Audio workflow for this project

Chris Johnston

Chris Johnston

Veteran
I do a weekly tv show (cooking show) and am trying to speed the workflow a little more so let me explain the setup and see if I'm doing this right
The show is a 3 cam deal but for the sake of simplicity I'll just simplify the explaination a little and explain HOW I shoot this.

The main "A" cam has 2 lav mics input to the cam (1 goes to chan 1, the other goes to chan 2)

I import the clip, which is generally about an hour long. (the show is a half hour but I just let the cam run recording and just cut out the parts where we break for commercials and prep for the next item of food to be cooked)

I duplicate the clip and pop it on top so I have vid track 1 and 2, and audio track 1 and 2. I then delete video track 2, put fill left on audio 1, and fill right on audio 2. (now I have the one "A" cam video and each of the "actors" on their own audio track.

The problem is....Chefs are always stirring the food then banging the spoon on the pan to get off the excess. BANG BANG BANG, which pegs out the audio
(or they make a joke and both laugh loudly and pegg the audio ect ect)

Some times they don't talk as loud for a couple seconds while he/she is concentrating on some particular dice job or whatever.

I should probably be using the compression effect to bump up where it needs bumping and chopping the tops down where it's needed but I'm not really familiar with that effect yet. What I have been doing is getting fairly good audio in the camera from the start and the bumping up the gain a little on each audio track (to get the low spots up) and then popping the "dynamics" effect on each audio track and setting the limiter to chop the tops down. Probably not the best way to do it but since it's one giant hour long clip, it saves me time keyframing down the high spots and keyframing up the low spots.

Now that I've explained the problem, a couple questions.....
Is this the best way to do it? (given that it's one long file, sure does save me a lot of time....)
Is this destructive? (sure sounds better to my untrained, non audio guy ears...?)
Should I just be putting the "dynamics" effect on the master audio, or on each actors audio track?

ps, It's a low budget show so ...no audio guy, just me.
 
Certainly compression is a "time saver" if you don't want to "ride the fader" or in your terms "keyframing." However, writing automation will sound more natural and will cause you to use the compression less. In short, yes, in your case, it is probably the best thing to do. It's not destructive if it sounds natural. Compression is often used incorrectly and when used incorrectly, compression artifacts can become very audible. For my mixes, I tend to write a ton of automation and have less compression (if I have a little bit of time). The compressor that I usually have set is very very light. I typically put a compressor on a dialogue "submix" or "bus" track and ride the individual dialogue tracks INTO the bus - to me, this sounds more natural. However, if needed, i will pop in a compressor on each actors audio track for those unbearable peaks (shouts, screams, etc.). Hope this helps.
 
Chris Johnston said:
The problem is....Chefs are always stirring the food then banging the spoon on the pan to get off the excess. BANG BANG BANG, which pegs out the audio
(or they make a joke and both laugh loudly and peg the audio etc., etc.)
Click to expand...

The word you are looking for is "limiter". This is one of the reasons to use a field mixer like the SD MixPre-D -- for the limiters. They will largely take care of this problem for you.

If you wait to post to deal with it, you can put a filter on it and specify a hard limit (like -6 db), but since the sound is already clipped, it will still be clipped -- just at a lower volume. Which brings us back to using a field mixer...

As to compression, in my experience nearly all presenters can use some small level of compression to make what they are saying more easily understandable in the final product (DVD, blu-ray, web, broadcast, whatever). But its pretty easy to over do it. Subtlety is your friend.
 
Thanks for the comprehensive reply Ryan. While I don't know much about audio, I've squeaked by using the "sounds ok to me" approach on a lot of my projects. I was really uncertain if I would be better served by doing the effect on each audio track or just one time on the master track which I think would provide the effect coverage on all the tracks???? (not really sure if thats how the effect works on the master track.....I think so...."he said without testing it" ) EDIT, just did a quick check....No you can't put it on master track, just each audio track.
 
Last edited:
Chris Johnston said:
Thanks for the comprehensive reply Ryan. While I don't know much about audio, I've squeaked by using the "sounds ok to me" approach on a lot of my projects. I was really uncertain if I would be better served by doing the effect on each audio track or just one time on the master track which I think would provide the effect coverage on all the tracks???? (not really sure if thats how the effect works on the master track.....I think so...."he said without testing it" )
Click to expand...

With this type of stuff, it's hard to tell without hearing the material. The use of compression has so many variables and it depends on many things, such as how much unwanted noise is lurking into the recording. If there is a lot of "unwanted noise" present in the recording already, it's only going to get worse with compression.. which leads me to say "it depends!" Well, if the compressor is on the MASTER TRACK (which, to me, means the bus that the dialogue, music, and effects are routed to), then music and effects will be compressed along with the dialogue, which is probably not what you want to achieve. I'm not sure how Premiere routes audio, but I'd advise to route the individual dialogue tracks to a bus called "DX BUS and have the DX BUS routed to the Master Track. I would then place the compressor on the DX BUS instead. A soft compressor on the DX bus with fast attack times would clamp down on the peaks, but like I said, it would be a very light compressor coupled with a ton of volume automation (or I guess "keyframing") on the individual dialogue tracks.
 
Good thought Bruce, I didn't have the budget for a field mixer this time around. (actually, didn't even think about it....DUH) but looking at the big picture of my workflow, I probably spend the vast majority of my time doing audio correction/fine tuning vs color correction/image manipulation etc. I've heard for years how the audio is more important than the visual end......I'm beginning to realise that now.
 
ryanjf said:
With this type of stuff, it's hard to tell without hearing the material. The use of compression has so many variables and it depends on many things, such as how much unwanted noise is lurking into the recording. If there is a lot of "unwanted noise" present in the recording already, it's only going to get worse with compression.. which leads me to say "it depends!" .
Click to expand...

Luckily for me, this particular show is shot in a studio kitchen, so no (or not much) unwanted noise (other than stuff like sizzling food when they pop something in a hot pan) So it seems like this would (and apparently does) work well in this situation. Probably not so advantagious if I was shooting a short or something.
 
Jim Simon said:
Try getting new mics. Sounds like you're using omnidirectional, where you should be using hypercardioid.
Click to expand...

No, not on a cooking show. Been there, done that.

Cooking requires the chefs to do a lot of physical motion and they often bend their heads down to look at something (while talking of course) or turn their heads to the side and talk to someone while working with something on the counter. If you use a hyper under these circumstances it sounds pretty bad -- you can really hear them moving in and out of the sweet spot. Omni lavs are the way to go, even with the extra noise pickup.
 
You must log in or register to reply here.
Back
Top