Adobe Premiere Exporting Master Audio File

T

taras4444

Well-known member
Hey guys. I'm coming to the end of a feature film I have been working on the past two years and am readying everything to be exported (ultimately for a DCP). I'm going to export a master copy as a DNxHD file for the best quality but do not have much experience in exporting/mastering audio. I've been advised it's a good idea when exporting the film to have three audio tracks (a centre track for dialogue, and then a left and right for effects etc). But I have no idea exactly how to go about this. In my adobe cs6 project I have upwards of 15/16 tracks. Each track is for something specific ie. tracks 1-6 are all mono dialogue. Tracks 7-10 Effects, tracks 11+ music and so on. Is there a way to condense everything audio wise into three when exporting? In export settings I can only see an option usually for stereo or mono audio aac files.

Any and all advise will be much appreciated. The DCP company haven't really been very helpful as they are doing it all off book for me as a test (a big company with a lot of paying clients). Thank you!
 
Assuming it's mixed to sound the way you want, the way I'd do it is to mute the dialog tracks and export a stereo mix of the effects, and then unmute the dialog tracks and mute the effects tracks and export a mono track. But really, this all depends on how you've mixed it. Also, for a feature I would have thought they would want 5.1 or 7.1, in which case you may need more separated tracks.
 
This will need mixing and balancing to suit the delivery format that is chosen, just exporting tracks from an edit is not doing the quality control and level adjustments that are required for encoding to the chosen delivery format for playback.

For stereo you would normally just deliver a stereo final mix and a mix minus M&E (Music and Effects) for international dubbing if it has been requested and budgeted for. It may also be that mix stems are to be delivered and they will be additional individual stereo mixes of Dialogue, Effects and Music.

5.1 and 7.1 demands a final mix in the required format you wish to have with a mix minus M&E and the stem mixes if required.

This is all very complicated and should really be completed in an audio post facility as the levels and balancing for stereo and surround is far more than just exporting tracks from an edit suite, in addition 5.1 and 7.1 have strict criteria that is laid down by dolby or the other format providers and encoders especially for theatre delivery.

There is one other older dolby format called LCRS or pro logic and that may be where the centre side is coming from but once again it is a delivery and mix format and not an export format as the LCRS signal is encoded in a dolby plug in or box and then decoded for playback, the same goes for 5.1 and 7.1.
 
Last edited:
I agree with Gary. This really should be done by an audio post facility to meet the specifications. Was your project mixed by a post sound person? Lots of times, sound effects aren't only in the left and right channels, they are also in the center with the dialogue...

Furthermore, remember that DCPs have a requirement of 24FPS (I believe this is still the case as of today). That means you will need to sample rate covert your audio if your timeline is not at straight 24fps.

This is common procedure for any film with a 23.976fps timeline. The sample rate conversion for the audio should be done by an audio post facility.
 
Gary is right but he is really talking more about a 35mm film release.

I will agree with everyone that if you are going to the trouble of doing a DCP which is only for theatrical projection then you NEED to have you sound mixed by someone who knows how to mix for the big screen.

As to specs. that really depends on a lot of things. There basically aren't any for a theatrical release. Broadcast and cable and streaming will all have specs you need to meet, and they will mostly/probably be different.

You don't need to do an M&E unless you have a distributor who requires it. For a low budget film you can probably do a trashy one to "satisfy" the requirement since it is not gong to used in all likely hood and it's only "required" because some bozo saw it on a list someplace. They are NOT going to spend the money to dub and remix your film. If they do then they are planning on making enough money that doing a proper M&E after the fact is not an issue. 99.9% of english low budget films that play in non english speaking locals are subtitled.

Plus you can't do a real M&E with out a "real" sound post and mix.

Unless spec'd someplace you are going to be delivering stereo. There is no LCR format for projection. To keep the discrete Center you would need to mix to a surround format like 5.1 or 4.0 (which is the ProLogic "Dolby Surround" format that Gary mentioned).

The catch there is that unless DCP is really happy with discrete tracks you start getting into proprietary encoding that you probably have to lay a license fee to use.

Gary is right about the requirements for Dolby Digital, though the bigger issue for you would probably be the $20,000 or so license fee and the $450 an hour or so mix stage fee (10 hour minimum, figure two days -20 hrs- to do the mix).

Now if DCP systems are perfectly happy with discrete tracks then you could do that for free. And you could to a LCR with nothing in the surrounds or sub. Basically a 5.1 with 2.1 of blank tracks.

In any case your mix as it exists might be OK for the small screen. Unless you had someone do the mix it's probably not great and might be actually quite bad. You could burn a DVD and hand it to some folks who will be honest and have nothing to do with the film and preferably don't know anyone in it and see what they say. Having screened and judged a few film festivals the odds are not in favor of a good mix.

But assuming you knocked it out of the park that small screen mix will not play anything like it does on the big screen.
Even experienced folks get fooled. I worked on a film that had these black outs between sections. I had put in sounds that lead into the next section so you got a beat of black and quiet then a few beats of sound leading into the next section. In the mix the picture editor, fairly experienced and a really good picture editor - who had been a big screen music editor, argued successfully to stick with silence in the black.

I liked the momentum of the prelap sound but on the small screen it really didn't make a lot of difference. Then I saw it on the big screen. It seemed like the black went on forever. Small screen there is stuff to keep your mind busy if the screen is black, on the big screen it's all you have. My instinct was right but I got talked out of it because I couldn't prove on the small screen that it was needed. If we had mixed on the big screen it wouldn't have been changed.

I have a bunch of those kinds of small to big stories so just take our word for it you NEED to have someone who knows how it will translate, or better mix in a proper mix stage or you will very likely have some rude surprises when you screen it.
 
I must admit that while I'm a happy Premiere user, I'd not consider it an appropriate tool to handle multi-track audio in anything other than basic stereo. I also avoid anything in any form of surround, as my studio is set up for simple stereo. Quality extra channels are damn expensive at the bass end, and the room is pretty critical, because you need to monitor in a room of adequate dimensions and audio treatment to make those ever so subtle decisions. I've been doing stereo for years, but have not produced one in a surround format that sounds good on other people's systems. Too damn hard without the kit and experience.

Gary's had too many years of balancing music and effects for TV, and he makes it sound simple(ish) I find it much harder when sounds start coming from different directions!
 
paulears said:
I've been doing stereo for years, but have not produced one in a surround format that sounds good on other people's systems.
Click to expand...

Or worse - the posh 5.1 mix may sound like crap on a laptop, tablet/phone, or anything with just stereo speakers.
I go with stereo for anything that goes to festivals around the world.

Heck Woody Allan made mono movies till 2011 - Midnight in Paris was his first non mono film.
Can't remember anybody coming out of an Woody Allan flick and saying "nice film but the mono track sucked".
 
Last edited:
Hi all first of all thank you everyone for the indepth responses and discussion. It has given me a lot to think about that is for sure! It's difficult I am a bit naive to exporting for sound especially and have no knowledge for theatre. I was just trying to figure out how other no budgets cope, i.e paranormal activity etc.

We've done extensive tests on the small screen with many TV's that range from decent audio set ups to rubbishy tinny ones and importantly dialogue, effects etc all sound pretty darn good but in relation to the big screen I'm sure like you all say it won't touch on it.

It's difficult as we are literally a no budget (you can see one of our pages here www.facebook.com/theprofesionalthemovie) so it will be almost impossible to get a decent post house to mix it. The only reason we are getting the DCP is good fortune, a friend of mine works quite high up at one, which i can't say for legal purposes, but they are willing to do one for free with the qc and everything and write it off as one of their tests, but as a result we can't really get much assistance. I was told to just send a copy of the film over when all is ready. I'm under no illusions the film will be released in theatres really it's more the DCP will come in handy for film festival screenings etc.

It's a bit depressing for a no budget filmmaker when you can't really afford a good mix for theatre! Though I'll look into post houses and see what qoutes I can get. In the meantime I might have to try and get a 2.1 stereo version as best I can then as suggested with the discrete tracks. How would one go about that though roughly? Like export the effects/music with muted dialogue in stereo, then dialogue with muted effects as mono? Then add both these tracks back into the project and select 5.1? Sorry for the ignorance but i really do appreciate all the help. Thank you.
 
For TV, DVD, On-line and even some cinema stereo can be perfectly acceptable and is the easiest route to delivery, add on an M&E seperate mix and you have delivery for international dubbing covered as well.

All of the other options are due to budget or delivery demands from those who are willing to pay for that service, that can be a 4.0 pro logic, 5.1 dolby digital or pro logic2, 6.1, 7.1 or even 22.1 but all them including the stereo option need to be balanced and mixed to suit the delivery medium.

Extra deliverables such as stem mixes for dialogue, music and effects may only be relevant for high budget 35mm cinema but I have also done them as standard for TV with the help of a decent digital desk such as the AMS Neve DFC etc and spooled them out to a DA88 or Akai DD8 etc at the same time as the M&E mix.

Everyone always thinks that 5.1 is the defacto for cinema or even HDTV but as others have said you need the dolby approved dub theatre and licencing to allow that to happen as their quality control is very high. I do very little in 5.1 as most TV just do not have the budget and indie cinema has even less money for it.

A stereo mix is more compatible though and for multi platform delivery at low cost is the best option and it can be played on virtually everything.

As for mono, it is actually a good way of learning how to balance as you have only one channel to deliver through and I spent the best part of my 35 year career mixing mono for TV on a single fostex self powered speaker with the main tannoy or dynaudio monitors for quality control.

You can also split the mixing using digital technology if you have virtual mix automation and do a dialogue pre-mix first without any other effects or music so that all of the dialogue is equalised and pre balanced. You could then do the same for the effects and music and then just have three stereo faders as the masters to balance out the final mix, of course you need the hardware and software to do all of this and that is why a good session in a proper dubbing theatre will give you the best outcome!

I have done fulll mixes at home though on a digidesign 002 with pro tools version 8.05 but as Paul said I have done this for many years so before I even fade anything up I have the set-up to deliver all my stem mixes and final deliverables in stereo. I do have 5.1 but it tends to be a burden rather than add anything substantial to most things.

Good luck and hope the charge card has enough credit! :Drogar-BigGrin(DBG)
 
taras4444 said:
It's a bit depressing for a no budget filmmaker when you can't really afford a good mix for theatre! Though I'll look into post houses and see what qoutes I can get. In the meantime I might have to try and get a 2.1 stereo version as best I can then as suggested with the discrete tracks. How would one go about that though roughly? Like export the effects/music with muted dialogue in stereo, then dialogue with muted effects as mono? Then add both these tracks back into the project and select 5.1? Sorry for the ignorance but i really do appreciate all the help. Thank you.
Click to expand...

Do you have access to Adobe Audition? You can export an OMF/AAF out of Premiere and import it into Audition and do a mix in there. You will be able to separate your dialogue and FX and print "stems" from there, which will be your discrete tracks.

What's frame rate is your timeline set to anyway? If applicable to you, will your friend at the DCP house convert your non-24fps timeline/audio/video to 24fps?
 
Hey thanks again guys and yeah i have access to audition (used it for fixing and doing all the audio so far). So i will play around with what you suggested and try a few mix options in there, see if i can export the stems there. Would I export the dialogue as mono then (as it was recorded) and the other two tracks in stereo from audition? I found this too http://strypesinpost.com/2012/11/exporting-multichannel-quicktimes-in-premiere-pro/

And mine is 25fps but they mentioned (and the festival also accepts) 25fps DCP as I shot it in the UK. So hopefully should be ok...
 
taras4444 said:
Hey thanks again guys and yeah i have access to audition (used it for fixing and doing all the audio so far). So i will play around with what you suggested and try a few mix options in there, see if i can export the stems there. Would I export the dialogue as mono then (as it was recorded) and the other two tracks in stereo from audition? I found this too http://strypesinpost.com/2012/11/exporting-multichannel-quicktimes-in-premiere-pro/

And mine is 25fps but they mentioned (and the festival also accepts) 25fps DCP as I shot it in the UK. So hopefully should be ok...
Click to expand...

Can you post the audio specs from the DCP house?

An example of specs would be something along the lines of:

Channel Layout -

1) 5.1 Printmaster (Left)
2) 5.1 Printmaster (Right)
3) 5.1 Printmaster (Center)
4) 5.1 Printmaster (LFE)
5) 5.1 Printmaster (Left Surround)
6) 5.1 Printmaster (Right Surround)
7) Stereo 2.0 Printmaster (Left Total or Stereo L)
8) Stereo 2.0 Printmaster (Right Total or Stereo R)
9) Mono Dialogue Stem
10) Stereo FX Stem (Left)
11) Stereo FX Stem (Right)

OR

1) Stereo 2.0 Printmaster (Left)
2) Stereo 2.0 Printmaster (Right)
3) Mono Dialog Stem
4) Stereo FX Stem (Left)
5) Stereo FX Stem (Right)
6) Stereo Music Stem (Left)
7) Stereo Music Stem (Right)

OR

1) Stereo 2.0 Printmaster (Left)
2) Stereo 2.0 Printmaster (Right)
3) Stereo Dialog Stem (Left)
4) Stereo Dialog Stem (Right)
5) Stereo FX Stem (Left)
6) Stereo FX Stem (Right)
7) Stereo Music Stem (Left)
8) Stereo Music Stem (Right)

... etc.

The above are just short/simple examples of what you might see within a more complex spec sheet.

Based on the above deliverables, you would actually export (or setup and route your tracks in Audition to compliment your delivarables) a variety of things from Audition so it really depends on what the DCP house wants... The U.S. TV network I'm working with now requires a mono dialogue stem and a mono FX stem (verses stereo or 5.1 for both). So it really depends......

You mentioned that you're already using Audition, so maybe half the battle is already done, and you just need to setup your Audition project/session in a way that satisfies your deliverables?

Glad to know the DCP house is ok with a 25fps timeline!
 
1) Stereo 2.0 Printmaster (Left)
2) Stereo 2.0 Printmaster (Right)
3) Stereo Dialog Stem (Left)
4) Stereo Dialog Stem (Right)
5) Stereo FX Stem (Left)
6) Stereo FX Stem (Right)
7) Stereo Music Stem (Left)
8) Stereo Music Stem (Right)

From above this is what I would generally deliver with a picture master that also has Final mix on tracks 1+2 and M&E on 3+4.

All the stem levels would be balanced as per the final mixes so that if the content needs re-versioning the editor has everything to hand.

Following on the conversation I use FCP for my in house picture editing and will generally do most of my audio track-laying in the edit, I then export an OMF of the tracks for use in my Pro Tools for final finishing and then produce the delivery masters as per above and re-import them into FCP so that the final edit time line has all the final mixes for versioning.
 
As to the money issue. You got a camera, a DP, actors, etc. Apply the same strategy to sound. You don't have to pay scale for sound. You need to sell someone on doing sound you can afford.

Big points on your small screen QC. From what you describe you are probably in pretty good shape for the small screen.

There is not a lot of reason for most films to be in surround. Having worked on a number of commercially released 5.1 films I can tell you that all too often they are played in stereo. Streaming is in stereo, a lot of folks watching on the small screen will be watching in stereo... So for a low budget stereo is your gold standard.

Unless the DCP requires STEMS you do not need to try and create them.

They are a usual step in big budget films and the main reasons are #1 for dubbing into other languages #2 to be able to make alt mixes cheaper. A big film will often to alt versions for airlines (lots of language restrictions) and broadcast (less stringent language restrictions). Having the dialog separate makes it easier to dub in a "darn" to replace a word not allowed.

Neither of these reasons really apply to no-budget films. IF by some miracle your film gets a foreign distribution they are not going to pay to do a dub, they are going to subtitle. The only lower budget films that get dubbed are ones dubbed into english because americans have an allergic reaction to subtitles. The chance of someone wanting airline or broadcast versions is just a bit lower than your chances of winning the Irish sweepstakes.

So I would not put in the time now to do it. It is actually a complicated thing to do. It's not just separating the dialog from everything else, and that done properly is a LOT of work since done right it is JUST dialog, no BG, no FX, nothing. It sound sound like it was done in a vocal booth. Every non vocal sound needs to get moved to PFX tracks, all non principal dialog needs to get moved to BG vocal tracks and then you need to do a full Foley pass to cover everything you can't move like cloth. If you just isolate your production tracks then the Stems are practically useless.

Since I'm thinking your small screen mix is probably OK, you can get a decent idea of the big screen mix by talking a local movie theatre into letting you screen your film through their system during down time. If you get an owner OK you might be able to get the projectionist to do it for a few beers. Ideally you would like to do two checks, one in stereo and one through the "Dolby Matrix" with out noise reduction.

If you can only do the stereo that is fine, probably the DCP won't get routed through the Matrix, but that might also be a reason why, if you can deliver discrete, that you would deliver a "5.1" that only has L & R since 5.1 will never go through the Matrix.

A really short simple explanation:
The Matrix is a way of sticking 4 tracks into two tracks and then extracting it back to four tracks on playback. The problem is that the way it is done will sometimes take a plain stereo track and toss odd sounds into the surrounds because it can get fooled. When you create a matrixed mix (a LtRt) you have to monitor the decoded version so you can catch the sounds that have decided to wander and corral them back to where they should be. Mostly this problem will happen with music but...

Pretty much all films in "stereo" are run by default through the Matrix so any 35mm film that is stereo really needs to be checked. I'm just not sure if the stereo file on a DCP would be assumed to be a LtRt, since really all recent stereo tracks on major films would be, or if it can be flagged as a true LR and so always be played direct.

If I were you I would find the nearest facility that does theatrical mixes and see if you can find someone willing to lend you a hand. If it's a bigger studio the mixers might be unreachable but a "mix tech" (american term, don't know what the english equivalent would be) is probably your best bet. It is also fairly common in the US that mixers and mix techs can get time on a stage during down time for projects they want to do. Even editors can sometimes get a project in. I was owed some favors and got a no budget feature a two day final mix on the down days of the Godfather REDUX mix. I helped out at Skywalker, along with a bunch of other editors, on a documentary someone brought in that ended up mixing on the same stage Titanic had been on a year before. A low budget short I worked on got a free private screening at a local theatre to check the sound because the director asked and they said fine. Dolby labs in SF lets filmmakers screen in their screening rooms for free and you can do a whole "opening Gala" there either for free or Very close to it.

Especially if your film is good and you are serious about filmmaking (as opposed to doing it for fun) a lot of places and people will be willing to lend a hand. A) they love film or they wouldn't be doing it and B) you might be someone who "makes it" and then it was a good business decision to lend you a hand. Plus it doesn't cost them much besides time. So as long as you can work around their down time and be really flexible (and grateful) you will be amazed at the help you can get.
 
Hey thanks for all the messages again very helpful indeed and I appreciate the time it takes to respond to them all. I've just moved over to a small town outside of Vancouver so I might follow your advice and see if I can get the small theatre close by to play my film so I can test the stereo sound, and also look into getting someone involved in the post sound for the best possible results (I managed to convince the actors!)

Still a little ignorant/confused as to the specifics of certains things though. Was experimenting exporting an OMF and had two options, one to export each audio seperately and the other together, I guess I should choose seperate?

And for example if i followed the following layout 1) Stereo 2.0 Printmaster (Left)
2) Stereo 2.0 Printmaster (Right)
3) Stereo Dialog Stem (Left)
4) Stereo Dialog Stem (Right)
5) Stereo FX Stem (Left)
6) Stereo FX Stem (Right)
7) Stereo Music Stem (Left)
8) Stereo Music Stem (Right)

Would I export and have my DNxHD master picture and then these 8 tracks in it (so if i imported it into premiere after id have automatically the eight tracks and one video track? Or would I export my DNxHD as stereo and then do each stem seperate?

For the time being I may avoid the stems until I am asked and try and export the best audio tracks for viewing in the cinema/sound booth. If hypothetically everything was mixed and it was just a case of exporting, how would I then go about that? When I export my DNxHD I get one audio track with it which is stereo and included all my tracks together. However, I saw somewhere about opening a new sequence and selecting the tracks you want (ie i could then select four tracks possibly) and then pasting everything in there and trying to get all the audio into that.

Or alternatively (if everything was mixed hypothetically) should I export all my dialogue tracks (and nothing else) as a wave form, then do the same for all my music, then effects. Then reimport these into a new sequence with my footage so I have four complete tracks and not all the seperate ones? Then export this DNxHD so I have the one video track and fou audio tracks and take this to the cinema?

Sorry if I am just going around in circles as you can see I'm still a little confused as to the actual exporting side of things! Not sure if this would be ok...
 
For the OMF export - just do what works for you. For big budget feature films, we request to have separate audio files (not embedded in the OMF). There are other technical reasons for doing it this way, such as maintaining BEXT/iXML metadata but I don't think that should be of concern for your project.

For exporting your DNxHD master picture, with or without audio in it, I'd consult the DCP house. The DCP houses I work with actually request the audio separately and not married to any picture file. They usually just want the RAW .wav/.aiff tracks.

I could be wrong about this, but doesn't the latest version of Premiere CC have the option to "map" the audio layout when you Export the timeline. It was highlighted in their "newest features" release before the October update to CC.

To answer your workflow question, it roughly goes like this:

Production -> Picture Edit -> Locked picture (hopefully) -> Send to Post Sound via OMF/AAF -> Sound Editing -> Final mix -> Sound folks send the final mix files (whether it be stereo or 5.1) back to picture editor -> picture editor lays back these files into a sequence only containing the final mix files -> Export master video file (with or without audio per deliverable specs)
 
"The DCP houses I work with actually request the audio separately and not married to any picture file."
- Same here in NYC, most layback is done with just a PCM stereo or poly file(s), which is almost fool proof when the picture is locked.
Regarding 5:1, if not done correctly, a down-mix can have disastrous results, and most will stay away from the format altogether, unless the audio can be auditioned at the theater prior to screening.
 
Ok awesome guys thank you. I'll see if I can get more info from my buddy at the DCP house. In the meantime for a theatre check I will try and export the DNxHD picture. Then export all the dialogue as one wav track (should I choose mono for this then as that's what it was recorded and edited as?) And one music track and one effects track both in stereo.
 
taras4444 said:
Then export all the dialogue as one wav track (should I choose mono for this then as that's what it was recorded and edited as?) And one music track and one effects track both in stereo.
Click to expand...

I think there's still some confusion. When you export the dialogue as one wav file, then export just the music track as one wav file, then export the fx track as one wav file, this is referred to as exporting/bouncing/printing "stems". In your previous post, you said, "For the time being I may avoid the stems until I am asked..."

The question for you is: Are you required to do this for the DCP?

If you are not required to do the above, then read below:

The (normal) way that you would deliver the final mix is by "mixing" everything in Audition. When you are happy with your mix, you would export dialogue, music, AND fx to one stereo .wav file. This stereo file is a "composite" mix of all dialogue, music and effects. You would take this single stereo wav file and import it into Premiere. Duplicate your master sequence (and rename it to something like Final Mix). In the duplicated sequence, unlink all the audio regions from the video regions, select all the audio, then delete all the audio. Then drag the stereo .wav file that you just imported onto a track. Then export this new sequence to whatever you need it to be (DNxHD).

Note: I don't think you need to export/print/bounce "stems" for the DCP house. They probably just want a final "composite" mix. If the DCP house specifically tells you that they need dialogue, music, and fx stems, then that is a different scenario.

So... can you post the audio specs provided by the DCP house?
 
You must log in or register to reply here.
Back
Top