I’ve been thinking about how to write relevant blog posts during the COVID pandemic. I’m saving some of my rehearsal and gesture stuff for whenever we decide what normal looks like again. That being said, I’ve been building a lot of virtual choir videos since March, and lately I’ve been working with people to help them learn to build their own. So it makes sense for me to write about how to create a virtual choir. Today I’m going to just describe my process and a few of the lessons I’ve learned along the way in the hopes it might be helpful. In the future I’ll plan to go into more detail.
Software:
I use two programs when I make a virtual choir video: Mixcraft Pro to edit audio, and Magix Movie Edit for the video. I’m sure other software works just as well. This is what I have.
Singers don’t need anything more than a relatively up-to-date smartphone.
Singer Materials:
Working in recording studios in the Midwest, I learned very early on that higher quality on the front end leads to exponentially less work on the back. In other words, getting good product from the singers is crucial to saving time and sanity. In order to help with this, I create the materials that my singers will use. If it’s my own arrangement or composition, I make sure the score is clean, easy to read, and VERY IMPORTANT FOR RECORDING has page turns in places where they can either be done quietly, or where the singer isn’t singing so I can edit them out.
The next thing the singers need is something to sing along with. If the piece is accompanied, I record the piano plus one voice on each vocal part. I then export a different recording for each voice part to sing along with that has their part prominent and the other parts in the background. That way if you’re a tenor, you hear the tenor part prominently, but also understand your context in the music. If the piece is unaccompanied, I send either the piano playing parts or the voice parts singing with the parts, but rarely both. If the singers sing on the first beat of the piece, it is crucial to include a count-in. If the piece has a lot of rubato or needs lots of subdivision, it can be useful to include a click track.
The final material to send to singers is detailed instructions. For example, I’ve found the video editing is much quicker and cleaner if everyone holds their cameras in landscape rather than portrait mode. This keeps all the video ratios the same for all singers. It can also be useful to remind the singers to stay engaged even when they’re not singing, to perform the piece rather than just sing it, and to be aware of their attire and background. Perhaps the most important thing to remind the singers of is that they will never sound on their video the way they sound in choir. The vast majority of them will listen to their recordings and think they will sound terrible. There are a lot of reasons for this that we won’t go into here, but you can lose a lot of participation if people get discouraged because they aren’t prepared for how different their recordings will sound.
Process:
I have singers send their videos to me via file sharing programs (iCloud, Dropbox, Google Drive, etc). If they email or text them, you either won’t receive them, or they will be subjected to additional layers of compression that can make them impossible to work with. (I recently did a video for a group in Charleston, SC, where the director collected the videos and shared them with me. I couldn’t understand what was going on with one of the videos until I talked to the director who told me that singer had texted the file. Once we got the file via Dropbox, everything worked perfectly.)
Once I have the videos, I start in my video editing software. I bring in my audio file that I sent out to use as master audio, and then bring in the first video. I’ve started renaming the videos by voice part first (sop Pat, tenor Alex, etc). I bring in all of the bass parts, then all of the tenors, etc. That way they’re easy to find and all together if something in the music calls for video structure that highlights a certain voice part. Here’s the process:
- I bring in one video file
- I separate the vide and audio onto two tracks
- I mute the audio that came in with the video
- I line up the video so it matches the master audio
- I mute the master audio and unmute video audio
- I export the audio to a .wav file named the same way as the video (alto Alex , tenor Pat, etc)
- I delete the video audio
- I make sure the video is locked in its place on the timeline and muted
- I make a new track for the next video, and repeat until I have all my video imported
Once the video is in, I turn to the audio editing. I make subgroups for each voice part, and import by voice part. This is a place where some time on the front end will pay dividends on the back end. One thing I hear is that people don’t want their product to sound over-produced. When you think about the fact that everyone has recorded their part with a different camera/mic in a different room, you quickly realize that a lot of production needs to go into getting the sound to not sound overproduced. Spend the time now. It will save lots of time in the long run. Compressors and EQs will be your friends as long as you use them to counter the distortions to the sound that are already there. A little reverb on the whole track, or on the subgroups, can help too, but don’t overdo it! A little goes a long way. Panning voices both in your subgroups and panning the subgroups can also help to recreate the sonic space of a choir.
Once the audio is mixed and mastered, I import it into the video software and replace the old master audio. This becomes my new master audio. Now it’s time to edit video. I use the score to get an idea of who I want to see where. Then I go through and separate my track into sections, deleting the video of any parts I don’t want to see in any specific section. Once this is done, I go section by section and apply my layout ideas. Once layout is done, I create crossfades, build titles and credits, and apply any static backgrounds. Then the video is done… sort of. I export it and watch it a few times. If possible, I check every face all the way through to make sure nothing has come out of alignment or that I haven’t missed a fade. Sometimes these little tweaks take a few mixdowns, but it’s worth it to have a product that is clean and professional looking.
And that’s it! You can see some of my virtual choir work on my MEDIA page. I’ll go into more detail on some of my process down the road, but please contact me or write in the comments if you have any specific questions or anything you want me to write about in more detail. Stay safe and stay sane!
Paul H. Thompson 