I've been editing a six-episode, 53-minute-per episode TV series to run on the community access cable channel.

Their standards given to me: -18dB average, no peaks over -15 or valleys under -25.

I've done plenty of audio-only mixing over the years, and audio for video mixing for youtube and the like. It's always been 0db is the max, otherwise loudness to taste.

So, I'm a little stumped as to how to get my mix into these parameters.

I'm using Sony Vegas. I've been building and mixing the show with the master at 0, not aiming for anything other than "sounds good" and "doesn't clip" by just using levels and some conservative compression and plugins.

What I'm wondering is ... - ought I re-mix the whole thing and aim for a lower peak, or just keep mixing with 0dB as my ceiling and bring it all down and into the specified window of peaks and valleys afterward before delivery? - if mixing normally and handling the specs after the fact is the better strategy -- where to go from there? Drop the master to -15 and compress from the bottom up? Maybe mix the whole thing down to one long audio file and take it into a DAW, put Ozone on it, as though I'm mastering it, and lay the mastered audio back in?

I feel like the answer is staring me in the face and I'm just psyching myself out about this. As though it's as simple as "go into Ozone, use preset x, threshold at -25, ratio at 4:1, limiter at -15, EQ to taste." Thoughts?

3 Answers 3


Welcome to broadcasting :)

In Europe we have a EBU R128 regulation, which is a loudness normalisation with ITU 1770 measurement. Our loudness is -23LUFS and our true peak -1dBFS.

I am doing a lot of TV commercials so I hope i can help you with the mix idea:

Normalisation of audio gave me a real headache when it was introduced, but it is quite simple. Just open up a Metering device on your mastering channel. Be sure that the metering is measuring the right parameters via the right standard(for e.g. itu1770...).

Then just level your loudest elements (normally its either dialogue or music) that they nearly (roughly 1-5dB depending on how loud the episode will get (action scenes, car crash, explosions etc)) full fill your RMS or momentary loudness parameter. Its important that you do not look on your normal dBFS meters. Then mix everything as you were used to, but do not adjust your "anker" element. (be sure to compress, highpass and denoise the individual elements before you get into mixing so that you only need to fine-tune everything.)

after that get a measure the whole episode (some devices like the Visual LM allow for offline measuring or measuring while you bounce). but the file into a new project an then adjust the average level via a gain.

after that check your dynamic range. if it is to high -> compress the master channel with low ratio, high release, medium attack and low threshold. look for the highest peak in the waveform to find your makeup gain

after you adjust your makeup gain be sure to check your peak. normally you can calculate it or use the audio suit to render the compressor in the loudest peak to see where your new peak is. use a limiter to limit at your desired peak level. (always limit for safety issues)

after that measure it once more for safety and to hear if there is any point where the compressor or limiter won't work. at these points use volume automation before the device to reduce the artefacts.

i hope that helped peace and gl

  • Great advice, there. Thank you. Sounds like Izotope Insight will be a big help on this one. I appreciate your insight, too. Thanks! Feb 3, 2014 at 19:38
  • I like VisLM-H from Nugen for metering since they have most of the loudness standards . Great I could help! Feb 3, 2014 at 23:41

It's not standard practice to do any "Mastering" when mixing to picture for TV shows, commercials or film. In most cases not mixing to the correct level will wreak havoc on your delivery specs such as if you have to provide splits or stems of the different elements.

The first thing you should do is make sure your mixing at the correct level for the size of your room. They have lots of info about it in the ATSC/85 documents.

Most people mix with subgroups of their different elements so you could preferably put a limiter on each buss master with the output level a couple db lower than -15dbfs for when a couple different elements get loud at once.
If all you have to deliver is a stereo mix then you could probably just strap the limiter across your master with the output at -15dbfs and adjust the other settings to your ear. -18 average with a -15 peak seems kinda a strange requirement but you have to go with the specs they give you are those measurements with the same type of meter?.


A follow-up...

... I pressed further with the broadcaster, and they indicated that the -18 average was not RMS or loudness after all. They just want no peaks beyond -15, with most of the program around -18. It looks like a mix of me overthinking things and the broadcaster being kind of vague with their technical standards.

My final plug-in chain on the master: Ozone dynamics to glue things together a bit, Ozone maximizer to bring up the loudness, Waves L2 as a final limiter.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.