Jump to content
Bubu

Putting new mix tracks into a Polyphonic recording

Recommended Posts

Hi all,

I´m have a "small" problem.

I have quite a few polyphonic audio files (22 shooting days!)  witch have no Production Mix tracks (Boom mix and lav mix), I have a long shot way of doing this in Wave Agent and ProTools (Bounce new mix files in PT and Combine that with the orginal in Wave Agent later.

The problem is that the editors want the tracks in a specific order that I can´t seem to be able to change, making the files in wave agent. Although the option seems to be there, the order always stays the same.

So I guess my questions are:

Has anyone had to do this as well?

Can you guys think of any other programs that could do this without wiping out track names and other metadata?

 

Thanks you!

Björn Viktorsson.

 

Share this post


Link to post
Share on other sites

Thanks for the replies!

You can change the order of the files in PT, then export them an OMF or AAF.

p

 

Exporting Omf´s and AAF´s is not an option with the editors, they want new polyphonics, I´m basically trying to find a good workflow to replace the polyphonic orginals, put the mix tracks in there and export,  but keep all the metadata.

You can also open the tracks in (free!) Audacity, re-order, and re-export.

Can´t find a way to make a polyphonic file in Audacity and exporting strips the files of their scene, take and frame rate metadata witch is essential!

 

Edited by Bubu

Share this post


Link to post
Share on other sites

Input the files from a DAW back into the original recorder with a timecode feed.  Remix like it was live, but this time recording the mix.

Some recorders allow chase lock to timecode too.  Otherwisr metadata should remain more or less the same with exception of TC start / stop times.

I'm curious how this problem originally came to be.

Share this post


Link to post
Share on other sites

Input the files from a DAW back into the original recorder with a timecode feed.  Remix like it was live, but this time recording the mix.

Some recorders allow chase lock to timecode too.  Otherwisr metadata should remain more or less the same with exception of TC start / stop times.

I'm curious how this problem originally came to be.

Thanks!

Interesting, this can be a solution if it chases to timecode, otherwise not really, we need the TC. 

I could not use my 663 on this one so it´s really a matter of finding the right recorder with 8 digital inputs and the chase to TC option i guess.. 

Any Ideas?

The mix files where not recorded with the original polyphonic recordings!

 

Share this post


Link to post
Share on other sites

I know that Nagra VI offers this feature (chase lock) but only has 2 AES inputs for 4 tracks.  I think the Aaton Cantar does this too, but never tested it.

Share this post


Link to post
Share on other sites

Recording and mixing all of the audio of 22 shoot days again is quite a task. I am curious, too: why was the mix track not recorded? 

In any case, I think you should tell the editor that there is no mix track. Explain the reason if you can, but that's it. They can get by without a mix track much easier than you can re-create one

Share this post


Link to post
Share on other sites

Not knowing all the particulars on this I'm going to say that Constantin has it right.   Post can creat a mix much easiier then you will be able to. Is it that you are afraid to let post know the situation? You will be amazed at how a big issue in your head is a little issue in the heads of those in post.  In short, We tend to worry about things that isn't a big deal to post.  Be up front with them before you start frankensteining the tracks you have. You'll save yourself a lot of trouble.

Share this post


Link to post
Share on other sites

Hi Bubu!

My MixInjector App for OSX does just that! You may choose which ISO tracks to combine to create one or two mix tracks. You may also choose the scaling factor for the mix. The app is recursive, so you could choose an input folder containing dozens of daily subfolders and the output folder would contain an exact replica of the input folder structure but with mix tracks added to channel one (mono mix) or channels 1 and 2 (stereo mix) of the BWF files, shifting the ISO channels back by one or two channels. MixInjector maintains all Metadata of the original files. You may either inject the mix tracks into the files or generate new files containing only the mix track(s).

Of course, this will not produce a good mix, but rather a rough/guide mix for picture editorial. It works extremely well though if the ISO tracks were recorded post-fader on the set.

I wrote this App for my own needs a few years ago when I was working with SD744T and Sonosax R4 recorders. I just tested it on OSX El Capitan and it still works.

Give it a try, i've published it under the Beerware license, so it's free (as in beer)!

You may download it here:

MixInjector v2.9.637

Cheers,

Jürg

Contents of the Help Window:

Welcome to MixInjector!

This unique tool is designed to batch-inject mono or stereo Mix Channels into WAV audio files while maintaining Metadata such as project, scene and take names, timecode, time and date, channel, reel and tape names, in fact any iXML or BEXT Metadata embedded in the input files. The Mix Channels are the result of a mixdown of any combination of the input files' audio channels.

MixInjector is primarily intended for Film Sound Recordists who like or need to work with small lightweight multitrack field recorders which do not feature built-in mixing facilities, such as the wonderful Sound Devices 744T and Sonosax SX-R4 recorders. As the track count on these machines is limited, the user may prefer not to record Mix Channels during the shoot and rather rebuild them at a later date, usually right before the film is sent to the cutting room. That's where MixInjector enters the game.

MixInjector processes WAV files with embedded Metadata, such as files recorded by the vast majority of professional field recorders, as well as sound files without embedded Metadata recorded by most consumer-grade recorders, such as the Zoom H4n Handy Recorder or the Sony PCM-D50 Linear PCM Recorder.

MixInjector can handle WAV files with up to 16 channels and will ignore any other file type present in the input folder.

The <Recursive Processing> mode processes all audio files in the input folder's subfolders as well. It does this by replicating the input folder's folder structure in the output folder.

The hidden <Process In Place> mode overwrites the input files with the processed files for the processing modes <INJECT MONO> and <INJECT STEREO>. To enter this operating mode, either press and hold the <ALT> key on your Mac's keyboard while launching MixInjector or press and hold the <ALT> key while clicking on the input folder selection button once the application is running. Please make sure you have a backup of the original files in the unlikely event that something goes wrong!

You may choose which channels of the input files are mixed into the respective Mix Channels by clicking on the green channel buttons. Clicking on the blue <LEFT>, <RIGHT> or <MONO> buttons alternatively selects all channels or no channel.

You may also choose the scaling factor for the mixdown of the channels. "Linear" means that when a Mix Channel is mixed from multiple (n) input channels, each input channel will be scaled by a factor of 1 divided by n. This method has the advantage that there's no way the Mix Channel could distort, but the overall level of the Mix Channel will be quite low. "Square" means that when a Mix Channel is mixed from multiple (n) input channels, each input channel will be scaled by a factor of 1 divided by the square root of n. The advantage of this method is that the overall level of the Mix Channel will be louder, but could occasionally result in distorted audio.

A Log File with the detail of every processing step is saved in every output folder for troubleshooting purposes.

The processing may be cancelled at any time by pressing and holding the <ALT> key for a few seconds.


Hover the mouse pointer of your Mac over any interface element to get additional help.


Processing Modes:

Inject Mono:
Injects a mixdown of the selected <MONO> channels of the input files into the first channel of the output files, shifting the original channel allocation of the input files back by one channel.

Inject Stereo:
Injects a mixdown of the selected <LEFT> channels of the input files into the first channel and a mixdown of the selected <RIGHT> channels of the input files into the second channel of the output files, shifting the original channel allocation of the input files back by two channels.

Extract Mono:
Creates mono output files containing a mixdown of the selected <MONO> channels of the input files.

Extract Stereo:
Creates stereo output files containing a mixdown of the selected <LEFT> channels of the input files on channel one and a mixdown of the selected <RIGHT> channels on channel two.

 

Screen Shot 2015-10-21 at 12.31.52 .png

Screen Shot 2015-10-21 at 12.49.12 .png

Edited by jlempen
Added screen shots

Share this post


Link to post
Share on other sites

Thanks Jürg that's brilliant, when I'm next in Lausanne, I'll be happy to buy you as many beers as you want......

Now, as you obviously know what you're doing with all this including timecode and metadata etc. I dream of Sound Devices employing you for a very handsome fee to improve, and finally make Wave Agent the definitive production sound manipulator.....:)

Share this post


Link to post
Share on other sites

Is the intent to auto sum the ISO's and call it a mix track?  Where's the mixing?  That'd just be a sum track, and having those tracks on the edit timeline would do the same thing every time they are played back.

I must be missing something, I don't see how it's worth anyone's time to remix 22 days of production audio before the edit, when it would take much much less time to just mix the audio used in the edit.  I also don't see the benefit of having a "mix" track that's just a sum of the iso's.  

There's obviously something I don't understand about the workflow going on here.

Share this post


Link to post
Share on other sites

Hi Wandering Ear,

You are absolutely right, the MixInjector App's purpose is to batch inject auto-sums of selected tracks into one or two channels of polyphonic WAV files.

This won't create proper "mix" tracks of course, but it will provide the editor with files containing "mix" tracks if these tracks haven't been recorded on the shoot. I wrote this App for my own needs and for a very specific workflow a few years ago. Let me explain:

A few years ago I used the Sound Devices 744T recorder (paired with a Sound Devices 302 or MixPre to control the inputs 3 & 4) extensively for documentaries. This machine is excellent, but has not been designed to do real-time mixing in the field. Furthermore, wasting two of its 4 (744T) tracks for "mix" tracks would not allow me to record all the mics on their own ("ISO") tracks. As the 744T is not what today they call a "mixer/recorder", there's no way to record the inputs pre-fader on ISO tracks and use the knobs as faders to create real mix tracks. The knobs are in fact gain knobs. Every input is recorded "post-gain-knob" on a track.

Now I know many (usually older) picture editors who don't want to work with ISO tracks in editorial, but only work with two mix tracks.

So what I would do back then was to monitor a sum of the tracks in mono and set the relative gains of the mics in a way that sounds just about right to my ears (the "mix"), but I would not record that mix on its own tracks because of the relatively small track count of the recorder.

Then I would use my MixInjector App to re-create these mix tracks in the files before handing them over to the editor.

This trick allowed me to work let's say with a boom mic and three radio mics on a 4-track 744T but still deliver every mic on its own track, as well as a two-track mixdown, thus expanding the 744T by two tracks, making it a virtual 6-track recorder.

I've also seen quite a lot of people use the same workflow with the Roland R4 recorder or even with the Sonosax SX-R4. The latter actually is an 8-track recorder and has 6 analog inputs with a routing/summing matrix, so on the SX-R4, you could actually record 6 inputs on their own tracks and record two summing tracks, but I've seen many sound recordists who for whatever reason didn't record the summing tracks on the shoot and were very happy to be able to re-create them with my little MixInjector App.

Maybe I should have called the thing "SumInjector"...

But anyway, the app does something no other app does and it has proven itself very useful for myself and a few fellow sound mixers over the years.

Cheers from sunny Lake Geneva,

Jürg

Share this post


Link to post
Share on other sites

Pindrop, Christian and Wandering Ear,

Thanks for your kind messages!

I already shared the App on this board about a year ago in some thread and have received very positive feedback from the few people who downloaded it.

I actually quite enjoy writing code in my free time, and I also have a few other little helper Apps for location sound people in my projects folder. The problem is that I don't have much time to code them properly, maintain them, adapting them to new OS versions and API's, bugfixing, etc., making them ready for prime time and publish them. But hopefully I'll find that time some day, and then I'll probably make a website for this collection of production sound tools. At one point, I even thought of selling them for an adequate price, but then I'd need to stand behind my products and give timely support, which is quite impossible when you are shooting in a remote location without Internet.

There are mainly three other Apps in various development stages I'll share with you guys at some point:

- MidSideTool : a batch converter for stereo WAV files. Converts XY stereo to MS and vice-versa. This App also converts WAV files without Metadata into proper BWF files with embedded iXML and BEXT Metadata, thus allowing you to record atmospheres and wild tracks in stereo XY let's say on a Sony PCM-D50 handheld recorder and deliver MS stereo files with a timestamp based on the RTC clock of the recorder. The Tape Metadata is based on the creation date of the files, and the user may set default Track names and a default Note, Wild Track and Circled tags, Project and Scene names, Originator and Originator Reference, as well as the Frame Rate. MidSideTool runs stable but still has a few error handling issues I need to fix.

- MovieSlateInjector : batch injects data from a MovieSlate XML file into BWF files. Allows the user to manage all Metadata in the iOS MovieSlate App and inject this data into the files. This is a tricky App to maintain, because it relies on MovieSlate XML files which may change from one release to the other. I've successfully used this App on several TV series with MovieSlate V7.x and it works perfectly as long as you are very consistent in maintaining your MovieSlate report vs the recorded files. The App has a nice feature in case there's a mismatch between the MovieSlate Data and the recorder's Metadata, where a window with an integrated QuickTime player opens to let you hear the problematic take and decide which data is correct. This App did run fine on OSX Mavericks and MovieSlate v7.x, but I'll have to do some very thorough testing with the latest OSX and MovieSlate v8.x.

- MetaInjector : a tool to very quickly review your dailies and correct and complement Metadata. This one is kind of a sister App to Sound Devices' Wave Agent. Where Wave Agent shows you a list of all the files in one or several folders and lets you batch-edit your Metadata, MetaInjector sequentially loads the files in a folder and immediately shows all Metadata of the current file in a single window while playing it in a multitrack audio editor, allowing you to listen to the take, solo single tracks, etc. As soon as you press Enter on the keyboard, it loads the next file until there's no files left in the folder. MetaInjector is still in development and I have yet to decide how to handle the player functionality. I'm not skilled enough to program a graphical multitrack player myself, so I need to rely on other products which have to be scriptable. Felt Tip Sound Studio works almost perfectly and is scriptable, but their scripting support is broken in some versions. Audacity is not (yet) scriptable, but that would be a fantastic Open Source player for this App. Maybe I'll start off with a simpler version relying on the deprecated QuickTime API, I'm still not sure on how to handle that.

Well, I'll let you guys know how all these projects evolve,

Cheers,

Jürg

Share this post


Link to post
Share on other sites

Hi Jlempen!

Thank you! Thank you! Thank you!

and a few more times thank you!

And i´ll buy you beers any day!

This has saved me from a long and boring days with PT and Wave Agent!

And yes the case was that here in iceland we rarely use the mix tracks anymore. And since they where not asked for in the spec&workflow documents, I decided not to record them. Probably a mistake in this day and age..  I´ve felt that they get in my way when I do sound post. Saying that the track naming in the recorders has become much better in the recent years so these things should not slow you to much down in post. Anyway this is a Danish co-production and the editing is done there by some old timers :) , and they said that there was no way for them to work with it like this. It turns out that when making the Spec&Workflow document they forgot to mention that little thing about the mix tracks. 

And I guess every country is different, here on a film project, there is rarely done any on location mixing so the mix track culture hasn´t caught on here yet.. I guess i´ll be recording them from now on!

Really like the other ideas you have though and would be very interested in following that in the future if you keep on developing them. Cause although Wave Agent is i good program it is far from perfect.

Another thing I also had to do for the editors was to combine multiple recorders into one polyphonic file for each take, when we had that situation, be it another sound devices or sometimes we put a mic on the Alexa (it was all long handheld takes, with quick 360 turnarounds). That I did with Wave Agent and was a bit tedious and WA proved to be faulty when keeping tracks in the desired order above 9 tracks. 

Thanks alot Jlempen, i´ll make sure to get you on the thank you list for the movie!

If you swing by Reykjavík stay in contact and you´ll get those beers!

Cheers!

Björn V.

Share this post


Link to post
Share on other sites

Hey Björn!

I'm truly glad my App saved you a few days of tedious and boring work! That's exactly why I wrote the thing in the first place! :-) As I'm planning to travel to Iceland for holidays probably next year, I'm looking forward to having a few beers with you in Reykjavik!

Around here, recording two mix tracks has been the de facto standard since the DAT-days. And I actually like the fact that many editors still prefer to work with my mix tracks and not mess with the ISO tracks, which makes for a cleaner and clearer timeline. After all, my mix usually sounds more or less exactly how I wanted the scene to sound, but still allows the editor to change the balance between boom and wireless mics to some degree. And if I fu*** up the mix at some point, they could always import the ISO tracks for that take and play around with them.

You are absolutely right about the fact that they never include this information in the specs & workflow documents. I got burned once on a production for not providing mix tracks and I've recorded them by default ever since.

I've had to merge files from multiple recorders only once, but I recall that it was a tedious task in Nuendo. Maybe I'll write an App for that some day! Should be quite straightforward if the recorders were synced by TC.

Cheers from sunny Lake Geneva!

Jürg

Share this post


Link to post
Share on other sites

Is there a default order to how most want their mix and poly files arranged and named? I'm just starting to work with higher quality recorders that will allow this control and am wanting to be confident I am following the industry standard if one exists. Thank you for sharing these helpful apps.

Sincerely,

Morgan

Share this post


Link to post
Share on other sites

The most widely adopted standard, at least in the US, is a mix on track 1 and iso discrete pre-fader tracks on 2, 3, 4, etc., delivered a polyphonic bwf. There are some who are delivering 2 mix tracks, on track 1 and track 2, but this less common (again, only referring to common practice in the US).

Share this post


Link to post
Share on other sites

Beyond what Jeff's already stated, I try to think like the folks who will have to deal with my tracks and as a result go for Consistency. That's a capital 'C' Consistency. Deploy TRX from low-numbered to high-numbered as with character numbers on the call sheet, and so with the tracks. Consistently.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×