Q It Up: Back in the analog days, voice tracks were often recorded direct to multi-track after the music bed was recorded on separate tracks, thus allowing the VO artist to hear the music in his/her headphones while reading the copy. This makes for an arguably much better VO performance for many reasons. Fast forward to today’s DAWs where setting up busses and consoles for this type of monitoring is not as simple as it was in the analog world when each individual track had its own fader on the console. The question is, do you monitor the music bed while recording voice tracks? If not, have you ever done so and why don’t you now? If you do, tell us about your setup and experience recording this way. Granted, too often we don’t even know what the music bed will be when doing voiceovers, so this question applies only to those times when you are the producer AND the VO artist, or when you’re the producer with the option to bring the VO artist to the studio after you’ve laid down the bed (or via ISDN, etc.). Add any other thoughts on the subject!
Denzil Lacey: A lot of the time I would play music for the voiceover using "sends" in Pro Tools. I think it's very important to get the right energy for a promo particularly. For our main station voiceovers, these are recorded via Skype and ISDN. Our Male Station Voiceover is based in the .K and we conduct it via Skype and he records his end and has my mic muted while recording which makes things very easy. The same goes for the Female Station Voice, she is Irish but living in London and we do the sessions via Skype in a rented studio.
Luckily here at FM104, we have great resources and the equipment is great, which makes it much easier to have the ability to do things like that. I have a Yamaha O1V96 Console which is routed via Digi002 (Retro!) which is then fed directly into Pro Tools where I have set up a lot of routing options for different studios and remote ISDN/Skype sessions. We have Axia desks in all studios apart from Production and I can record into Pro Tools using multitrack - which is great for doing live sessions too!
Jay Helmus, Jay Helmus Productions, Toronto, ON: Some VO talents find it distracting (probably because they're not used to it). But assuming it's possible with the routing, and the VO talent can handle it, voicing with the music is usually the better option from my experience. Hearing the energy of the track has a tendency to instantly put the VO in the sweet spot. Often I do it the other way around though, simply because I don't always know what music I'm going to use until after the spot is voiced.
Josh Goodman Voice-Overs: Great question, and if the music is known and available, I love to read to music. It helps create a movement and energy to the copy that dry v/o doesn’t have. AND, I believe it’s one of a few sets of skills that people with radio experience have that can actually be an advantage in the v/o world compared to those coming from different backgrounds. Radio people have a great sense of timing and are totally used to talking over music, which in some instances, can be a real advantage in v/o.
Gord Williams: There are checks and balances when you consider the old versus the new. If you are a one man/woman band, being able to hear what the result of just you in the headset is a plus. It may make a dead performance for some but I think it’s a lot like David Bryne said about the then ‘new’ Talking Heads sound. “The music could be implying one thing and the lyrics another.” Okay that’s not an exact or direct quote, but let’s use our imaginations! Isn’t that what we are supposed to do?
I remember getting mild to all-out war crap for playing ‘hit the post’. It was cool in the boss jock days then it was shut up and say nothing ‘music radio’. So environment has something to do with it for sure. There are those who never played by the rules, like Dan Ingram that I have heard talking through station jingles. ‘Pray to the gods of production that Dan’s ok’, we thought at the time. Following this mantra would mean not hearing the bed or the effects would be a good thing. Was Dan actually listening to them?
On the other hand in a DAW what I will do is record sequentially, mostly for dialogue. “Hmm,” Sally says “did you leave the toilet seat up again” (punch in) “No Sally, it must be the hired help here at this budget hotel” (punch out). The advantage is that you do not need two whisper rooms to accomplish the same thing without bleed through, or heaven forbid a date at some recording studio that changes on your way there.
The only thing that seems to me to require good monitoring would be producing a live radio type talk show. You want someone to operate it ideally so you can spend your millions during breaks (bring your cellphone) but if you’re not that level of budget, you do want other channels to cue things up. For me this is a non-starter as I haven’t done live to anything in a while. I don’t think anything I do lacks the live appeal or is over polished. If you stick to a just a couple of punch ins it’s an amalgamation of live/produced.
It comes back to what we all say about production of any kind. Listen, it’s in your ears.
Jean Hetherington: Low latency interfaces are your friend! I create jingles and parody songs while monitoring the music on my system. It’s not a complicated setup. I use a Focusrite interface (doesn’t matter which one) and Adobe Audition. It works great.
Peter Morley, WWSW-FM & WPGB-FM: Good question. I used to monitor music in the analog days while voicing, maybe between 1984 and 1999. And I thought that was best.
I never do that anymore. I use Pro Tools and edit together the best possible voiceover that I can - eliminating breaths (although leaving enough pause to imply breaths). I then choose music that fits the subject and voiceover. I edit the music where appropriate to enhance the read.
I don't like voiceovers that are obviously edited, with phrases bumping into each other. Overlapping for effect is another matter. That still sounds good these days when done right. Jumpy editing does not. The goal is to sound clean - not amateurish.
Voiceover with lots of distracting inhales doesn't sound clean - nor does an overly gated mic attempting to minimize breaths. Loud inhales can happen when the voiceover is a "slave" to the music bed; using the music to influence the v-o, basically in a single take.
There are of course voice actors who can cut it in that regard (complete takes without edit). And their breathing is often strategic as opposed to distracting.
I'm not one of them. HAH!
Dave Savage, Vice President, iHeartMedia Creative Studio: I occasionally listened to the music while recording a VO back in the analog days but don’t ever do it today. I don’t think I need to anymore because I am more experienced and feel I can do a better VO without the distraction of the music. In fact, I rarely wear headphones to hear myself because that’s distracting too.
Rafe Sampson, Sampson Media Inc.: First off, thanks for saying “analog days” instead of old days. Cuz I go back to pre-multi-track production, where you had a music bed cued on the turntable or second reel to reel, sfx in order needed prerecorded on cart and a script and stopwatch in front of you. Then you start all your machines and do the read while punching the cart start button and record the whole thing on the fly. Multi-track made it much simpler, especially concert spots or any with a lot of music changes. Cut the bed together, do a rehearsal read or two to get the tone and timing right, and then let it rip while recording.
Now, I seldom, if ever, listen to the bed while recording the voice track. I hear it before and again during the actual assembly. But the actual recording comes in sentences, phrases or chunks. That way it’s easily pieced together over the rest of the production. Even if you’re attempting to match beats with phrasings, that’s done more effectively in post. Plus it’s much easier to concentrate on the performance. I think when you try and record with the music in your headphones it actually pulls your attention away from the vital aspects of the performance. Also, it’s much more effective in maintaining necessary “white space.” Pauses and silence are a vital, and often overlooked aspect of voice over performance.
This workflow is likely not a universal approach, but after 15 years of radio prod and, lemme see, 18 years of full time VO, it works for me.
Jim McCutcheon, Entercom, New Orleans, LA: I use Pro Tools on an iMac and a Mackie mixing board. There really is no way to listen to the Pro Tools in playback while I’m recording. When I use solo mode I get a weird digital delay on my voice as I try to talk over the bed. So everything has to be laid down on its own then moved around and mixed. But I’ve been doing it this way for so long it seems like second nature to me. If I had a regular console I could isolate those tracks in a mix minus to listen while recording. But I don’t see an advantage in that. Just lay down a killer read then make the music and sound effects match that.
CJ Goodearl: 9 times out of 10 I'm reading dry VO. If the producer has music sometimes they'll play down the line to give me a feel for the tone of the spot, which is always helpful.
Bottom line, whatever the producer and creative want, I'll do!
Archer and Valerie Productions: With DAW multi-tracking it is a rare occasion when I get to do a VO while monitoring a bed, but when I do, I truly savor it. I even put on-air-style processing in my monitor mix to get that true experience of posting and rhythmic phrasing, complete with a gentle finger on the music fader. It's a rare occasion because clients seldom have music picked out ahead of the delivery of voiceovers. But in those instances of imaging or doing full copy-to-produced spot production, I love doing it "the old way," though some of the younger people in a radio building have no clue nor desire to know what that "way" is. On the upside of the digital multitrack approach, there are tools like ducking and level stabilizing, plus the ability to split or slide bits of your VO around over a track, but I still prefer doing it the old way, with one exception: Instead of sending the audio directly to a reel or cart, I send the voice and music to separate tracks, so it's not the final mix. There's always that client who thinks the music is too loud.
Ty Ford: I think in order to do it well, you need to have a built-in music clock in your head and not everybody has one of those. Yes, hearing the music can make a huge difference and in the case of donuts - music with singing in and out — it’s vital.
I don’t know what trouble others may have had. I’ve had no problem with it in the analog days or later using DSE7000 and Audicy workstations or Pro Tools from 6.0 up through 10, where I am now. I don’t do it as much now because I don’t work that often with jingles.
Morgan MacGavin, Salem Music Network, Nashville, TN: I started in radio when I was 16 years old at this little mom & pop station in Montana. Despite being on the cusp of a new century—complete with the apocalyptic scare of Y2K—they were using carts, reels, cassettes, and turntables. (If anyone needs some tape spliced, I’m your girl!) I didn’t have a choice but to learn to do production by recording my VO at the same time as the music bed and even sfx. A few years later, though, I was at a new station being introduced to Cool Edit and the ability to edit, and I can’t say I really looked back.
These days I record my VO dry and mix the bed later. There are a few reasons I do that:
- It’s faster. Back in the day I can’t tell you how long it would take me to do a single 30. I mean, if you flub up, you have to start over. But if my audio is dry, I go through and in just a few seconds have everything edited and ready to set on top of the bed where I can do quick volume adjustments and move on to my next project.
- I don’t always pull from one read. I’ll try a few inflections and mix and match. Sometimes I don’t know which read I want until I put it with the music and I’ll wind up swapping out clips. You can’t do that if the music is already there.
- Adding sfx or song hooks either becomes impossible or a real chore.
- I like to beat match and edit my beds to fit my VO. Especially if I’ve pulled a 30 second bed that is actually 28 seconds.
Nostalgically speaking, I’d be all about going back to the analog days. In reality? I’d rather lie on hot coals than spend an entire day without a DAW.
Kyle Whitford: This is a great question and deserves real attention. Listening to music while performing voice over can make a huge difference. The only time I do it now is while singing a jingle demo. But in the old days having music inspiring the vocal performance created a synergy that's mostly lost now.
I no longer use music for a live guide but I do have a feel in mind. Plus, with digital editing I can customize the music bed much quicker around the voice. But I might listen to literally 30 or 100 beds to find the right fit.
I once did a personal study of the top 30 or so award winning commercials for a few years. Surprise - Most did not use music beds under the spot. They used targeted music only for a brief and specific purpose. Like the theme from Jaws for a few seconds to create tension. So maybe the idea of music under an entire spot is a bit 'local' when all is said and done.
K.M. Richards, K.M. Richards Programming Services: When I did production at KHTY/Y-97 in Santa Barbara in the late 1980s, we had a four-track production studio, but I was the only one who ever used it as intended. Everyone else would record everything -- voiceover, music bed, stingers, effects -- at the same time. Even our production director said he only used the four-track machine to playback dubs from agencies and other stations. (I did production in the evenings after my afternoon drive airshift.)
The real advantages to recording my voice track separately from the music bed? First, it gave me much better control over the level of the bed compared to the copy during mixdown. Second, if I got 45 seconds into a one minute spot and blew it, I could punch in at the end of the last sentence I got right and not lose that. (Try doing that when you were recording with the bed in a single take!) That also minimized the chance that I'd blow a line earlier in the copy on later takes.
I think anyone in this digital age that doesn't have a multi-track studio set up to record voiceovers on a track while playing back the bed in a headphone mix is doing it wrong. If anything, digital recording gives one the ability to edit the voiceover independently, perhaps to better hit a stinger in the bed. Couldn't do that with analog ... you were more dependent on hitting all the posts right during the recording.
One good thing about the evolution from analog to digital: At least we don't have to remember to switch the headphone monitor to the record head instead of the playback one!
Todd Franklin, Tee One Productions LLC, Columbia, SC: This is the "TRUE APPLICATION" of multitrack recording that a plethora of producers have forgotten about. I happen to apply this technique to all my production and clients "truly feel" the essence of the message and everything tends to flow more flawlessly. I have several different ways of utilizing this method and the only thing that I can say is that is highly effective. The soundcard is the key with the proper loop that will dispense this effect in unison with a mixing console (at least for me)!!
And production music in the box first has an intro then a HEAD then more intensity following. Commercials are not all like that nor should they be. No to cookie cutter worlds. Now true in the day the old LPs with "The Jolly Commercial Singers" did inspire me to write but I prefer to grab the direction of the commercial I'm producing then seek out the music and punctuations later.
Paul Lowman, Abu Dhabi, UAE: Monitoring of the main background track is an essential part of what I do. In designing my own home studio the ability to monitor was a key focus of the design and equipment acquisition. I believe the ability to listen to the music while tracking enables the VO artiste to deliver the script with a more accurate pace and intonation which makes for a better delivery.
Unfortunately, humans cannot voice to a track running at 128 BPM without a reference. To do so runs the risk of altering the authentic sound of the VO to match the music. Most basic mixers today are equipped with at least one mix bus. Content producers should forward a sample of the music bed to VO studio/artiste. I believe it enriches the experience of the VO artiste.
Andrew Frame Soundworks: I've never worked at a facility where the board was set up to allow monitoring the music while doing the VO.
The closest thing was starting the vinyl, and doing a read over the music, recording it all in one take.
Since DAW's have come in, I don't use a board at all. Everything is direct in to the computer. Once the VO is done, I'll go match a music bed to it.
Jay Rose CAS: For what it's worth. I've almost always gotten better VO performances by leaving the music off. Maybe play it once for the talent, in the control room, so they get a feel for its structure and energy. But not in the phones... unless there are a lot of very tight cues within the piece.
I've found two reasons why this is true. But with a caveat: I moved from station production to boutique agency production early in my career. Agency studio budgets are a tiny fraction of the media buy, so spending twenty minutes to perfect a voice track doesn't bother them... particularly if it results in what we all think is a better spot. The pressures of a station production room can be very different.
My philosophy was, every piece of well written copy has its own rhythms. Syllables are stressed, and overall energy ebbs and flows, based on the story and the sell. A good voice actor or director recognizes this, and tailors the read -- moment by moment -- to the message. But spoken rhythms keep changing. Ad copy doesn't match a constant iambic pentameter, nor does it adhere to a strict 4/4 musical meter. If the talent is hearing a constant drumbeat in their ears, or struggling to hit a post, it gets really hard to pay attention to the stresses in each phrase and the overall curve of the message. Remember: the copy and performance sell the product. The jingle's job is to be a branding mnemonic.
So in analog days, we almost always recorded multiple voice takes to 1/4" tape, and then either mixed with music playing on a second machine or laid the edited 1/4" up to a multitrack. The quality hit of another generation of full track at 15 ips, on a good recorder, was insignificant.
If we got a perfect read, it took just a minute or two to mix it with the donut. And maybe another minute, if we wanted to try the same read a fraction of a second earlier or later... sometimes, this can catch the music's rhythm in better places. (It was pretty common, when mixing filmed commercials, to improve things by bumping one of the tracks "a single perf" earlier or later. In 35mm, that was 1/96th of a second. These days I keep my DAW's "bump" control set to 1/4 frame, about 1/120th second.)
But the second reason kicks in when you don't get an absolutely perfect read. If you're going to edit, you'll have to re-lay the music anyway. Editing let us opt for the perfect setup from take 3, the great selling line in take 7, and the wild pickup where the talent finally pronounced a critical name properly.
And if a read was pretty darned good but a second too long, you didn't have to scrap it. Just pull out one or two of those sucking announcer breaths, replacing each with a shorter silence. Replacing breaths with shorter silence usually improves the rhythm as well. And it lets you use more compression without sounding ugly or destroying the intimacy, so the spot can be louder on the air.
As far as the "fast forward to today" part of this Q It Up: There are still a few times when I want to record digitally against music.This was never a problem with DSE7000/Audicy, and it isn't with Nuendo, ProTools, or other modern multitrack DAWs: the record bus and the monitor mix are separate. It may be hard if a DAW is wired to an older broadcast board with has only one output, but that same problem would exist if an analog multitrack were wired the same way. A patchbay can come to the rescue.
As I said, this works for me. YMMV.
♦
Thanks to all who responded. Your input is valuable and appreciated. If you have a question you’d like to see posed to the RAP Q It Up panel, email it to