I just used Audacity to edit a 16-track session, and I thought I’d share what I learned along the way.
I’d done multi-track sessions before, sometimes even more than sixteen tracks simultaneously, but those were sessions where I was layering on different instruments or sound effects to the base tracks I was starting with. That’s a very different process.
In this case, I was given a live concert, in sixteen separate WAV files. Each file was an hour long. They had to stay in sync with each other throughout the entire concert.
1. Bass drum mic
2. Snare drum mic
3. Left overhead drum mic
4. Right overhead drum mic
5. Bass A feed
6. Guitar A feed
7. Bass B feed
8. Guitar B feed
9. Keyboard A feed, left
10. Keyboard A feed, right
11. Keyboard B feed , left
12. Keyboard B feed, right
13. Saxophone mic
14. Flute mic
15. [empty]
16. Vocal mic
I say 16 tracks, but really it was 32 tracks because each of these files was stereo. Since the tracks labeled as “feeds” had no ambient sound, just the sounds of the direct-feed instruments, it would be difficult to synchronize them by ear. The tracks labeled “mics” all had bleed-through from the rest of the band on-stage.
To begin with, I took two files－an overhead drum mic, and a keyboard feed－and put them up onscreen together. I was able to verify (YES!) that they started in sync and stayed in sync throughout the entire concert. That’s one major headache averted.
Next I took the four files (eight tracks) of drums and decided to do a stereo premix, to make them easier to handle. It’s unlikely the drum sound will change during a concert, or need panning or spot EQ. All they might need is volume balancing, and that would be together in unison. I found the left and right overhead mics could be cut down to mono tracks, since the opposing channel was just live bleed-through of what the other mic was capturing. The stereo bass and snare drum mics each had a lot more ambience, so I left them in stereo to give the drums room presence. I EQed the bass drum to make it stand out (+10dB at 32 Hz, -3dB at 64 Hz, -5dB at 128 Hz, flat the rest of the way up). This gave the drum a satisfying ‘kick’ without creeping into the range of the bass guitars.
The snare mic was usable pretty much as-is (good engineering!) I rolled off the bass on the overhead mics, to give me better control later over the bass drum and bass guitars.
I laid these six WAV files next to each other in Audacity (bass drum, snare, left overhead, right overhead) and fiddled until they were nicely balanced. Because the overhead mics each caught a little bit of what the other was capturing, I found I could pan them hard-left and hard-right without making the drums sound ten feet wide. I saved this drum premix to a stereo file.
Next I thought I should do a premix of the keyboards. That’s another eight tracks I could bounce down to two. Both Iago and Ricardo had multiple keyboards, and their stereo feeds were already panned and balanced. I wanted to maintain their individual settings, but still give them their own space on-stage, so I panned one from hard-left to 75% right, and the other hard-right to 75% left. This sounded fine on the premix, but as they switched off between instruments or took solos, their individual instrument volumes would need some adjustment later. I therefore saved this premix as an Audacity project file (.aup3) which preserved the eight tracks separately for later mixing.
Next I did the same thing with the horn files. Again Pablo and Dubi traded off on flute, tenor sax, soprano sax and clarinet so I needed individual control of these instruments later.
Finally, I looked at the direct feed files. The guitar and bass files were the same left & right, so I could treat them as mono files. The vocal mic picked up the band around him, but again there was no advantage to preserving it as a stereo file. The drum overheads picked up plenty of band ambience.
Now my 32 tracks were down to a more workable eight tracks: drums, keys, horns, two basses, two guitars and vocals.
I laid these eight tracks alongside each other in Audacity, and verified again that they stayed in sync (note to self — do not delete anything!) I saved this as a “master premix” AUP3 file. I discovered of course that when Alberto played bass his guitar channel was silent, and when he played guitar the bass was silent. Good, that makes mixing even easier, six tracks instead of eight.
With all eight channels up on screen of the entire one-hour concert, I simply highlighted each song, one-by-one, pulling the cursor down through all eight files. Then a copy-and-paste to a new window. All eight channels came across, still in sync. I started a little bit before the song started, and a little bit after the song ended, for later trimming.
Once I had a satisfactory 6-channel mix I exported it out to a stereo file for each song, then opened that song file and normalized it to 0dB (I discovered very early that when combining 8 tracks you need to stay well below 0dB on each of them). Occasionally the additive sum of bass & drums created excessive peaks, which I had to scale back a little with Clip Fix. Finally I trimmed and faded the beginnings and endings, which is a lot easier to do on one stereo file than eight sub-files.
There were a couple of small hitches in the process.
In one track, “In The Land of Grey and Pink,” Iago switched back and forth between a Rhodes piano and an acoustic piano while Ricardo held down the organ chords. I found his acoustic piano sound was way too soft compared to everything else going on around him. In certain parts of the song, the piano really needs to stand out. Since his mix was already pre-balanced, whenever I boosted the volume of the piano it changed the entire keyboard balance. By using Channel Mixer, I could separate the left from the right in the premix, which gave greater control over both. The organ solo in the middle could be boosted to center stage, because Iago laid out during it.
When I shared a mix of this track with Alberto, he said “There is a mistake. The keyboards are a half second too early, so they come in too fast.” I went back to the original keyboard premixes and retarded them (moved them to the right) just enough to line up the organ chord changes with the rest of the band.
However, after doing this I got some echoes in the piano. The overhead mics had picked up the keyboards, and the ambient mics were no longer in sync with the direct feeds. What happened? I’m guessing the keyboardists wore earpieces, and there must have been a slight latency delay in their on-stage feed. Maybe? At any rate, it’s only noticeable when the piano plays (which is not often), and it’s very subtle (being ambient feeds only). You can’t tell at all on the organ chords. Unless you’re wearing headphones and listening super-critically, I don’t think anyone will notice.
Put all ten songs in order, adjusted the beginnings and endings for consistent volume, and wrapped the whole concert up with a bow. This was a fun project, and Audacity more than lived up to the expectations.