The ease at which your organization can get their content livestreaming, leaves us with a lot of media content awaiting an audience. You've gone out and got all the gear, such as great cameras, a switcher, cables, tripods, capture computers, and web encoders. In addition, you have your livestream license and have hooked up your audio.
The single biggest point of failure for video production, both live and in post, is audio.
Your video quality looks amazing, and you have incredibly creative video directors. You're ready to rock, and may have been streaming for some time.
So what was that last piece you mentioned when you assembled your system? Oh yeah, audio.
The single biggest point of failure for video production, both live and in post, is audio. It's often the most mishandled and last piece, particularly when working on video. Bad audio can also be the single biggest distraction when trying to hold a viewer's attention to the media content.
Awhile back (July 15, 2016) I was able to write more in depth about the how-tos to assemble your broadcast mix from the same console you have to mix FOH from. In today's article, I'm going to address two final steps before sending your broadcast audio off to join with its video counterpart.
Polishing Your Broadcast Mix
You've worked hard to get all your buses, matrixes, pan, EQ, and room mics all dialed in for a great sounding broadcast mix. Now we need to get it all into a final package to be delivered to broadcast.
There are just a few final touches here that we can do to step it up.
If you have assembled your final mix from your FOH console or even a dedicated broadcast room, begin to study the recorded mix in as many types of listening spaces you expect your viewer to listen from. For example, if your final mix is designed for livestreaming, listen to it from anything such as a laptop, tv speakers, cheap ear buds, home stereo speakers and in your car. If your final mix is designed to be played from another video campus, you need to go there, and listen to how it sounds on that PA.
Most of the time, your final mix from a controlled studio or your FOH headphones will not translate the same in other sources. The really good studio engineers in the recording business always take their mixes home to listen on other sources, and so should we as broadcast mixers. The only way we can know what needs to be adjusted is when we allow ourselves to become part of the audience.
You might find some overall compression is too audible, or the EQ is too muddy or harsh. While the mix can't sound perfect from every listening environment, address the bigger issues from the extremes, to meet in the middle for an average. For instance, from the car or home theater, it's too "boomy," so you'll need to address the low end, sometimes applying a high pass filter up to 50-60hz can help. If it's super shrill from small computer speakers, you may need to address a bit of the high end.
Either way, you'll likely need to assemble a final chain of processing to the broadcast mix for some fine-tuning. For example, my broadcast chain consists of three or four Waves plug-ins, Puigchild (for a light compression, but it also does magic with warm mid-tones which are important in livestreaming), a simple overall EQ plug-in, Vitamin (which addresses tighter or wider control of the stereo image by frequency bands and also some final EQ), and L2 (a brick wall limiter).
Each piece of your final processing chain should have purpose and help you obtain the final polish. Changes here likely will be fairly subtle, amounting to overall small changes.
If you find yourself making massive EQ changes, you might need to go back and make some better decisions on the EQ of your inputs.
Getting Your Final Levels Right
Fasten your seat belt here, as we are going to get a little technical with numbers, but I'm going to keep it within reason. When it comes to final product levels, it very well can be summed up as a numbers game.
The two biggest things you're going to strive for are consistent and clean levels. There isn't anything more annoying than constantly having to turn the volume up or down, when the program content changes. This includes transitioning from a full band, to video playback, then to the spoken word. The goal is to have a very close average of those signals smoothly being delivered to the listener.
The other one is hearing constant crackles of distortion, because the signal along the way has been crunched too much by a compressor, to then be clipped again with the final mix running too close to 0dbFS [decibels relative to full scale]. (For the sake of the remaining article, we are going to use the dbFS scale)
In audio, there is a constant goal to strive for a consistent gain structure, and to get as hot of a signal to 0dB that doesn't clip. But in broadcast, when it comes to our final stems and mix, we are going to be pretty far from it.
Learn to leverage your brick wall limiter to not only cap your final volume, but also shorten your dynamic range. The brick wall limiter will keep your final mix from clipping, but you can adjust the threshold to push the perceived loudness in the final mix up into the limiter a bit.
Unlike a compressor that just pushes down volume by a ratio when it reaches its threshold, the limiter isn't a compressor, it simply doesn't allow audible transients over the given ceiling.
The compressor actually does subtly the same thing, if it's also in your final processing chain, but I recommend one use it softly enough, where you won't hear hard pumping when music content is going full tilt. I use it just to tickle the audio signal at its peak volume. Also of note, using a compressor at 100:1 ratio effectively is a limiter, however, most live sound consoles using this setting with their onboard compressors will hear drastic pumping. I highly recommend using studio limiters (analog or plug-in) in all cases, for a final broadcast limiter.
So how much dynamic range should remain between music and spoken word? I look for a 3db difference, meaning spoken word will land about 3db quieter on average. It does mean some soft moments in the music may fall below even spoken word levels, but it shouldn't by much. TV standards often say they don't want instantaneous peaks great than +2db. But if your levels are drifting more than 10db, you have too much dynamic range in your mix. Try to tighten it up, to less than 6db.
Now that we have consistent average levels between all of our incoming content, we need to send our final mix to the broadcast room for distribution. This final level can vary based on your needs, but the key thing you will want to know is never send a final level at 0db because it'll clip down the road, when it meets up with its video counterpart. The same holds true with any stems being sub mixed along the way.
The broadcast brick wall limiter standard is -2db, but most of the time, it will be set far lower. If your final mix is going through additional mixers, getting your gain structure lower ensures headroom throughout the audio chain. For example, many TV stations have minimum requirements from -24db to -10db of final delivered levels. Whereas radio deliverable content will target -6db.
What do all these numbers mean for the average church that is just broadcasting to livestream or archiving to YouTube and Vimeo?
It means there is more happening after you are done mixing. This also will make you think about your stems upstream in your mix, and leaving headroom there allows for a cleaner delivery to your final mix.
For your final mix heading for livestream, a good starting point with your brick wall limiter is setting your threshold to -10db and your ceiling also at -10db.
In your video room, adjust your level feeding your livestream software to leave about -6db of headroom as a starter to ensure you don't clip your livestream encoders.
The reasoning for -6db is digital encoder systems have an absolute maximum ceiling of 0 dBFS. Ddepending on the manufacturer encoder used, though, overshoots and spikes of 2 to 3db can occur. Leave headroom, as you can fine-tune those levels to be tighter, once you can predict any encoder spikes.
If the final product is an archive video, you can go all the way up to -2db with the limiter ceiling, as this is your final product and not being delivered elsewhere.
To Wrap This All Up
Finalizing your broadcast audio takes some tweaking. The goal of bringing you these guidelines is to get you closer to understanding some benchmarks that you can fine tune to produce a broadcast that is distraction free, for its intended audience.
House of worship video production is not like mainstream TV broadcast (unless you happen to actually televise), due to the use of so much varied equipment choices for churches to purchase.
There won't be a one-size-fits-all here, and as we know, many of you may be rockin' it with simple home consumer equipment. However, it's my hope that this gives you encouragement and a greater picture that excellent broadcast or post audio is obtainable and only a few polishes away.