|
|
Capturing the Hurricane -- Don Hayden's TracksNovember 29, 2008 -- After a brief Thanksgiving pause, it's time to get back to the project. Having taken several shortcuts to get samples of the raw tracks out. I dumped those sessions, pulled fresh copies from the back-ups and started over. The standard procedure that I use for vocals didn't work for Don. He's got a deep bass voice that this particular vocal mic is extremely responsive to. If this was one of Cara's tracks, for instance, booming p's or b's can be eradicated by killing anything below 100 Hz and moving up, if necessary, until it disappears. Since her primary range is in the 1-4k area, this barely effects the quality of her vocal tracks, and if it seems to, it can be overcome by adding a little back in the 600-800 Hz range. Doing this with Don's voice sucks all the awesome out of it. This time, I also want to be more gentle with the compression, but at the same time, the fact that he was also playing guitar was one of several factors that effected his proximity to the mic, which wasn't real consistent. Additionally, the more you increase the compression with the vocal mic (I'm gonna have to find out the model number), you start picking up more room ambiance. This can be a good thing or a bad thing. As someone who began recording on open-reel tape decks, I think the most amazing thing about software like ProTools is simply the ability to see sound. During the recording process, you're trying to get the best possible signal without distortion. Sometimes, keeping the peaks under control without compressing the audio as it comes in, sacrifices the overall presence of the track. Here's one of Carl's guitar tracks. At left is the original, at right is after applying gain to bring the greatest peak to about -1 dB. Looking at it, there are about 5 spikes that are limiting the rest of the content. One of them is the ending. Carl obviously hit that note harder. So before I consider automate compression, I like to do some manual work first. I'll go in, find those specific peaks, drop them 1 or 2 dB at a time until they're in line with the rest of the neighboring audio. Then maybe another dB or two off of that fat section in the middle. For the acoustic guitar tracks, this may be all the compression we need.  I do the same to the vocal tracks, but I will run it through the default vocal levelizor compression, with the attack reduced to about 7 milliseconds. Even then, there are still going to be spikes. In the example below, a single hot frequency still dominates after compression. So I'll fix that manually, even before I apply gain. It'll add 3 dB to the rest of the vocal track.  Next we have a vocal track (in green) and a guitar track (red). The vocals have already been compressed, but the attack was slow this time. Bringing down those stray peaks and about four vocal phrases will seriously fatten up this track. The guitar track would benefit just by reducing three peaks. Spend a little more time and, without any automated compression, the midranges will improve, the highs will be a little crisper. 
To approach the problem of the boomy low end on the vocals, I picked a track, added an EQ plug-in and experimented. After playing around with the low-end, it seemed much more effective to use a notch at around 44 Hz. I'm not committing to it at this point, which is why I'm using a plug-in. The perfect setting is going to vary from song to song. But it's a good starting point. At this point, I take the headphones right back off. Then I go through every song in the batch and do the same thing -- gain and compression, add the EQ plug-in. Then it's a process of listening to the song over and over, fixing each flaw as I hear it, then listening again. After spending an inordinate amount of time on these songs, the first song ("Clementine") and the last song ("You Are My Sunshine") still have a lot of background noise in them and "Sunshine" has some odd artifacts as well. Also, Don did 8 songs and it adds up to less than 15 minutes. If we're going to do anything at all with this, we need another session and probably another 10-15 songs. This will give us a chance to fill up an album and get cleaner takes of the problem spots. But this finishes these off enough to move on to the the band tracks. Finally. |
Raw Tracks |
