Just wondering what you guys do or recommend in post to keep/have/make all your audio even and level? Programs, tips, tricks, and so on. I'm looking to "remaster" the audio from my first film/documentary and this is pretty much the only issue.
If you're going to release on youtube, your peaks can go to -6db. Premiere might have a limit of -6.02 or something, but your youtube peaks can even go all the way up to zero ( I wouldn't go above -6db).
If you're going to level for broadcast, it depends on broadcaster specifications, and they seem to vary a lot between US and Europe. US broadcast peak levels are between -10dB to -12dB.
although on this site it says US peak is 0dBFS, lots of other sites suggest -10 to -12dBFS. ... The dBFS values is what you're interested in. Whatever your NLE says is the dB, is the dBFS value you're looking for.
I used a program called Camstasia and it has a really nifty audio leveling feature.
Youtube converts your audio into the AAC format which involves a resampling process. Without going into technical detail, it's wise to limit peaks no higher than -2dBFS.
Mmm, no! Not sure where you got this info but it's seriously outdated. North American and European broadcast specs are now more similar (though still not exactly the same) and neither specify -10dB or -12dB anymore.
According to wikipedia, the CALM act only restricts commercials from being louder than the programs on the station. It doesn't specify dBFS values for the station. I think that would still leave room for the stations to choose their own dBFS levels. Of course, I'm not sure of what I'm talking about. I read someone asking the audio leveling question as he was preparing his program and someone suggested asking individual broadcasters for their specifications, hence my conclusion.No, there's two serious problems with this advice:
1. dBFS is NOT the scale that North American or European broadcasters use anymore. In other words, the dBFS values provided by NLE's (and DAWs) are effectively useless as far as broadcast specs are concerned. BTW, this is NOT broadcaster specific, it is federal law (CALM Act 2010).
2. The dBFS scale in effect measures the amount of energy encoded in digital audio, not how loud it will sound. Therefore, two different lines of dialogue (or two different actors) could measure the same peak or average dBFS level but sound different volumes/loudness. In other words, dBFS levels cannot be used to level dialogue (or a mix in general).
If you need to hit broadcaster specs, you really need a commercial Re-recording Mixer. The OP does not say s/he needs to hit broadcaster specs though and Youtube, like many/most hosting platforms, effectively doesn't have any audio specs. So the OP can have whatever level s/he wants, although it's wise not to have any peaks higher than -2dBFS.
G
Also, I'm now confused that two separate audio peaks of the same dBFS level can have different volumes. Now I have to figure out what that means.
[1.] According to wikipedia, the CALM act only restricts commercials from being louder than the programs on the station. [2.] It doesn't specify dBFS values for the station. I think that would still leave room for the stations to choose their own dBFS levels.
My understanding from preliminary readings was that dBFS is the unit used because it is independent of device, as with analog devices the same energy levels would produce different dB levels. Since for NLEs and DAWs we're working digitally, I assumed that the dB values must be the dBFS values.
... from my readings on the internet, it seems that higher dBFS is recommended for youtube because a lot of youtube users watch on laptops, which might have bad speakers, and so some sites recommended higher dBFS. That's where my suggestion comes from.
By "a commercial Re-recording Mixer" do you mean an audio post house, or a piece of software or hardware?
On a side note, RX4 has a leveler that allows you to target RMS.
I gave complete authority to the audio person to stop a take, or call a cut or a "back to ones" if something went wrong, during the shoot. And since the shoot, I've spent more time trying to figure out the NLE / DAW workflow and to understand what is possible in the DAW, than I've spent actually editing so far.
I'm trying. But you say that there are specs that MUST be met, and I'm wondering if you mean for minimum technical requirement reasons or for 'perfection' reasons.
I just can't believe that all these filmmakers sending their films to festivals and the short films being played on public tv stations are all meeting audio requirements by themselves. It's just too much work.
So what I'm suggesting is that there has to be a gap between minimum technical requirements and the levels that require a professional Re-recording Engineer.
The only thing I got yesterday was that if RMS was higher, things feel more loud.
I'm not sure if you've maybe gone a little too far ... and I never thought I'd say that!
G
BTW, as far as audio specs are concerned, do not confuse broadcast specs with theatrical specs, they're not related!
G
Just wondering what you guys do or recommend in post to keep/have/make all your audio even and level? Programs, tips, tricks, and so on. I'm looking to "remaster" the audio from my first film/documentary and this is pretty much the only issue.
One more question: Should I leave the gain adjustments alone while I'm editing, and wait until I take my audio into the DAW at the end, or is it okay to adjust now and finalize later? thx.
There are a few auto levelling tools out there. They tend to work by averaging out dBFS levels, essentially like the AGC feature in some cameras. I've never used Camtasia so I can't say for sure how it works or how well it works. I can't imagine it's better than some of the more professional audio levelling tools but even the professional levelling tools tend generally to be avoided by professionals!
You seem to be aware that your auto levelling feature is not ideal and is more of a "quick and dirty" solution aimed at presentations than a serious filmmaking tool. Depending on what features Camtasia contains, my advice would be to switch the auto levelling feature off and do it all manually, rather than to use it even as a starting point. This is because you might inadvertently find yourself in effect fighting whatever it is Camtasia is doing and ultimately reducing the quality of your dialogue.
G
I have a really simple setup which is extremely effective. My workflow is:
1. Put the speech bit in at -6db
2. Add in a load of foley from my library or wherever I can get it.
3. Get Mike McGuill to compose some awesome music
4. Mix it all together, personally until I like it.
5. Give it to an audio professional to sort out the mess.
Last time I did this, the theatrical sound held up well next to other shorts.
Totally understandable. I guess I should have included that this is, more so, for outside the professional field. That is, too, to say and assume that professional tools are going to be quite expensive and outside of a budget.
I have some idea of what's being discussed in this thread but it's obvious that I would have to dive well into the field of audio alone to get something I'd consider good (better than what I can do). With this, I'm thinking outsourcing may be needed for my audio needs. Yay nay?