as a web-app for what an end-user would see. Ideally if we can use some of the audio playback and record features found in HTML 5 then a client could be developed which would be extremely cross-platform - as long as you use a browser supporting HTML 5 then you'd be good to go. Anyways, just something to think about. > > No Cowboy > > Our system is rivendell, we just engineered with lots of help in the > technical > side a way to do remote voicetracking > > the convo is talking about a way to make it better > > I think > > (sorry im writing on pain meds today everyone) > > > > _______________________________________________ > Rivendell-dev mailing list > Rivendell-dev@lists.rivendellaudio.org > http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev _______________________________________________ Rivendell-dev mailing list Rivendell-dev@lists.rivendellaudio.org http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev
