Vicarious recording fun for yez

[Bob Womack]

You know what they say about the shoemakers kids - they have no shoes. Getting up the energy to record my own stuff after a day or week of recording is tough. But I've gotta do it.

Bob

[Bob Womack]

Tuesday was consolidation and guitar overdubs day. By consolidation I mean organizing the tracks to give me more monitor control for the next round of overdubs. I also balanced the string section: Because each of the instruments is on a separate track and I have time, I can balance things out a little better than I could during the tracking session. I also took time to experiment with panning and the overall stereo image. I tried a trick this time: I reversed the stereo image of every other pass of the quartet (there are four passes), placing the violin and cello on opposite sides so that you don't end up with all the highs on one side and all the lows on the other. It gave the impression of a much larger string section, frankly. Were this classical music, I wouldn't do it. But it is pop!

Next I did electric guitar overdubs. The song has some places where there is a build in the tension and I felt like they needed some strength. Last week I suggested to the producer that we could ratchet things up with a crunchy rhythm guitar and she like the idea. She asked, "Who could play the part?" I said, "Oh, I'm the cheapest and easiest to get guitarist you've got." She said, "Go for it!" Knowing I was going for a full sound, I chose my old Les Paul as my weapon. I took it in one hand and slung my heavily-tweaked POD2 over my other shoulder with a bag of cords. I really had only a loose idea of what do play as I walked in. I arranged a part that replicated the voicing of the strings in one place and comped chords while following the bass guitar's and drum's rhythm in two others. Getting some guts took some doing because the song was in the very pianistic key of F minor, but I was able to do the hand contortions to make the parts fit nicely. I doubled each part so that the sound could be panned outboard for some stereo interest. I had a little fun playing and today my producer gave my interpretation the thumbs-up. Success!

Mid-session: I pulled my Swiss Army knife out of the change
pocket in my jeans to protect the back of the guitar.

My next post will cover today's choir recording.

Bob

[Bob Womack]

So today was choir day. We were slated for a four-voice section to be built up to a gospel choir but we cut it to three, eliminating the bass, because we are getting a fair amount of bass build-up with the strings already. Two hours before the session I turned on the Avalon VT-737 preamps and let them heat up. I get a little anxious every time I work with one of our preamps because it wakes up with a little bit of bacon frying sound in the background. Thank goodness by session time, as usual, it was nice and clean again. Apparently it just likes to marinate a little.

The group intended to work as a section so I set up for the trio with an ORTF (Office de Radiodiffusion Television Française) stereo array on a large studio stand. The official definition of the ORTF system is a pair of crossed cardioid mics set at a 110' angle and 17cm (approximately 7") apart. For my ORTF, I used our matched AKG C451BSTs. Because I wasn't going to close mic, I ran them at about eye level and pointed at the mouths. They were about two feet away from the singers. I ran them flat and with only a tiny bit (say 3db) of compression.

Our producer/arranger/choirmaster was also the soprano of the group. I arrayed them, left to right, alto, tenor, soprano, to put tenor in the center because he would least conflict with the lead vocal, who will be a soprano. The ORTF array nicely reproduced acoustically what I saw from the control room window. The group ran down the parts in the control room for about twenty minutes, polishing timing and blending, then went into the studio. There, we got the headphones going for them as they continued to practice. I made a few positional changes in both the mic and the singers, and it was time to hit the red button.

It took probably ten takes to come up with a master and then we punched-in to the last phrase to help them get the intonation perfect. Once we were satisfied, they doubled their parts on a second pass. Interestingly, they asked to punch-in on the same passage. Then, to fill out our simulated choir, we ran two more passes and called it done. I invited them back to the control room and quickly reversed the stereo left-right imaging on two tracks before they arrived. After a quick review, the choir was dismissed, the producer left, and I began consolidating the tracks and setting up for tomorrow's lead vocal recording session.

Tomorrow we record the lead vocals with a gal who recently made fifth place in American idol. To give you an idea of how little I follow American Idol, I don't even know who she is. That will make tomorrow's session a fun surprise, eh?

Bob

[sir_strumalot]

Bob, you da man!

[Bob Womack]

Hehehe... You made me laugh!

Today was lead vocal overdub day. I chose to mic the female singer with a Neumann TLM-170 large diaphragm condenser run through a Avalon VT-737 preamp. I ran the EQ flat and used at max only about 2-3 db of tube compression to catch any overs that might threaten the recording. Franky, I ran the levels conservatively enough that I didn't need the compressor. Thankfully, the system is quiet enough to support that sort of operation. Our singer did a fine job and we took about two hours to finish up the soprano lead, stacked alto and soprano harmonies, and special effects of various types.

The biggest challenge of the session was the fact that members of the entourage brought their 18 month to 3 year old children into the control room where they ran around, banged various items, shook a maraca shaker, screamed and cried. I heard, "No, honey, don't touch the knobs," over my shoulder more than once, with results later. Once we finished, the entourage left, and I went back to evaluate the tracks in a silent control room, I discovered that the background vocals were quite dark-sounding. When you are recording them you run them behind the the lead vocal, where much of the detail is lost. If there is noise in the control room, it is virtually impossible hear background information. When I heard that the tracks were dark, I spun around and looked at the mic's preamp and discovered that one of the kids had grabbed the high-end EQ and had spun it down pretty far. Thank goodness the Avalon EQ has a clean, predictable high-end EQ that could be reproduced and then inverted on a plug-in to fix the problem.

Ah, the other side of recording!

Bob

[Bob Womack]

Today, we recorded some of the final bits and bobs of vocal work for this piece. In this case, it consisted of a verse of rap and some a rap overlay for the final chorus, as well as some vocal elements for a montage for the end of song. The gentleman rapper with whom we worked was a professional and eager to help us get the parts we needed. He was a pleasure to work with.

We discussed the beginning of the song and I put forth the idea that we create a "soundscape" as a lead in, something along an Alan Parsons/Pink Floyd-ish construction to set the stage for this song, as it is a very serious one that tells a story. The idea went over well and I'll be starting that next week.

Meanwhile, I'll be bringing my guitar and rig back to the studio with my EBow to add some special effects. More in a few days.

Bob

[RogerC]

Originally Posted by Bob Womack

Thank goodness the Avalon EQ has a clean, predictable high-end EQ that could be reproduced and then inverted on a plug-in to fix the problem.

Bob

**head in hands, shaking it back and forth** I have so stinkin' much to learn. where's the translate button for that post?

[Bob Womack]

Originally Posted by RogerC

**head in hands, shaking it back and forth** I have so stinkin' much to learn. where's the translate button for that post?

Lemme see: The high-end EQ on the Avalon Preamp/channel strip is a shelf EQ, much like the shelf on the right in this pic:

The kid turned down the boost/cut knob and took out a generous amount of high-end gain, making the actual response resemble the lower curve on the right above. I knew the EQ's nominal frequency selection (10k hz). I started by inserting a plug-in EQ into the channel and selecting a shelving EQ. Then I set the nominal frequency to 10khz and came up with a reasonable curve (or "Q") between the turn point (where the frequency response is 0) and the nominal frequency to reproduce that of that Avalon. Finally, by A/B-ing the background tracks with the undamaged lead track, I ADDED gain on the shelving EQ to make up for the subtracted gain on the Avalon's shelving EQ until the damaged and undamaged signals matched to my ears. Essentially, I created the upper right curve in the plug-in to counteract the lower right curve the kids put on the Avalon and ended up with neutral influence. Thankfully, my original signal was clean enough that there wasn't a significant amount of noise added by the high-end boost. Voile'! Band-aid.

Does that make it clearer? It isn't a perfect or purist way to handle a signal, but it beats bringing the singer back in. I really don't think anyone but me (and now you) will know that anything was damaged by the kids in the first place.

Bob

[sir_strumalot]

Bob, you're going to have to change your profile location from "Between Clever and Stupid" to "Between Clever and Yes I Can Save Your Butt".

Amazing work, really. Now, where did that "No Children Allowed in Control Room" sign go?

Totally OT: You don't happen to follow the AV Rant podcast by any chance do you?

[Bob Womack]

Thanks, Sir Strumalot! No, I haven't followed it. What's it about?

Bob

By the way, this is the sort of thing you learn when you are doing Audio Post for Video and Film. You constantly have sources coming in that have an overbearing capsule resonance or a group of frequency peaks caused by proximity to a wall or something like that and you have to learned to rapidly identify and eliminate them to make the track sound natural. I bet most music recordists would consider APVF purgatory but between the skill sets developed, the sheer workload, and the fact that APVF is accomplished under extremely tight deadlines when compared to music, you end up well-equipped for music.

[sir_strumalot]

Originally Posted by Bob Womack

Thanks, Sir Strumalot! No, I haven't followed it. What's it about?

Just call me Jeff.
I was playing a hunch, as it's a podcast from audioholics site. A podcast or two ago they mentioned an email from a "Womack". I was thinking, "Cool, small world!", but apparently a long lost relative instead. Oh well, sorry to bother.

By the way, this is the sort of thing you learn when you are doing Audio Post for Video and Film. You constantly have sources coming in that have an overbearing capsule resonance or a group of frequency peaks caused by proximity to a wall or something like that and you have to learned to rapidly identify and eliminate them to make the track sound natural. I bet most music recordists would consider APVF purgatory but between the skill sets developed, the sheer workload, and the fact that APVF is accomplished under extremely tight deadlines when compared to music, you end up well-equipped for music.

Speed and skill! I hope you make a great wage, because man, you earn it!

[RogerC]

Originally Posted by Bob Womack

Lemme see: The high-end EQ on the Avalon Preamp/channel strip is a shelf EQ, much like the shelf on the right in this pic:

The kid turned down the boost/cut knob and took out a generous amount of high-end gain, making the actual response resemble the lower curve on the right above. I knew the EQ's nominal frequency selection (10k hz). I started by inserting a plug-in EQ into the channel and selecting a shelving EQ. Then I set the nominal frequency to 10khz and came up with a reasonable curve (or "Q") between the turn point (where the frequency response is 0) and the nominal frequency to reproduce that of that Avalon. Finally, by A/B-ing the background tracks with the undamaged lead track, I ADDED gain on the shelving EQ to make up for the subtracted gain on the Avalon's shelving EQ until the damaged and undamaged signals matched to my ears. Essentially, I created the upper right curve in the plug-in to counteract the lower right curve the kids put on the Avalon and ended up with neutral influence. Thankfully, my original signal was clean enough that there wasn't a significant amount of noise added by the high-end boost. Voile'! Band-aid.

Does that make it clearer? It isn't a perfect or purist way to handle a signal, but it beats bringing the singer back in. I really don't think anyone but me (and now you) will know that anything was damaged by the kids in the first place.

Bob

That does clear it up somewhat. Thanks, Bob

I had no idea what kind of demon hobby I was getting myself into when I started recording. I just wanted to plug in some mics and play, but noooooo. I had to find something that would require me to study a whole new language and use a part of my brain that I swore I'd never touch again after college

[Bob Womack]

It was RED GUITAR DAY at the studio yesterday. Actually, I did some more guitar overdubs, this time involving the EBow, to spice things up a bit. Lot's 'o fun!

Bob

[Bob Womack]

Today was vocal treatment day. We pitch corrected and time corrected the lead and background vocals. Now, we didn't use an auto-correct program this time because 1) our lead vocalista uses quite a bit off "body English," aka sliding on and off pitches, and we didn't want to negate that, and 2) she was pretty well on with only a few isolated items needing correction. In order to handle an isolated pitchy note, I'll go in and split the bad section into its own region and manually pitch correct that section only. If it is part of a held word (ie the word spans two pitch regions with a slur) I'll split out the bad section, pitch-correct it, and then cross-fade between it and the rest of the region. When done well, it can't be detected. Voile'!

And then there is time correction. Have you ever been a part of or watched a choral rehearsal? The conductor will direct the beginnings, middles, and ends of notes so that all the voices synchronize. Many modern vocalistas haven't had any formal training and don't know about this business. They are very expressive, but undisciplined, so when they sing their own background vocals, they simply aren't aware of the business of sync'ing up the parts. As a result, you end up with a host of un-synchronized entries, pitch changes, and exits. The goal is to sync them up with the lead vocal. You can especially hear it in the "s" and "t" sounds at the heads and tails of words. The good news is that background vocals can be edited pretty freely without artifacts. So, a typical word has an entry, mid-part, and exit. We sync up the head then the body and are late. As an alternative, you can cut off the head and move it to leave the body and tail in sync. Often nothing syncs, and you have to clip it into three chunks and time 'em up. Add cross-fades and you are good to go. Now to add complexity, "S" sounds can often be different lengths, so you have to adjust the lengths by cut, paste, and crossfade. By the way, to help things along, many BGV singers omit the final "S" so that they don't have time sync issues in the first place. During this phase of operations, the whole process is aided by temporarily panning your voices radically so that the bad guy can be identified geographically and the track identified.

So, after two hours of editing, we are 90% done pitch and time correcting the vocals and they are cleaning up pretty well. Once we finish, I'll be automating the levels to get a reasonably consistent performance level that can sit above the orchestra and rhythm sections.

Monday we're on to the last 10% of pitch and time and the we build an exit montage. Once these chores are done, I've got to try to convince my pianistic co-producer to allow a little lead guitar closure for the dénouement of the piece.

Cheers!

Bob

[Bob Womack]

For those of you who have followed my little session blog here, two weeks ago I created a montage of sounds for the head of the track, a street scene, as an emotional scene-setter for the song. I added in drone tones under the montage in the song's key via the Ebow. I used my Strat-ish guitar with its extra octave of overtones for one track to get a richer sound so that when I pitched it down an octave as a bass drone it was rich and busy. I had also created "birds", ie. seagull sounds, with the slide 'way up between the two humbuckers on the ES-335 for the end of the song. Last week I was on vacation so there was no progress on the song except in my head.

I've spent the last three days honing the mix with little automation tweaks and the final touches. One of the things I discovered was that the piano track was "thinkening" things up. I gently backed it off, allowing more detail in the strings and guitars and a little higher mix of instruments-to-voice.

I've also been working up to today's vocal overdub session. We are changing the delivery on the first few lines to ratchet back the emotional intensity a notch. We'll launch into that after lunch. After the overdubs, we integrate the new tracks into the mix, spend a couple of hours on the last polishing touches, and perhaps the song is done. As of right now, it's 95 tracks of joy. Woohoo!

Bob