I'm trying to decide the best way to record dialogue between two puppets - they are both on screen simultaneously and need to have proper timing between voice and movements. My question is how do I record audio for multiple characters on screen and get the timing right?
What I normally end up doing is recording the audio track for a character, do any audio editing (e.g. noise reduction) in Adobe Audition, then chop it into little audio clips. You can then import the audio clips into Character Animator, slide them around individually in the timeline etc. I just found this easy to understand.
(You can do simple cropping etc inside Character Animator as well - there are multiple ways to achieve what you are after. I just do it in Adobe Audition because it seems easier to understand. I have an audio clip when something is being said.)
Once you have them placed correctly, I then use “compute lipsync from audio” inside CH to get the visemes. You can do this per audio clip as well. You can slide around the audio file and visemes together later if you need to.
I frequently do multiple passes over a project. Throw it together, put the audio clips in, don’t both animating the characters (just drop them in a scene), play the scene - does it “feel right”? What areas need to be longer or shorter? I then go through making passes to add in more details, remembering to look at the global feel and local feel in balance. Sometimes I find a scene is just too long and slow so I need to cut back. You can spend a lot of time doing the animation, so getting it right before you start (the right balance) can save you a lot of time.
Oh, you have a question of how to create the audio recording. I don’t have a professional sound studio - I find a quiet room with lots of soft surfaces (curtains, carpet on floor etc) to reduce background echo from the room. Having the mic close to your mouth helps too. Hard surfaces around you are bad (create echos) which are almost impossible to remove. I then bring the audio records to Adobe Audition and use the noise reduction features etc there to clean up the audio files. I also do pitch bends to make it sound more “cartoony” for the voices. Then I load the final files into CH. You can record directly in CH - I just don’t because I care about the sound quality.
I remember OKSamaurai had a video on using Adobe Audition as a multi-channel mixer. Maybe he got all the timing right there, got the sound levels consistent, then exported One channel at A time and loaded up the full scene audio file because he had it all fixed in Audition. That works too I guess and you end up with less audio files. You can still splice the files in CH to adjust timings to fit in with animation as well.
As always thank you Alan. I will record it all at once and then chop(edit) it into proper segments - I didn't think of that as an option.