Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Broadcast Levels (Read 4961 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Broadcast Levels

I have a few questions about radio broadcast standards. 

Background Info: I recently purchased an audio editor called 'Hindenburg Journalist'.  The software analyzes audio as it is imported and then automatically adjusts levels.  In many cases it lowers the level of my clips considerably. 

Hindenburg Systems says their program adjusts audio clips "relative to each other for a professional and balanced mix."  That part seems fine.  When importing audio into Hindenburg Journalist, all levels are automatically set to the following levels: Peak 9 QPPM; Narration 21 LUFS;  and Music 23 LUFS.

I notice that the finished output file has peak amplitude way way below 0dB. 

I understand leaving some headroom when recording my initial tracks.  But what is the advantage of leaving so much headroom in a finished, mixed output file?  Isn't the listener just going to have to crank up the volume?

Wouldn't it be better not to drop the levels of everything so much but rather keep the final mix closer to 0dB to lessen noise?






Broadcast Levels

Reply #1
I have a few questions about radio broadcast standards. 

Background Info: I recently purchased an audio editor called 'Hindenburg Journalist'.  The software analyzes audio as it is imported and then automatically adjusts levels.  In many cases it lowers the level of my clips considerably. 

Hindenburg Systems says their program adjusts audio clips "relative to each other for a professional and balanced mix."  That part seems fine.  When importing audio into Hindenburg Journalist, all levels are automatically set to the following levels: Peak 9 QPPM; Narration 21 LUFS;  and Music 23 LUFS.

I notice that the finished output file has peak amplitude way way below 0dB. 

I understand leaving some headroom when recording my initial tracks.  But what is the advantage of leaving so much headroom in a finished, mixed output file?  Isn't the listener just going to have to crank up the volume?

Wouldn't it be better not to drop the levels of everything so much but rather keep the final mix closer to 0dB to lessen noise?


Judging by the LUFS units, it is likely using ITU BS.1770.  There is plenty of information about it on here on hydrogenaudio, as well as the related implementation recommedation EBU R128 or the American equivalent ATSC A/85.

Is your audio mixed material (music + speech) or just speech?  If it contains significant musical or other broadband content, it will likely lower the volume level away from 0dB peak.  If it did not lower the volume of typical musical content, the only way to adjust speech-only content to match the same perceived loudness would be to adjust to it well over 0dB peak.

Broadcast Levels

Reply #2

I work in North American broadcast TV and nominal operating level is -20dB FS for the tape formats used. Duplicates for Europe (particularly Germany) insist on an average level of -21 +/- 1dB LUFS. That's TV but radio is completely different, The rules for FM are stay within your channel and the correct power. For AM I think (was never a radio engineer) you can hit +125% positive modulation but may NOT cut off the carrier on negative peaks as it causes out of band splatter. Radio people have found that American consumers are very dumb and will select a station based on how loud it is. It doesn't need to sound _good_, just LOUD. It actually makes your job quite easy as all you need to do is make sure you don't clip the audio _more_ than it already is.

Keep in mind that _your_ audio is likely 16 bits while video recorders are 20 bits. That minus 20 business effectively cuts TV back to 17 (with peaks at -8 dB LUFS which gets near 19 of the 20 bits ) but for you its like 12 bits and no peaks because the radio guys already squashed it to death to stay legal. In other words, turn it up so the peaks are around -3 dB FS and you will be fine.

I'm not sure how your software shows levels but in Adobe Audition it's so simple my 3rd grader was doing fine with levels.


Broadcast Levels

Reply #3
Quote
I notice that the finished output file has peak amplitude way way below 0dB.
Loudness perception is more related to average levels than peaks (although it's a bit more complicated than that).  Some different program material might end-up with peaks at or near 0dB after being processed through the same software.

Broadcast Levels

Reply #4
The rules for FM are stay within your channel and the correct power.

The spec relates to peak carrier deviation only: +/- 75KHz = 100% modulation.  Average levels are the result of processing and can be anywhere, but typically would vary station to station by only a dB or so, special formats like classical are the exception of course.

For AM I think (was never a radio engineer) you can hit +125% positive modulation but may NOT cut off the carrier on negative peaks as it causes out of band splatter.

Correct, but that's because some voices are highly asymmetrical, and since AM is often a high noise floor medium, the extra few percent positive can help sometimes.  They don't actually "force" all modulation to do 125+.  That would be nasty.  They just either let the highest asymmetrical peaks go positive, or scramble it all up so it doesn't matter.  Out of band splatter is a function of the NRSC mask, which is mostly a frequency response issue...unless this new-fangled digital stuff is present.
Radio people have found that American consumers are very dumb and will select a station based on how loud it is. It doesn't need to sound _good_, just LOUD.

That's the programmers and managers ego at work.  They have all the data that shows none of that is actually true.  But just imagine the ego sag if you punch up your station on the dial and it's quieter than the competition.  The reality is, a listener has a knob called a volume control, and he twists it all the time, station loudness being only one cause.

Broadcast Levels

Reply #5
Correction: AM's do force to 125% positive post phase scrambling, and it is nasty.


Broadcast Levels

Reply #7
Radio people have found that American consumers are very dumb and will select a station based on how loud it is. It doesn't need to sound _good_, just LOUD.

That's the programmers and managers ego at work.  They have all the data that shows none of that is actually true.  But just imagine the ego sag if you punch up your station on the dial and it's quieter than the competition.  The reality is, a listener has a knob called a volume control, and he twists it all the time, station loudness being only one cause.

I too doubt that finding. Seems to me from logic, common sense, and experience that most people choose station based on what music is playing. They hit the button until they find something that doesn't sound bad, and failing that, hit it some more until they find the least bad thing. Usually they stick with two or three stations, like my dad will never even try anything besides the oldies or classic rock stations. Some styles of music are more popular in certain areas and I'm sure a classical station pumping up its music louder than its competition will still be trounced by country music in most, if not all, parts of America. I'd love to see this supposed data. I wouldn't even mind if it showed my assumption to be wrong and that indeed there is a statistically significant relation between higher volume level and more listeners within the same range.

 

Broadcast Levels

Reply #8
Perhaps I wasn't very clear, but the data available to radio programmers does not support the supposition that people choose the loudest station.  The audience analysis data has never supported that.  The data supports that people choose stations for content first, reception second, and audio quality dead last.  If anything, there's a little bit of data to show that an over-processed signal is fatiguing and has a negative affect on time spent listening.  But, you see this is a competitive game, and very high pressure.  Stations are demanded to show a continued increase in profit, yet the number of broadcast channels is growing (including stations own HD channels), and the total radio audience is shrinking because of alternative choices.  Stations are under pressure to win each day part, and not having the loudest station on the dial, or at least louder than your competition, is very hard for programmers and managers to deal with.  The data doesn't matter unless it shows their station "winning".  I've sat in many meetings when the latest ratings came out, and the station slipped a fraction of a point.  Fingers get pointed all around the room, including squarely at the engineer for not making the station louder. It's silly, but that's the way it is.