Blog: Roundabout synchronize of AV for multi cam edits: Premiere and Audition
Premiere... ...Audition... ...They do work together. However, Audition often works on the input for the beginning, or on the audio mix for the end of an edit. By that, I mean it usually creates a single mix to use in Premiere, or it allows you to clean up the audio from a video file, rendering the audio to a preview for use with your edit.
Here, I'm going to use it a little differently, and I'm going to make some notes for using other helpers like Plural eyes, or WOOWAV.
When you shoot constant video feeds with solid WAV output, you can get files that go beyond the WAV standard. If you use an audio app that records WAV with this ability, you'll have trouble using the audio in any other program. I have run into this a lot.
I like 96k audio, as it keeps the "Feel" of a room when shooting live performances. Later I drop it down to 48k, and the conversion is done only to the final output. Premiere, however, cannot open a WAV file larger than 4gb.
I save my movie files normally with uncompressed audio for the output stage. Then I save a proxy with a compressed format audio.
Note: Since I only care about audio at this point, I'm going to go ahead and just output an AAC file at 48k 16bit (audio only) from adobe media encoder for each video in my multi cam project.
When I want to synchronize audio, I have several options. Manually editing, I'd need to adjust by frames or samples, which can be tedious and inaccurate, but close counts. If I find that most of the job is short, I might do this manually to boost labor, and make the job worth it; or I might just check the audio waveform for matching points and line it up if it's an obvious match.
However, I usually prefer to let the computer do it when the project is unbearably long. 2 methods work here.
1. double sync
2. single sync
The second is the method most know and love. In CC, it's automatic. In cs6 you can use Plural eyes, or woowav plugins, but there are problems when the audio is too accurate (highest bit representation or sample rate). Dropping the sample rate will fix this, but you can lose quality, and you'll have to do more work anyway to bring that back.
The first method is one I've come to know and love, as it works with almost every workflow, and it's often the ultimate end of the other method anyway.
You output audio only files at 48k with low bit width, in wav format. Open premiere, and prepare your multi cam setup sequence using one of your videos at the desired resolution, then add the audio on it's own tracks. Remove the video files from the sequence, then edit the sequence in audition (or send it to your plugin). For Audition, it's difficult to match with too long of a playback. SO... ...You should limit this. You can shorten the files to a few minutes, where there is plenty of audio crossover. Now sync the speech in audition, and send it back as stems to premiere using the export menu in audition. This will send the audio back into an xml sequence as separate tracks, then you just insert tracks in between those, and line up the clips properly, and you can select them both, click synchronize, clip start, then check each with the different audio tracks. At this point you can delete the audio tracks you stemmed, and nest the sequence.
If you have a plugin, you can use the full WAV files at lower sample and bit width. It will lose quality, but will have more points to use for sync (though you technically need only one).
NOTE: This only works with audition if the videos aren't far apart for sync (within 2-4 minutes) and if the spikes are similar (so you do need to find a close match at the front).
If you have plugins, you can do this with full WAV from the sequence, and sync them up if the cameras are farther apart.
Just so you know, I prefer Audition, but I have used plugins. Audition is faster, but plugins copy the WAV info, and then line it up.