Thread: Changing capture input in VSA?

Tweet

I'm trying to program our skull in VSA. The audio file i' using already has background music in it, so I can't use the Wave Motion Analasys. I'm trying to use the joystick to capture events, but it's incredibly hard because I have to hold in the trigger, and then move the joystick side to side to make the mouth move. Is there any way I can just use the trigger to capture, so I can just "click" when the mouth opens?

crazybob, is it not possible to just manually click and drag the events for the jaw servo and then manually set the values for each of those events? I know this would take a lot longer, but it may be an option.

Yeah... if all else fails that would be an option. The problem is that the speech in the track i'm using is very fast, so it's damn near impossible to do manually. I emailed Jon and he suggected playing with the tracks min/max settings to make the joystick super sensitive, but I don't know if this would be a good Idea wil a Skulltronix? Would that change the servo limits at all? I've even thought about using Vixen so I can "tap out" the words first, but vixen doesn't have the servo limits built in.

Or do what HalloweenBob does (There seems to be a plethora of Bobs on here...) He records himself doing the lines along with the song, then does the wave analysis on his track. he doesn't even have to sing on key (Although I'm sure he does) or even do more than say "la" for every syllable. Then copy the new track to the main routine.

Yeah...What he said...

Are you using a custom track that you made yourself or a song? If so, what song?

The easiest way to do it the way I do is to burn your original track on a CD.

Then put on a set of headphones and listen to the CD. It's best if you lift one side slightly off your ear so that you can hear your own voice as well.

Then, just speak or sing in sync with the music recording it into your computer at the same time.

It may take several takes until you get it right on, but it will save you TONS of time in the long run.

Once you have your new track, open it up in some Multi-channel audio software like Audicy (Free) or the one I use, Adobe Audition.

Then you can easily sync up your new track to the original. You can edit it so that every syllable is exactly matched, then Lower the volume on the original track to zero, mix the tracks together and save as a wav file. You will end up with a perfectly synced version without any music to run waveform analysis on.

Always work with wav files. Why, you ask? I'll tell you...

Even though they take up more space and VSA can use mp3's you will have syncronization problems if you go to mp3s.

This is because each time you create an mp3 and save it, there is a small but noticable amount of blank space inserted at the beginning and the end of the file. This is non-audio data that includes ID3 tag info and compression information, but you don't really need to know that. The problem is, that it will no longer start your words at exactly the same number of seconds in that it did before.

Soooo.....

once you settle on your original file, the one you will be hearing in your routine, convert it to a wav file, and make all voice separation tracks as wav files after that. This will keep it all in sync.

So, is that more than you ever wanted to know about the process?

By the way....I usually do NOT sing in key! If you hear me singing in key in any of my routines, it was probably take #45 or so before I got it right.

Someday I might post some of my voice separation tracks just so everyone can get a good laugh!

I didn't even think of that! That should do the trick! thanks!