Is this an alright way to go about mixing your audio? I'm not entirely too sure what all I should be doing for post audio work. Setting levels in the audio mixer? what else? Someone told me on here to fade the audio after before and after cuts to get rid of ambient clicks. How low should I fade out? -1 db?
post Audio Mixing in premier and in general
- Thread starter skywalks
- Start date
ProfessorU
Veteran
You would fade out as low as it will go, probably -90dB or so. Premiere is totally okay for audio work, but it isn't very powerful. If you were doing a project with serious work on the soundtrack, you'd want a different application.
There may be an auto fade in/out option on the sound clips. Check around in the preferences options... or maybe someone here knows where it is.
There may be an auto fade in/out option on the sound clips. Check around in the preferences options... or maybe someone here knows where it is.
SMuir
Well-known member
While fading out and back in across cuts will get rid of the snap of a hard edit, a cross-fade will blend over the seam in the audio.
Although its entirely possible to mix a film in Premiere, its not a good idea. I've said it before, and I'll say it again: mixing and cutting audio in premiere/FCP is kind of like using a crescent wrench to drive nails.
The industry standard for audio post-production is ProTools, although Digidesign is facing fierce competition from other platforms (eg. Soundtrack Pro, Adobe Audition). Locking the picture and migrating it to an audio editing/mixing platform will make the proccess a whole lot more efficient, and give you access to far more powerful audio sweetening tools.
You may find it to be more cost-effective to hand the project over to an experienced sound editor for the audio post-production than to do it all yourself.
Although its entirely possible to mix a film in Premiere, its not a good idea. I've said it before, and I'll say it again: mixing and cutting audio in premiere/FCP is kind of like using a crescent wrench to drive nails.
The industry standard for audio post-production is ProTools, although Digidesign is facing fierce competition from other platforms (eg. Soundtrack Pro, Adobe Audition). Locking the picture and migrating it to an audio editing/mixing platform will make the proccess a whole lot more efficient, and give you access to far more powerful audio sweetening tools.
You may find it to be more cost-effective to hand the project over to an experienced sound editor for the audio post-production than to do it all yourself.
Steve House
Veteran
I might add to SMuir comments to say that Steinberg's Nuendo and Wavelab are snapping at ProTools' heels. Sony's Vegas actually got its start as an audio editing application and though it's now marketed as a video editor, it retains very strong multi-track audio capabilities.
It sounds like the OP has little or no background in audio. I'd strongly suggest he start off with some 'book larnin' and grab copies of Jay Rose's two books, "Producing Great Sound for Film and Video" and "Audio Postproduction for Digital Video."
It sounds like the OP has little or no background in audio. I'd strongly suggest he start off with some 'book larnin' and grab copies of Jay Rose's two books, "Producing Great Sound for Film and Video" and "Audio Postproduction for Digital Video."
Steve House
Veteran
There's much more to stereo than simply recording two tracks at the same time. It starts with proper positioning of the microphones and goes from there. Premiere and other workstations will import both channels as a single stereo track but whether you leave it that way is another matter. For instance, it is not at all unusual to record a mono signal to both channels with one channel set to a lower level than the other so that if a sudden loud sound forces the primary channel into distortion you have the other lower level channel as a backup. Dialog is virtually always recorded in mono and another common arrangement is to put a boom mic on one channel and a lav mic on the other, picking the one that sounds best in post as the one used in the final production. Yet another arrangement would be micing each of two characters or both an interviewer and interviewee, putting each mic on its own channel yet splitting them and using them as two mono channels in the edit.
davbeisner
Well-known member
Excellent advice from all so far. I've used ProTools for several years in my personal studio, but I also really like Audacity. It's an open source audio editor that, in my opinion, does more than just "nip at the heels" of ProTools. I've actually gone to using Audacity for almost everything I do, leaving ProTools only for the big stuff or when I'm actually composing music (something I try to avoid!!)
The nice part about Audacity is that its free and the interface is easy to learn. Check it out at audacity.sourceforge.net.
If all your using right now is Premiere, then I would definitely recommend using something else. Audacity is free, easy, and powerful. There are other programs that are as good and better than Audacity, but they tend to be expensive.
But more than that, I would highly recommend you do some more research and learning on audio. Sounds like you're new in the field and the more you learn, the better you'll be. There's a lot more to audio than setting up a couple mics, recording them to your camera, and fading in and out or cross-fading with your video cuts.
Oh, and on your cuts--Premiere's cross fade SFX and fade in/out SFX work very nicely and completely remove the "clicks" that so often come with cut audio. In Audacity you can also use a fade in/out feature or do a cross fade by drawing your waveform up or down on adjacent tracks. You can also go all the way in to do very tight waveform edits and get rid of your pops and clicks if you accidentally make any.
The nice part about Audacity is that its free and the interface is easy to learn. Check it out at audacity.sourceforge.net.
If all your using right now is Premiere, then I would definitely recommend using something else. Audacity is free, easy, and powerful. There are other programs that are as good and better than Audacity, but they tend to be expensive.
But more than that, I would highly recommend you do some more research and learning on audio. Sounds like you're new in the field and the more you learn, the better you'll be. There's a lot more to audio than setting up a couple mics, recording them to your camera, and fading in and out or cross-fading with your video cuts.
Oh, and on your cuts--Premiere's cross fade SFX and fade in/out SFX work very nicely and completely remove the "clicks" that so often come with cut audio. In Audacity you can also use a fade in/out feature or do a cross fade by drawing your waveform up or down on adjacent tracks. You can also go all the way in to do very tight waveform edits and get rid of your pops and clicks if you accidentally make any.
SMuir
Well-known member
David brings up another important point in passing: most video editing platforms will not allow you to fine-tune your audio edits beyond the start/end of a frame. In the audio side of things, a single frame is a rather long time, and I often find myself zooming in a great deal to make editorial changes that may only be 1/10th of a frame.
As for mixing in stereo: if memory serves, Premiere's default track setup is such that each audio track is a stereo pair (ie. 2 channels linked together). Each audio channel has a left and right pan control, which defaults to left=hard left and right=hard right. That's all well and good if you're working with stereo content, but production sound is almost always recorded mono. If both channels contain the same mic signal (ie. 1 mic equally split to two channels), then everythings groovy and your dialogue will come out of the center of the image. However, single-mic shoots are often recorded such that the mic goes to both channels, but one channel is a bit lower (in case of incidental peaks). In that case, you would have to take one channel, discard the unused channel, and pan it back to the center. Now, I haven't cut in premiere since v5.1c, but they may still call that opperation "take left" or "take right".
In cases where the boom was recorded on one track and a lav on the other, its often prudent to go with the best-sounding mic, and discard the other channel.
Just to drive the point home, there are two different sides to stereo sound: stereo micing and recording, and stereo mixing. For more information on the recording side, google "X-Y stereo", "A-B stereo", "M-S stereo", and "ORTF". Stereo mixing, on the other hand, usually involves mono as well as stereo elements "panned" from left to right in the stereo image, as appropriate. Dialogue, incidentally, is almost always kept mono and panned to the center.
As for mixing in stereo: if memory serves, Premiere's default track setup is such that each audio track is a stereo pair (ie. 2 channels linked together). Each audio channel has a left and right pan control, which defaults to left=hard left and right=hard right. That's all well and good if you're working with stereo content, but production sound is almost always recorded mono. If both channels contain the same mic signal (ie. 1 mic equally split to two channels), then everythings groovy and your dialogue will come out of the center of the image. However, single-mic shoots are often recorded such that the mic goes to both channels, but one channel is a bit lower (in case of incidental peaks). In that case, you would have to take one channel, discard the unused channel, and pan it back to the center. Now, I haven't cut in premiere since v5.1c, but they may still call that opperation "take left" or "take right".
In cases where the boom was recorded on one track and a lav on the other, its often prudent to go with the best-sounding mic, and discard the other channel.
Just to drive the point home, there are two different sides to stereo sound: stereo micing and recording, and stereo mixing. For more information on the recording side, google "X-Y stereo", "A-B stereo", "M-S stereo", and "ORTF". Stereo mixing, on the other hand, usually involves mono as well as stereo elements "panned" from left to right in the stereo image, as appropriate. Dialogue, incidentally, is almost always kept mono and panned to the center.
davbeisner
Well-known member
Actually (and I'm running with Premiere CS3 and looking at it right now), Premiere doesn't allow you to do any sort of panning of your audio tracks if its a stereo track, nor can you split a stereo track to two mono tracks. You've actually got to export it to SoundBooth to do any of that (at least as far as I can tell--if anyone knows differently, please tell me!!)
Goldmond
Active member
A related Question. Is there a set level that's preferable for dialogue? I seem to remember somewhere between -12 and -20 but don't remember where. If so could you set all the levels to something lower and then bring up the master volume in the final mix? Would that accomplish the same thing?
davbeisner
Well-known member
Ideally you want to maximise the potential of any particular source, getting all your levels as close to unity (0 db) at each source as possible. That will help to minimize distortion due to peaking and also minimize hiss and background noise that is amplified along with the desired audio when you discover you have to increase the gain in post.
I think the biggest thing to keep in mind is that, in the digital world, 0 db is actually the point at which peaking begins, so the rule I try to follow is to have things run as close to 0 db on any analog equipment (mixers, outboard gear, etc.) and then run about -6 to -12 db on my digital equipment (or somewhere in the neighborhood of three to four bars on a fifteen bar graph--remember it's not linear, its exponential).
That will give you the cleanest audio with the clearest tones and the best gain before noise.
I think the biggest thing to keep in mind is that, in the digital world, 0 db is actually the point at which peaking begins, so the rule I try to follow is to have things run as close to 0 db on any analog equipment (mixers, outboard gear, etc.) and then run about -6 to -12 db on my digital equipment (or somewhere in the neighborhood of three to four bars on a fifteen bar graph--remember it's not linear, its exponential).
That will give you the cleanest audio with the clearest tones and the best gain before noise.
SMuir
Well-known member
Different broadcasters will have different requirements, but most seem to prefer dialogue that sits around -12 dB, with maximum peaks not exceeding -6 dB. Other distribution formats will have different mixing standards standard; film mixes, for example, tend to be far more dynamic, with dialogue often sitting around -20 dB while some effects use up all the available headroom. Commercials, on the other hand, are often compressed to within an inch of their life (dialogue at -8, max peaks at -6).
The best way to judge the mix isn't with the VU meters, though, since there's so much more to mixing than simply hitting the standard levels.
davbeisner
Well-known member
Very true.
You can pan using the audio mixer in Premiere pro which is quite useful for basic stuff.
davbeisner
Well-known member
Thanks bimdas--I'll have to look for that. I've only recently started using CS3 so I'm not completely familiar with it yet.