Working with LevelNorm
The best way to mix audio files for broadcasting in the new paradigm may at first seem quite dangerous. Basically you just lower your modulation levels, ignore your meters and focus on the sound. Once in a while you take a glimpse at the digital peak meter to make sure you are not clipping. Don’t be afraid that you will loose resolution by modulating low, with current 24 bit high resolution equipment this is no threat anymore.
If you feel insecure working without visual feedback, please consider using a realtime EBU mode loudness meter. Alternatively use your VU’s or PPM’s but align them correctly. A proper alignment for your VU meter would be 0 VU to equal -18 dBFS. For analog PPM’s, keep the 0 PPM = -9 dBFS level. During production, avoid modulating hot all the time. But at the same time do not hesitate to use all that headroom. If you are using a digital peak meter next to the other ones, ATSC and EBU allow peaks to reach almost 0 dBFS, so don’t worry about short peaks above -9 dBFS.
The key to almost automatically achieve consistent results without using meters, is to adopt the film postproduction practice of aligning the monitor gain to a fixed acoustic level. In the broadcast case, the EBU tech 3276 document prescribes to align a -18 dBFS (rms measured) pink noise track to a sound pressure level of 82 dBA per speaker for stereo or 78 dBA per speaker for 5 channel surround. If you prefer to have your monitor SPL a little lower, just note the deviation from the calibrated level for later reference. The alternative is to simply normalize a well known file to -23 LUFS, play it back and set the volume control to your preferred listening level. The main concept is to now mark the monitor control position and work at this reference level position 90% of the time or more. Your ears will tell you how loud you should mix, you don’t need a meter for that anymore.
When you are satisfied with your mix and overall sound, just print the result to a seperate stereo or surround track, let LevelNorm analyze and process the file on this track and your audio is finished for release. After some practice you will note the necessary adjustment by LevelNorm will be minimal.
Broadcast and post production engineers in the US know that according to the ATSC recommendation A/85 they need to adjust the volume of their program based upon analysis of the dialog track(s) only. LevelNorm can be a big time safer here because it allows to very quickly measure the loudness of (part of) the dialog track. The 2009 edition of ATSC A/85 has no ‘relative gate’ in the measurement and a target of -24 LKFS (= LUFS) in stead of -23 LUFS like in EBU R128. The LevelNorm-ATSC plugin complies to that standard. Since the ITU has adopted the EBU R128 relative gate in 2011, it is now unclear if ATSC A/85 advices to measure ‘all with gating’ or dialog only. Grimm Audio will closely follow the developments.