Two of my clients have their voice
[Ryan Santos] "What is the normal practice, edit the video first then record the voice-over later?"
Only if you want to go very very insane. :)
For something that is very narration heavy (anywhere from fairly heavy all the way up to wall-to-wall voice-over) the easiest way would be to throw your narration pieces on the timeline and then edit the visual to them. In most videos like this the narration dictates the pace of the visuals, so that is by far the easiest way. You can always adjust the voice tracks (breaking them apart, sliding them to create spaces, etc.).
You CAN stick a mic into your computer and record directly into Premiere, but that is the very lowest of the low-end way to do it. You will get better results by either 1) having the VO tracks recorded in a professional audio studio and they deliver you an audio CD or a data CD with .wav files on it... or 2) recording yourself into a more specific audio program such as Audition, Soundbooth, Soundforge, or one of the flavors of Vegas audio. There you will have much more control over the sound files... can cut, paste, trim, eliminate breaths, fix levels, equalize, all that good stuff.
In our particular case we have a small audio booth in our studio... it is hardwired (mic and headphone cables running through the walls and floors) to each of our edit suites, where it first goes through a small Mackie mixer in each suite, and then into either the Canopus, AJA, or Matrox breakout box (depending on the suite). We used to have an Alesis compressor/noisegate in that chain as well, but compression can now be done so well in post we have eliminated it. As the talent performs in the booth, we record them into either Audition or Soundbooth... save the tracks as .wav files... and import them into Premiere later.
And of course...use the very best microphone that you can get your hands on (and a pop screen is a good idea, too).
Fantastic Plastic Entertainment, Inc.
Todd explained it pretty well. May I add that recording a narration track and only then dropping in bits of video to fit the narration is not usually going to create a superior video product. You are making a radio show with pictures when you edit this way. That's giving up half the power of video.
What gives superior results is a hybrid approach that combines the strengths of both the audio and the video in a complementary fashion. It requires writing to the images rather than just hoping somebody comes up with images that will fit what's been written.
[Mark Suszko] "What gives superior results is a hybrid approach that combines the strengths of both the audio and the video in a complementary fashion"
That's true of course... but to cut video first and AND then attempt to marry audio to it (without having planned that much) is a nightmare. If I'm doing something that is extremely solo-narrator heavy (an industrial film, for example), I always like to lay audio down, then scissor it and slide it as needed... five frames, 10 seconds, a minute...whatever... to give me time to let the visuals do the work. To me snipping and sliding audio tracks is easy as pie... video, well (depending on its complexity), sometimes not so much. I'd never suggest producing radio with pictures, hopefully I didn't give that impression. Just for me anyway, at least rough laying-in audio is technically a much easier workflow.
On the other side of the coin... I was always amazed at NFL Films, especially in the early days. Supposedly their cutters got the film fresh out of the soup and immediately started hacking away at the dozens of barely-dry 16mm reels... TOTALLY telling the story with the pictures only. AFTER they they had the visual part of the show complete, THEN a script was written to it and VO tracks were recorded..... of course by the late and legendary John Facenda... and I swear his pipes could have made your family's Thanksgiving day touch football game sound like the most important matchup in the history of sports.
Fantastic Plastic Entertainment, Inc.
I think Todd and I are on the same page here, I didn't bother addressing the alternate method because he covered it well already.
And obviously, as Todd says, it is easier to lay guide audio first and then cut video to it, than the other way around. You should just be careful that you don't miss opportunities to make the video and audio synergize, to let the video sometimes lead or trail but say what the audio track leaves unspoken, for a heightened effect. Otherwise what you get is so stultifyingly literal and blatant, nobody pays it much attention.
Sometimes this cannot be avoided, as in the case of a training piece that's literally about how to insert tab "A" into slot "B". In that case, yes, you pace a good sound track and cut video into it and collect your check. But for more artful projects than that, things that are more "conceptual", you might want to look at making video and audio deliberately clash at times, to point something up or just wake up an audience.
A good test of your visual-auditory synergy is to watch the thing with the sound off and see if you can still understand at least the basic narrative thread with visuals alone. Then go back and try listening without seeing any visuals. This may uncover areas where your story, structure, or script is especially weak and needs tweaking. Since you know more about the project than most people, it is good to have someone walk in with no preconceived notions and screen this with you... see if THEY get it or not.
I mostly work in very short form projects, PSA's and promos and short news VNR's and then some longer-form training pieces. Now not all, okay, not most, of it is "artful". :-) Still... When shaping a story, even one as short as a PSA, I try to think in terms of an overall general visual metaphor first. If I can come up with a clear one, I try to arrange everything including the narrative, around it, or to leverage off of it. Such a setup can make the piece able to communicate despite language or cultural barriers. For more on this, try a google or wiki search on "semiotics".
"Oh, you wanted to RECORD that?"
I don't think that there's as much of a gap between what Mark is saying and what Todd is saying as it would appear on first blush. Nobody wants to do "radio with pictures" but life is ohhhh so much simpler when you at least have something to lay down and start cutting to. Personally I start with the music track, add a few clips to see what works, then start adding in the voice over, then the rest of the video.
Of course you slide voice over all around to accommodate your footage. That's why we all went non-linear all those years back. And, in terms of creating the VO to fit the material, it's important to have a good idea at the outset what you'll be doing and what material you have available. I've also been in situations where I've had to go back to the talent and have them do a few additional or different lines as pick-ups. (And here's where it helps to have a good relationship with the talent.)
What hasn't been discussed is the importance of producing / directing the VO session. Maybe it's because I'm almost always the writer on my projects but there is simply no way I would ever have a client hand me a CD of a narration over which I've had no control. Voice over is a very sophisticated and highly focused form of acting and if it's my project, I'm in control of that highly critical part. It's also how I protect myself from: "Did I tell you that my best friend does opera -- so he's got a really great voice," or "My little brother is a DJ on weekends, he has all the equipment and everything else to do this." Gee, sorry. We're a signatory of the AFTRA union and therefore we're bound by contract to use union talent. And wasn't the quality of our work what brought you to us in the first place?
Scratch tracks can also be highly problematic and almost always result in additional editing once the real VO is brought in. Maybe it's just me but I find it very difficult to use a non-pro read to accurately predict the timing and rhythm of a professional read.
That's my over-stated and slightly off topic answer.
Nick and I seem to have the same method. Music, voice, video, adjust video, adjust voice, adjust music, then rinse and repeat as desired...
Tagging onto another thing Nick mentioned, I also have to echo the importance of producing the voice over session.
Getting a CD of the session from the client is no different than getting your video footage from the client, rather than shooting it. You can probably force it to work, but it immediately paints you into a corner. (Usually the corner that's farthest form the door - so to speak.)
It also instantly puts a limit on the quality level of the finished project.
It's like someone hiring you to decorate their house but insisting that you use the crayon drawings their kid made at school. The clients might still love it - just like a parent likes their kid's art work - but you won't ever want to use it as an example to potential clients of the quality level of your work.
[Timothy J. Allen] "Getting a CD of the session from the client is no different than getting your video footage from the client, rather than shooting it. You can probably force it to work, but it immediately paints you into a corner."
Oh god yes... when I said "get a CD from the client" I never meant to suggest not being in on the voice-over session, hope it wasn't taken that way. I've had clients provide voice tracks, i.e., take care of booking the talent they wanted, hiring an audio studio if it was elsewhere, etc., then delivering on CD (in the olden days, when the network was FedEx)... but I always tried to be able to direct the VO session, even if I had to do it by phone.
I think the only voice sessions I've used that I didn't personally direct have come from either Charlie van Dyke or Don laFontaine. And you know that whatever you get from those guys is gonna be gold, plus they are not very directable anyway..ha (actually, not true... Don actually does take direction well... he just usually doesn't need much).
Fantastic Plastic Entertainment, Inc.
[Nick Griffin] "Of course you slide voice over all around to accommodate your footage. That's why we all went non-linear all those years back. And, in terms of creating the VO to fit the material, it's important to have a good idea at the outset what you'll be doing and what material you have available."
Of course, even before I went non-linear, I went word-processor and the best way to have a good idea of what you have for material was to have shot it to a script in the first place...
This is what may be the largest, yet most overlooked point with new producers (as well as some old ones) these days.
In my experience, when the narrator has to be brought back in to do pickup lines for reasons other than the client saying "hey...can we add/change this...?" at the 11th hour, typically it is because of a script that either 1. didn't exist before shooting (as in the case of the NFL Films example, which is a perfectly legitimate situation) or; 2. wasn't able to be fully realized in images due to logistics or tight schedule, etc...
Shooting any communication piece (corporate or similar project.) without a script is a recipe for disaster.
(..obviously for news, sports, doco, etc. the situation is different.)
My personal example for clients is running to the grocery store without a list...you come home with a cartload of Ding Dongs and Frozen Burritos and remember about the time you need to visit the head that you originally went to the store because you were out of TP.
Creative Cow Host,