Nathaniel Kunkel and Alan Meyerson Discuss Aspects of the Audio Packing Problem

An active field of industrial research is devoted to devising packages that optimally pack into containers. The goal is to utilize as much container volume as possible with little to no room to spare.

Once audio tracks are recorded, the audio engineer faces a similar problem. The big difference is that none of the packages, one per track, have the same shape or size. The audio container holding the packages varies little from project to project. The variety of packages to arrange makes audio engineering interesting, but the overall goal is the same as that for manufacturers and distributors: use as much of the container space as possible without damaging the contents.

In the accompanying video, Nathaniel Kunkel and Alan Meyerson discuss many encounters with this audio packing problem. Their approaches to the problem involve signal processing tools, Quartet DynPEQ (though we must also mention Trio DynPEQ, which deserves more love) and other Pro Tools AAX plug-ins.

If you do not have enough time to watch the video right now, here are a few perspectives from Nathaniel and Alan reduced into bullet points. We strongly urge you to come back later and see the whole video.


Nathaniel Kunkel

• The engineer must prepare a lead vocal or instrument track to be rendered arbitrarily loud. In the process of turning the track gain up, aspects of the sound can be exposed that beg to be brought back down. These aspects are not necessarily part of the performance and often are confined to a subband of the spectrum. DynPEQ can be used to apply dynamics on the subbands of concern, allowing a vivid performance signal to emerge.

• Transferring old recordings to new audio formats can introduce ambience that obstructs the original recording’s ability to occupy the new space. DynPEQ can help manage the ambience, and the analog peak limiter in Quartet DynPEQ can smooth overruns to match level with the original format.

Alan Meyerson

• Mixing large orchestral recordings of intense music often leads to instrument sections in conflict for the limited audio space. DynPEQ can help regulate the interaction when large sections overlap and can help find critical space for solo instrument inserts. The analog peak limiter in Quartet DynPEQ enforces the container size while striving to avoid distortion, warm or otherwise, maintaining the intended sound at the desired gain.

Make your mark with Pro Tools

Create music or sound for film/TV and connect with a premier network of artists, producers, and mixers around the world.

Re-Recording Mixer Tom Marks on the Sound and Music of Sense8 Season Two

Tom Marks has mixed feature films, theatrical trailers, TV programs, games, VR media, webisodes, and film scores.  He has worked in many current and emerging audio formats, including 5.1, 7.1, Stereo, Dolby Atmos, Auro-3D, IMAX, and IMAX 12.  Prior to moving to Los Angeles in 2003, Tom worked in television and music while living in Illinois and Florida.  I talked to him about his role as the re-recording mixer on season two of the Netflix series Sense8.


Wholegrain Digital Systems (WDS): Sense8 credits many shooting locations all over the world.  Even if the crew is the same everywhere, the audio characteristics of the locations will vary.  What is your role in turning these sources into a single soundtrack?

Tom Marks (TM): Our dialogue editorial department will go though all the takes and mics to help solve on-set issues.  Then it’s my job as the re-recording mixer to smooth out all the angles, cuts, and scenes.

WDS: What do you mean by angles?

TM: It’s a film term used in reference to different camera angles.  Once a scene is cut together picture-wise, there can be a lot of variation in dialogue–level, tone, noise–for example, when the audio cuts from a boom mic to a lav.  Using volume, filtering, EQ, noise reduction, and other tools, I smooth out the cuts so the dialogue is consistent.


WDS: Sense8 has a large principal cast.  I assume sometimes there are not enough boom microphones to go around for ensemble scenes.  How often is the cast outfitted with lavs?

TM: There’s mostly one boom mic, occasionally two.  When shooting with at least three cameras, it can be difficult for the boom mic to get close, so actors are usually wired with lavs but not all the time.  Some action scenes only had one boom mic placed far away, so the fight efforts had to be looped.


WDS: How often will lav signals pick up clothes rubbing and other unintended sounds?

TM: Quite a lot.  That’s where dialogue editing is crucial.  The editors will remove unnecessary sounds between words or use dialogue alts/ADR for certain syllables.  In my dialogue predub, I’ll continue to work on these problem areas using iZotope RX6 Advanced.


WDS: Do you wind up processing dialogue more than other elements when mixing?

TM: Every show is different as far as the work involved, but generally speaking, dialogue takes more time.  In the schedule, we start off with a dialogue predub to build the rest of the show around cleaned up material.


WDS: What is the target channel configuration for the Sense8 mix?

TM: Netflix’s delivery spec calls for a 5.1 surround and 2.0 LtRt mix.  The stereo is a downmix of the surround using Neyrinck’s SoundCode LtRt Tools.  We have a Pro Tools rig dedicated to recording all the deliverables.  We use Dolby’s Media Meter plugin to ensure we meet Netflix’s loudness requirements.

WDS: In Sense8, are the surround channels used for ambience, spatial cues, or something dependent on context?

TM: If you soloed the surrounds, you would hear many things depending on the scene: 1) music ambience from the 5.1 orchestra stems; 2) reverb from dialogue, music, backgrounds, foley, and sound effects; 3) panned sound design, backgrounds, production crowds, and sound effects.


WDS: The Sense8 Christmas special includes a comissioned recording of Leonard Cohen’s Hallelujah featuring Daniel Martin Moore and the Apollo Chorus of Chicago.  How many microphones were used on this large choir?

TM: The choir formed a circle around the conductor, and one of the engineers, Mike Tholen, used his two Lucas mics in the middle, with X-Y mics above them.  There were also two wide mics with two more in the balcony, so 8 total.


WDS: Was the choir recorded in a studio, soundstage, or a naturally reverberant space?

TM: The choir, piano, and acoustic guitar were recorded at Gottlieb Hall within the Merit School of Music in Chicago.


WDS: Did you emphasize stereo aspects of the choral mix to clear more space for the lead vocal and instruments?

TM: The lead vocal was recorded (close-miced) at Daniel Martin Moore’s studio, while the choir was recorded with the room’s natural acoustics.  So most of the separation resulted from the recording.

photo credit: Netflix

Make your mark with Pro Tools

Create music or sound for film/TV and connect with a premier network of artists, producers, and mixers around the world.