I recently started doing sound mixes for tv programmes and I can’t seem to get the programme level as loud as the surrounding audio i.e continuity announcer, ads etc. My settings: if I take the narrator, I adjust the volume gain until the narrator peaks between 4 and 6 PPM (4PPM IS -18DBFS). I insert the protools audio suite vocal compressor across the dialogue buss with the following settings. ratio 2.5:1/ threshold -14 / gain 4.2 /knee 10db and attack 1.2ms. I then apply a L2007 mastering limiter across the master mix , max output -10db and threshold -11.5db. If I reduce the threshold on the limiter the sound seems to get overly squashed. The surrounding audio on the tv channel seems to have a better dynamic range with out a loss in volume. From these settings would anyone have an idea of what I might do to improve on the volume issue?

2 Answers 2


The simple answer is you won't unless you are prepared to squash the life out of your programme. The key issue here is that the PPM is not an indicator of loudness, it is designed to tell you where the peaks are which was crucial in an analog world. However the ads are processed very heavily so that my ad is louder and any other ad and the poor viewer has to dive for the remote. We are slowly moving over from normalising to peak level, which is what we do now, usually PPM 6 in the UK and across to normalising to loudness -23 in the Europe and -24 in the US. Take a look at my Loudness overview post on Pro Tools Expert

As to your workflow, I usually put a mild compressor on the dialog buss and allow up to 6dB of gain reduction, but I put a different comp on the narrator which is more severe. This helps the narrator to sit over music and also has the effect of making the narrator sound more authoritative.

Hope that helps.


I am interested to know what sort of release times people are using on compressors and limiters to avoid pumping or pulsing on programme material while still maintaining adequate compression levels to keep the volume up ( without sounding unnaturally squashed).

  • I find on my dialogue premix sidechain I like my compressor in Opto (just happen like the sound of it better) with about a 7ms attack and 35ms release, about 6:1 compression beginning around -12dB. Basically, I 'mix into' the compressor and let it gently control the high dynamics when necessary rather than use the compressor to babysit the levels. With FX I tend to brickwall it just under what the spec allows and again, mix into it. I like to run the faders rather than using sidechain compression to determine the mix balance (which can lead to pumping etc) Jun 7, 2012 at 3:52
  • Ok..cool thanks. That's a shorter release time than I would tend to use, but if that's what works for you then that's great and intersting to know.
    – rockit1
    Jun 8, 2012 at 0:34

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.