When there isn’t a budget to hire out help, where do you start? Composer Walter Everton walks through his experience working on low budget Indie post-production projects. Being at the start of his career, he hasn’t had the fortune of finding the projects with really any budgets, but what he has found are dedicated and passionate people who want to create. In this article Walter shares how he has dealt with the process.

From Music To Post

I’ve had to learn the hard way how to best deal with independent film/TV/Web Series Audio Post. As I’ve expanded my career from the recording (music) side of things I was focused on in University, I’ve needed to learn through making mistakes to find a workflow that works best for me. What I hope to share here is how I deal with small budget (to honestly no budget) independent projects, and how I have taken things I have learned from my education as a music recording focused audio engineer to a Post focused engineer. 

Indies Don’t Always Consider Post In The Planning Stages

So far all of my projects have come from being asked or hired to record location sound.  Once the principal photography is done, I take the project through post-production as well.

This has had its benefits, however, from a purely post workflow standpoint.  It means that the ‘producer’ wasn’t really considering how they wanted to deal with post-production during principal photography.

I understand the need to focus on the task at hand and the rather monumental project that principal photography typically is. However, this would indicate to me that the typical creator of a small/low-budget indie project hasn’t really considered the post-production stage when I usually get involved in the project. 

As such, if you are only involved in the post-production part of the workflow, you may inherit a mess. So, if you are given the opportunity in at least one pre-production meeting or you are in some sort of communication before principal photography starts, make sure they are considering a few points:

1. How are you planning to sync audio?

   • Do you have a proper location recordist?

   • Use timecode if at all possible.

   • Ask the location recordist to label all takes appropriately with scene and take-number.

   • If wild takes are done, label them properly and document what they are (I like to tell my future self, usually at the end of the wild take, what it was for through the slate mic on my Sound Devices recorder). 

   • Confirm they are using a slate, and make sure whoever is slating on the set is calling out the scene and take-number, as well as showing the camera properly the slate, i.e., the person slating confirms with the camera they can see the ENTIRE slate in frame with enough space around the slate for it to move and so the sticks are seen when they are clapped. Then, when either a director and/or producer calls for camera and audio to record and each team verbally confirms they are recording, the person slating calls “Scene 1 Take 1” and claps the sticks together.

     • Hopefully, the reason I want to emphasise this is obvious. As the post-production person (and likely the only one on the project unless you are very lucky) this will save a huge amount of time and a disaster of a project if you make sure that everyone is onboard with properly syncing and slating on set.

     • No one wants to figure out sync after the fact, but in case you have to, and timecode doesn’t work for some reason (most likely someone either doesn’t understand it, i.e. has never worked with it before, or something happens on set that prevents TC from being recorded properly), slating will let you do sync manually without too much lost time. 

2. If they aren’t using a location recordist, ask them what their plan is for audio. If they don’t have one, urge them to hire/acquire a Location Recordist to deal with that for them. Again, it will make your life much easier once principal photography is done. Mike Thornton’s advice is that if you are being asked to do the post on a project where they don’t plan to have a location sound recordist, walk away very quickly.

I have found in my (admittedly) limited experience, half the job of a location recordist is making sure the audio files are named correctly so they can easily be found in Post. Of course, the other important issue the location recordist deals with is making sure the recordings are of a reasonable quality so when you work on them during Post, there are fewer things to fix.

My Post Production Workflow

Speaking of Post. The process I have found most effective is as follows: 

1. Wait for the picture-lock 

2. Receive picture-lock, and an AAF from the video editor

3. Do the dialog (Dx) edit

4. Import the Dx edit into the full mix template, and print the Dx edit for Music (Mx) edit/composing

5. Do the Mx edit/composing.

6. Confirm the Mx edit/composition with the Director/Producer before sending that to the full mix session.

7. Import the Mx edit into the full mix session

8. Do the sound effects (Fx) placement/foley fill in the full session

9. Do the full mix, and get approval.

10. Repeat this step until you get a mix everyone is happy with.

11. Print down the full mix and stems, and deliver to video.

Notes On My Post Production Workflow

Burnt In Timecode - I do tend to have issues with getting picture-lock from video. These issues tend to be either there is no timecode burnt into the video, the timecode starts at an absurd location (i.e. projects that start at 00:00:00;00 are not a great idea because you can’t have anything before 00:00:00;00 in Pro Tools and I tend to use the area around the picture to edit and have space to work on things, both before and after the picture location), or later on in the process inevitably someone decides to change the picture in some aspect and a re-conform has to be done on my side which is a headache. In case you are not sure 00:00:00;00 = hh:mm:ss;ff (h = hours, m = minutes, s = seconds, f = frames).

This brings me to my first suggestion…

Start At 01:00:00;00 - When you are doing your initial consultation with the project team (whether that be the producer, director, or video editor) ask to have the first frame of the Picture start at 01:00:00;00. This means that any leaders or pops need to happen before 01:00:00;00 (and what is most typical is to have a countdown in Picture with a pop 2 seconds out from the first frame and black after that 2s pop, meaning you’ll have a 2 pop at 00:59:58;00).

2 pop (sync plop) is an industry-standard 1kHz tone that is one frame long. This is used in many scenarios but most importantly it is there to confirm sync with Picture through a leader. The 2 pop is placed 2s before the first frame of Picture where there should be a 1 frame visual 2 pop too.

In Pro Tools, you can create your own 2 pop easily by first selecting 1 frame in your session then using the shortcut Shift + Control + Option + 3. After this, I tend to like to have a 1 second long file so I’ll then select the number of frames needed for 1 second in the project and use the shortcut Option + Shift + 3 to create a new 1 second clip with the 1 frame 1kHz tone at the beginning of it.

Confirm The Frame Rate - If you can’t tell by now, sync is incredibly important. So while you are in these initial stages, confirm with the video editor what frame rate they are delivering you picture lock at, and make sure your session is set appropriately. The other very common issue when receiving picture-lock is that I don’t know what the frame rate the picture is in and so I have my session set to the wrong frame rate or something frame rate related isn’t set properly or is imported incorrectly, the picture becomes more and more out of sync with audio in Pro Tools. If you find progressively worsening frame rate drift, this is indicative of a frame rate mismatch somewhere. Whether or not that is in Pro Tools will take some investigation. 

Getting It Right Makes Life So Much Easier

If everything is done properly beforehand and there is clear communication between you and the video editor, sync really isn’t an issue you should need to spend a lot of time dealing with.

Usually, you just need to import the AAF and video file into your DX edit session. Once they are in place, give it a watch through with the mix that is in the video file. The video editor will likely have some sort of mix, and this will be a good guide for what you need to do.

This also helps if there is any extra Dx laying around seemingly randomly. Its more than likely they used some off-camera dialog and comped it into a different take for creative reasons, and this is something that you obviously need to replicate just with more finesse.

Do A Dx Noise Reduction Pass

Since you are likely the only person working on this in your Dx edit, I would noise reduce the more problematic files at this point seeing as you need to comb through them anyway.

I have learned that in larger production the Dx editor won’t typically noise reduce the files at this point, rather leaving that decision on how best to deal with it up to the re-recording mixer.

Since you are the Dx editor and re-recording mixer, in my experience, it is far more convenient to deal with noise reduction at this point, and it will save you time not needing to deal with creating large handles for your Dx edit that would otherwise need to be done if you were handing it off to a different Dx mixer. This of course is dependent on how secure of a “picture lock” you get.

If any big changes might come down the line later, it could very well be worth it to go the old fashion way and create handles for each bit of the Dx. This is very time consuming, so it only saves you time if you expect the picture to change a lot once you have “picture lock”. 

The Dialog Edit

The aim of the Dx edit is to choose the best option(s) between boom and lav mics for the dialogue, to make sure everything that is said has the most appropriate mic(s) chosen for that moment.

For example, if you are watching a scene where the person speaking is inside a car, but the camera isn’t, there is a choice to be made. Is the intent of the scene to be spying on the person in the car, where a more natural sound like the boom would pick up better, or are we supposed to be in the car with the person in their conversation, in which case the lav would likely be a better choice.

Even if a more natural sound (per the location of the camera) is warranted, you will still need to make sure the intelligibility is there, so what is said can easily be understood.

Sort The Levels

Once you have chosen all the different takes and you have the required dialog intelligibility, the next step would be to get the levels generally in the same region of loudness so things aren’t too quiet or too loud. This isn’t to say whispering should be as loud as normal speech or yelling, but you don’t want to go from one scene to the next and go from barely being able to hear the dialogue to the dialogue being way too loud.

Consider Using A Loudness Meter

I use clip gain to make this adjustment, and I also have a loudness meter up (like iZotope’s Insight) to have an idea of where I am in the spectrum of loudness. It also helps for this step to be in some sort of calibrated environment so you have a known output volume/value that will let you consistently hit targets of loudness without you needing to watch a loudness meter. This will save some time for whoever is mixing it (which will likely be YOU), so taking care of this in your Dx session is just something that will make your life a little bit easier before you get into the complexity of the main mix session. 

Once the Dx edit is complete, I import the session data from the Dx edit session into my main mix session and create a mono Dx guide so I can start working on music.

My naming scheme for my sessions is as follows:

• For the main mix session: ProjectName_DxMxFx - WE Mix

• Dx edit session: ProjectName_DxEdit

• For the Music session: ProjectName_Mx - I will also append the session if I am writing music or not, if I am I add Scoring to the end, or if it is just editing then append MxEdit

Try to be descriptive so you know what the session is, and anyone else who looks at the session/folder/file structure will know what they are looking at.

Music Next

With the Dx Guide created, I  create/open my music session, import the picture (either through Import Session Data of the video track from the Dx Edit session or through video import of the video file), and import the Dx Guide I created in my mix session (via audio import).

Once this is created, I start working on editing and writing music. For much more detail on this check out my article How To Compose Music In Pro Tools.

Once all of the music is written, or close enough, I print the music tracks via my Mx Template I’m working in and create a file the same length as the video file.

I then import these newly printed tracks into my mix session using Import Session Data and fix the routing so they go to the correct location.

Also in this step, I import any sync music that I was given to edit in my music session.

In this case ‘sync music’ is songs from other artists, that were either written specifically for the show or were chosen by the director/producer to use in the show. That is to say, I didn’t write these songs, and so have little to do with them other than edit them as required.

For example, in the web series I mentioned earlier, I was both composing original music for the show and incorporating source music that they wanted to use in the show. There were some complicated edits in the source music that I made in my Mx session, and I knew they weren’t quite final so I brought them into my main mix session with Import Session Data to preserve the edits and make it easier to re-edit further down the road once we got to a state where it was appropriate to fine-tune the general decisions I made in the Mx session about the placement and edit of the source music.

Sound Effects And Foley

With the music written, edited and imported into the mix template, it is time to do the SFX and foley.

So far, I have chosen to do SFX and foley in the mix project, rather than create a separate SFX session as I did with the Music.

This is because I can do all of my editing in the project that the SFX/Foley will end up in any way. This saves a step and means I don’t need to worry about importing another session into my main mix session. I also prefer laying the SFX/Foley last. While it certainly isn’t the least important element (far from it), it is the element that doesn’t rely on either of the other two elements (dialog and music) as much as dialog and music rely on each other to make sense.

The approach I have taken so far on all of my projects is to first lay down my ambience tracks, then work on foley sounds (i.e. footsteps and any other foley elements that need to be augmented or filled in), and then work on SFX. For both Foley and SFX, I tend to get the linear placement correct, then the volume (typically through clip gain), and then the panning placement.

Let The Mix Commence

Once all of the Foley/SFX are in place, roughly mixed and panned, I go through and start working on balancing the whole mix and more closely monitoring my loudness and making adjustments as necessary to hit targets, to make everything sound like it was all created that way in the first place.

Make It Appear Real

My goal in mixing is to make it sound as if we had just set up a camera and captured all of it as if it all happened in real life this way. You don’t want things sticking out and distracting the listener, whether that be through panning or volume or a combination of both. You want to create an environment that makes sense for the story you are trying to tell.

Now I am not advocating for everything to be as real as possible, but there should always be a reason for making things sound the way they do. For example, again from the Web-Series I recently worked on in the opening of the first episode, the scene opens in a dream of the main character, however, there really isn’t any visual cue that indicates that.

If you just watched the video, you could easily be confused about what or what isn’t real until later when we see the main character wake up from the dream. So to help tell the story, I added some relatively on-the-nose elements to help indicate that this wasn’t real. With over the top and dramatic reverb and the mood changes in the dream from cherry to dark, the songbirds change to ravens, and the music goes from happy, fluttery and bubbly to a minor, dark, tremolo diminished chord, all to give a first time view the indication that something isn’t quite right. So when the main character wakes up, the viewer can retroactively realise that this was a dream, and now we are in reality. 

If the video doesn’t start in the correct place go to 01:01 and watch through to 01:45.

Support The Narrative

This is the point of the mix, to help tell the story and to augment it. This means that for a lot of the mix you’ll end up trying to make things sound real or hyper-real depending on the situation. You want everything to sound both loud enough to hear but appropriate for the setting on screen, and the 2 or 3-dimensional placement of sound elements should both make sense and help set the scene or help tell the story. (See my raven example earlier).

This is the balance that you will spend most of your time working on when doing the final mix, in addition to dealing with any other surprises that show up at this stage (i.e. going back to Dx Edit to add ADR and getting it to match, or conforming to a new music cue, or making a change, requested by your client).

In the end, you are here to help make this story possible, a cog in the visual storytelling machine. You, the expert in sound, are here to make the director/producer ideas take form.

Sometimes what they want may not be possible, but don’t let the impossible stop you. Look for another way to accomplish their goal, from a different direction, as it may be possible, and produce the same result.

Ultimately the director/producer has the final say because they have the overall story in their mind and they will be the best judge to what best fits the story. That isn’t to say let them walk over your decisions. You should advocate for what you do as you are the expert in the field, but be judicious with your advocacy, gauge how much your team (producer/director) does or does not like what is being discussed and go from there. Hopefully, you’ll have a good working relationship with your clients, so you are more likely to be working as a team rather than having no say in your project but always defer to the producer/director’s decisions. They are the storytellers and likely have the most complete idea of the story they are trying to tell.

Deliverables

The final step is to create your deliverables. In this step, you’ll want to make a final pass of your final mix that everyone is happy with and print that to your recorder tracks in your final mix session.

This workflow will give you a final chance to listen to everything to make sure you aren’t delivering any mistakes to the final product; so this is an essential step of QC that needs to be performed.

Don’t fall into the trap of bouncing down at this step as there will be a point that you will have made a mistake and if you bounce you won’t know until the project is released publicly when it is too late to change at that point.

The recorder tracks are simply your final audio tracks that your Dx, Mx, and Fx are being routed through and that you are listening to in your mix session. If you’d like a more detailed breakdown of how best to route this check out Damian’s excellent article Free Pro Tools TV Post Production Template For 5.1 Projects.

I personally use the abbreviation of PM for my final mix, but you are welcome to use something else, just so long as it is consistent and clear what it is. PM to me means Program Material, and it is clearly different than any of my stems that will show up in the file list in Finder or File Explorer for my Windows friends.

Just make sure you let your team know that PM is the main mix if they aren’t familiar with it, or ask for a spec you should deliver to if it hasn’t already been shared.

During your final pass, you should hopefully also be creating your stems and downmix, so you won’t need to spend too much time creating them after your final pass of the mix and re-record.

Make sure your re-record PM and stems are precisely the same length as the video and/or have a 2pop included in them (if not specifically excluded by your spec sheet), so the video editor or person creating a DCP doesn’t have to worry about audio sync. It should be as simple as lining up the audio with the beginning of their final project and sync should be perfect.

Once that is done, make a new folder in your project folder that is your deliverables folder (I like to name it ProjectName_Deliverable), and copy all of your deliverables there.

Once that is done and you have made a quick final check that your deliverables make sense and have audio in them, compress the folder and upload it to your team for your delivery of the final project. Likely, this will be your main video editor who is creating the final Picture with your audio.

Summary

So let me summarize the process…

1. Perform your Dialogue assembly/edit 

2. Import Dx Edit into your DxMxFx (main mix) template, and create your Dx guide for your Mx Session

3. Write/edit Mx

4. Print any original music in the Mx session

5. Import Mx session into DxMxFx

6. Place and edit Foley then Fx 

7. Mix all elements to find the right balance of everything, so your final mix is close to spec.

8. Do a final pass to make sure you are hitting spec and to make sure you didn’t miss anything 

9. Time allowing, submit the first draft to your Producer/Director team, and make any changes they request.

10. Do another pass to make sure your mix sounds good again and re-submit. Do this as many times as either your budget/schedule allows for, and get a final version everyone is happy with. 

11. Do a final pass, and print your final mix. I would highly suggest you don’t bounce down at this point as printing gives you the chance to make sure nothing strange has happened to your mix for whatever reason, and it gives you a final pass to make sure everything is sounding the way it should. 

12. Create any downmixes (if mixed in a higher-order format) or any stems as requested. Hopefully, this is part of your final print process. 

13. Deliver your “deliverables” folder to the person who will create the final product (usually your team’s main video editor). 

Conclusion

I hope this helps you start your first project in Audio Post or helps you improve your workflow for future projects. By no means is this the be-all and end-all of how Post should be done. This is the workflow I have developed from trial and error. Please take my workflow, and make it your own and do let me know how you improved/changed what I did, in the comments below. New ideas are always welcome.