In an engaging presentation at this year’s Mediatech Africa show, leading sound technicians Adriaan van der Walt and Cristo Hattingh shared insights into the multi-layered topic of Communication Protocols and Cross-System Integration in the Modern Live Production Environment.

In this white paper feature, the two professionals outline the challenges and opportunities of networked audio solutions in the context of live event and broadcasting.


From the outset our brief from the client (the producers of The Voice South Africa) was the driving force behind just about every decision regarding our workflow and gear allocations.

The brief was that the viewer experience of the audience watching the show live at the venue or in the comfort of their home on the broadcast should be as close to the same as possible.

Of course, the scale of such a request – when dealing with a show comprised of as many ‘moving parts’ as The Voice South Africa – is not something to be taken lightly, and could only be executed with utmost planning and precision, and by finding efficiencies throughout the communication protocols and cross-system integration chain. This means both effective communication between people – through hashing out the exact technical specifications and having collaborative teams in place to make it all work – and finding the best methods of communication between technologies to deliver on the required outcomes.

Crucial to delivering on the client’s expectations was working closely with the production team, both Technical and Creative, to understand how the four different stages of the show – Blinds /Battles /Knockouts /Live – would map out as we progressed. Also keeping in mind the key elements within the show, the Coaches/Talent/Presenters/Audience, and their impact on how we approached all the different stages.

The Talent’s journey through each stage would be unique, they would not only perform but live out an experience, and our goal was to bring that experience to the audience. This required unique solutions and approaches, including expanded RF microphone capabilities and innovative splitting systems to allow the storyline of the show to continue even after the talent had left the stage and started to interact with the ENG crew.

Here is some technical insight into the RF systems and how we mapped it out.

During the Blind auditions, each role player had to be mic’d up in order to give the Post editors the content for their storyline. This meant a Shure ULX-D Beltpack with Sennheiser MK2-Gold on Talent, Coaches, Presenter and key family members (pre-planned by production). Although the family was technically off stage while the Blind performance was taking place, all reactions and comments needed to be captured, we even had additional ambient mics in that area for larger family groups, as only two lapel mics were allocated to family members . After a performance, the Talent would join the family on camera but off stage, and even move to the holding room, so we added a second ULX-D receiver unit synched to the Talent belt packs in that room so the ENG crew did not need to re-mic and interrupt during spontaneous moments.

We also mic’d the steps and floor of the stage to assist in making the walkup to the mic as realistic as possible.

During the Knockout round, the lapel mic logistics got even more compounded, as all Talent started the session on stage and needed to be pre-mic’d: this was to capture any comments and whispers between talent before and after they had the opportunity to perform.

We used Shure ULX-D Systems with KSM8 capsules for the performance vocal mics. During the Blind/Battle and Knockouts this was up to four Talent mics; when we got to the Live show, the lapel count dropped drastically and the performance mic count went up – at the peak, 22 microphones were allocated in some episodes.


When designing complex workflows of this nature, it will be inevitable that the protocols being used will be diverse. Even using one protocol like MADI can be challenging, as most MADI connections are peer to peer. This means dedication of a port to a single purpose, but – when only 16 voices are needed – accessing the others become impossible unless additional hardware is used to facilitate this.

Now add Analog, DANTE, AES, AVB and 48K/96K to the mix, and the fun begins. Once you mix communication protocols, as with people, translation and context become vital. We used them all. Starting with the decision, based on required workflow, to use XTA Active splits in the capturing path meant that Broadcast could function completely independently from the Sound Stage.

Now, this doesn’t mean to say we couldn’t be completely digital.  It would definitely have simplified the stage patch in terms of the amount of copper cable being used – and the challenges this brings!  We felt, however, that having an active splitter between the venue I/O (Digico SD10s) and broadcast I/O (Avid S6L) had more benefits that outweighed the possibility of hums or buzzes brought on by using so much copper between the Digico and Avid platforms. 

Some of these benefits are:

  • No clocking issues between Digico and Avid platforms
  • Making changes to gain and other relative levels during virtual sound-checks could be done independently between venue and broadcast systems
  • A particular input or output signal could be distributed between the platforms without having to consider sample rate conversion which meant saving time solving challenges on the fly
  • Virtual sound checks could be done at different times: FOH & Monitors could only do virtual sound checks after-hours, once choreography and camera rehearsals were complete, which meant broadcast did not have to wait around at night while the venue audio team was busy fine-tuning their mixes

Side note: the venue used for the production still operated as a church on a Sunday morning, so as Broadcast was only connected via Active spilt, we could switch on and do system checks and virtual sound checks while the services continued. Conversely, broadcast could do a virtual sound check on a Sunday morning while the venue audio team were busy with church services

The Broadcast solution consisted of a SoundGP Music OB, and a SuperSport OB

SoundGP – all performance related elements:

  • AVID S6L using the AVB Protocol
    • 2 x 64/32 Stage Racks
    • 2 x MADI Cards 64ch 48k each
    • AVID ProPAck AAX and McDSP Plugins
  • Direct Out Technologies 96/48 SRC Unit. because the S6L is clocking at 96k we needed to convert to 48K to interface with Supersport. 
  • Direct Out Technologies M.1K2, routing the 2 AVID MADI cards to SuperSport and RME USB/MADI XT for WAVES MultiRack on an iMac mainly Loudness R128 Monitoring
  • RME MADI/DANTE used for the Music director interface via DANTE running ABELTON on the stage
  • Apple MacPro with Protools AVB connection 128ch, recording for archiving and Virtual Sound Checks
  • Monitoring was JBL LSR 6328 and 6312 for LCR&Lfe with Genelec 8010A rear, controlled from a SPL Studio Monitor Control
  • Apogee BigBen Clocking unit to sync up with SuperSport as well as all local devices

SuperSport – all the Talking heads and EVS playbacks (Operated by Ross Glicrest & Arie Kruger):

  • Calrec SIGMA Console
  • Hyrda I/O in Venue at FOH and Stage to facilitate the receiving and sending of sources.
  • WAVE Soundgrid with MGB MADI connection (via D.O.T M.1K2) for DUGAN Auto-Mixer on Coaches
  • All Genelec 8020D monitoring
  • ENCO playback system for all Stings/Underscores and tension beds

Meanwhile, at Front of House the following protocols were used:

  • Analog for talkback I/O, accepting inputs from various playback sources, Smaart I/O and also outputs to delays and other spaces like the production office and ENG feeds
  • MADI as the native I/O on Digico platform
  • Optocore as the transport platform on which the Digico MADI signal would run as well as Main and Back-Up I/O routing between the SD10 and the L-ISA processor
  • Waves Soundgrid, which also runs in parallel to MADI on the Digico, was used for I/O to an external host computer for plug-ins and also multi-track recording and playback
  • Audinate Dante which was fed from Monitors to the Lighting department who ran their own DAW for programming and talkback purposes
  • AES was mainly used as the main output signal to the PA system and this signal also ran on the Optocore backbone
  • FOH used Ableton Live as the main multi-track system and Waves Tracks Live for back-up, both connected to Waves Soundgrid on its own closed network

During the show, the ability to work on independent workflows was even more important. 

Both broadcast and venue would use dedicated snapshots for each performance.  The point at which each department would switch to the next scene was totally different.  As an example:  the moment a performance was done and the host would facilitate a discussion between the on-stage Talent and the Coaches, broadcast and monitors had to change to the next scene to prepare for the next performance.  FOH had to remain in the current scene during these post-performance conversations and could only advance to the next scene as the next Talent was announced.

A total of five closed networks at FOH were established to take care of the following:

  • Waves Multirack which was a connection between the SD10 data port, an integrated Waves card, a Waves server and an external host computer on which plug-ins were operated
  • Waves Soundgrid which was a connection between a Digigrid MGB (MADI to Soundgrid interface connected to the SD10) and another external computer which ran the main multi-tracks
  • Apple Audio MIDI network which had a two-fold role; firstly to transfer MIDI commands sent from the SD10’s snapshot control to various computers to do combinations of program changes and controller changes and secondly to connect the back-up multitrack DAW to Soundgrid
  • L-Acoustics L-ISA controller which ran a GUI to access both Main and Back-Up processors
  • L-Acoustics LA Network Manager for real-time monitoring of the PA system during live broadcasts

The five networks had to run on a managed switch to eliminate network flooding and ensure the stability of each platform.  This meant the traffic from one network would not interfere or interrupt the traffic of another.


Of course, the heart and soul of The Voice South Africa remains the music – and an essential aspect of the system’s design was to showcase the talent in the best possible light. We developed workflow solutions that made an impact at every stage of the production process to make sure the music performances took centre stage.

This same amount of planning went into the design of our channel lists. As the idea from the Show Director, Darren Hayward, was to start relatively small in the early stages of the competition, and then add more and more elements to the band as the show progressed, we had to ensure that this aspect was integrated into our design from the outset.

The pre-recorded stages of the production, namely Blinds/Battles and Knock-Outs, were filmed roughly six months apart from the Live rounds, but we did not want to treat them as separate entities because as important as it is to ensure continuity within TV productions, it also counts for the technical workflow in the audio department.

From the initial discussions, all the channel counts and console architecture was planned with the Live shows in mind.  During pre-records, the live band might be simpler and would have a lower channel count compared to Live shows – which meant that there would be many open and unused channels at the start of the production, which would then eventually be used as the production grew, as and when Darren and the team planned for these evolutions to happen.

Digital platforms like Avid and Digico are highly configurable, in particular the way in which the operator can change the layout of the mixing surface.  Having the ability to adapt the layout of the mixing console from one performance to the next makes the job far easier during the event.  Naturally all of this has to be pre-programmed during and after rehearsals and also verified during the final dress rehearsal.

The complexity of an audio ecosystem like this is a result of integrating different platforms and protocols – and, most importantly, being prepared for the challenges which are part and parcel of live TV productions.  When the director calls for something you have to be able to produce consistently good mixes and it is only possible through loads of planning, rehearsing and communicating with fellow team members.