Hey guys, new here. So I have to mix/master audio for a Music program (with some dialogue) that will be broadcast on TV. Any suggestions for optimally carrying out this type of work in an efficient manner is appreciated.

The TV Channel Broadcast Specs are as follows:


2.0 Compliance with Technical Specification Guide

2.2 All audio content has to comply with technical requirements set out in ATSC A/85RP. Dynamic range must be suitable for television broadcast with target loudness level -24 LKFS. Audio should be mixed such that the measured average loudness of any Lt/Rt mix shall be equivalent to the measured average loudness of 5.1 audio within +/- 2 dB.

2. Accepted File-Based Formats
Due to the wide variety of file-based formats, they will only accept file-based material after it has been thoroughly tested. After the test period, they will determine the acceptability of the file-based material from a supplier.

2.1 Generally, file-based material formats shall be:

2.1.5 Audio Codec: PCM AES, 24 Bit, 48kHz, 8 Channels (see Table 1 SMPTE 320M-B Channel Allocation)

4. Audio Formats & Channel Assignment

4.1 They will only accept material with stereo or 5.1 multi-channel audio. All audio must be able to be output baseband audio following SMPTE 320M-B channel allocation.

4.2 For file-based material, they will accept material which has channel allocation described in metadata.

4.3 If file-based material metadata cannot describe the channel allocation, and for all other material, they will only accept material that conforms to SMPTE 320M-B channel allocation.

4.4 If DV is not provided a stereo down mix is required on Ch 7 and 8.

Table 1 . SMPTE 320M-B Channel Allocation

Channel / Description
Ch 1 / Left
Ch 2 / Right
Ch 3 / Centre
Ch 4 / LFE

Channel / Description
Ch 5 / Left Surround
Ch 6 / Right Surround
Ch 7 /Aux A
Ch 8 / Aux B

5. Described Video

5.1 When available, Described Video should be provided for all material. Described Video will consist of a channel 1 and 2 stereo mix for material with stereo-only audio, or a stereo down-mix derived from channels 1 through 6 for 5.1 material, combined with center-panned voice description audio.

5.2 Described Video audio shall be assigned to SMPTE 320M-B “Aux A” and “Aux B”.

6. Audio Loudness

They will only accept material where loudness is maintained at -24 LKFS within +/- 2dB, based on the ATSC A/85 recommended practice.

6.1 For single-segment material, they will measure the integrated loudness over its full length.

6.2 For multi-segment material, they will measure the integrated loudness of the full length of one randomly selected segment.

6.3 Loudness measurements will be made with an ATSC A/85 compliant audio monitor to ensure compliance.

6.4 For material that contains multi-channel audio, described video, and/or a stereo down-mix, -24 LKFS shall be maintained for each component

6.4.1 The 5.1 multi-channel main programs shall be -24 LKFS

6.4.2 The described video program shall be -24 LKFS.)

8. Audio-Video Synchronization (Lip Sync)

They will only accept material where the audio material is tightly synchronized to the video material.

8. 1 The overall tolerance in audio-video synchronization shall fall within +15ms and -45ms. In other words, audio shall lead video by no more than 15ms and shall be delayed by no more than 45ms from video.

If you understand loudness specs, and if Pro-tools has a loudness meter, then that covers most of the above.

For a background to loudness specs and pointers to other resources, this is a great read.
https://www.soundonsound.com/techniques ... udness-war

If PT doesn't have a loudness meter, there are several out there that should work with-in PT.

You probably already have, but you'll need to check through the ATSC A/85RP document.
This one is a few years old, so there may be a more recent one out there?
https://www.atsc.org/wp-content/uploads ... ness-1.pdf

NewtoPT wrote:2. I will be using Pro Tools for the first time rather than the DAW I am currently on.

This is not the time to fight with a new DAW (and PT is a PITA, IMHO). Use the DAW you want to use and, if absolutely necessary, export the results afterwards.

The Elf wrote:

NewtoPT wrote:2. I will be using Pro Tools for the first time rather than the DAW I am currently on.

This is not the time to fight with a new DAW (and PT is a PITA, IMHO). Use the DAW you want to use and, if absolutely necessary, export the results afterwards.

Do any other DAWs have spotting from an EDL in the same way PT does? Genuinely asking. That's the quickest way to set up these types of sessions,

ken long wrote:Do any other DAWs have spotting from an EDL in the same way PT does? Genuinely asking. That's the quickest way to set up these types of sessions,

I don't know the term "spotting from an EDL", is it a ProTools thing? I remember using EDLs from video editors to comform audio into a DAW, but then OMFs and later AAFs took over.
Not sure if it's the same thing, but Cubase and Logic will happily import OMF/AAF files from Avids or other video editors.

Kwackman wrote:

ken long wrote:I don't know the term "spotting from an EDL", is it a ProTools thing?

One of Pro Tools editing modes which allows fast and accurate placement of files based on decision lists. It is part of a standard workflow in post.

Most decent DAWs support placing of clips by embedded timecode (when using BWAV files).

NewtoPT wrote:Hey guys, new here.

Welcome to the SOS forums!

2. I will be using Pro Tools for the first time rather than the DAW I am currently on.

Why? That sounds like stress you really don't need!

The broadcast specs may sound complicated, but they really aren't.

In your situation, with an off-line mix, I'd just assemble and mix the show in your normal way, and then check the Integrated Loudness of the whole thing once you've finished. You can then dial in an overall level shift to make it conform with the -24LKFS target and export... Much easier than trying to mix while simultaneously hitting the loudness target.

Obviously, you'll need to be aware of the (slightly) restricted dynamic range requirements for TV, and of the appropriate relative balance between interview voices and music peak levels. You can figure that out from using a previous similar programme as a reference easily enough.... but in general I'd aim for average voice levels to be no more than 8dB lower than the music peaks, and often 4dB lower is more acceptable -- it depends on the kind of show and viewer expectations.

Are you mixing this in stereo or surround? If the latter, be aware that the LFE channel is not included in stereo/mono downmixes, so DON'T stick you kick drums and basses in there!

In fact, it's best not to stick anything at all in the LFE channel. It's simply not necessary and only causes confusion and incompatibilities. The five main channels all have a full bass responses and all get included in stereo mix-downs AND any bass in them will be routed to people's subwoofers at home.

ken long wrote:

Kwackman wrote:

ken long wrote:I don't know the term "spotting from an EDL", is it a ProTools thing?

One of Pro Tools editing modes which allows fast and accurate placement of files based on decision lists. It is part of a standard workflow in post.

Sounds similar to the OMF AAF thing? Embedded is how our workflow (non ProTools) did it, but unembedded would put associated WAV files to match the EDL part of the OMF or AAF, so maybe the same thing?
I've very little post experience of using Cubase or Logic in post as I used them for music, but they handled audio exports from video editors with no issues. Even got the handles, which are a big bonus when fixing dodgy edits!

To the OP, which DAW is your usual weapon?

ken long wrote:

The Elf wrote:

NewtoPT wrote:2. I will be using Pro Tools for the first time rather than the DAW I am currently on.

This is not the time to fight with a new DAW (and PT is a PITA, IMHO). Use the DAW you want to use and, if absolutely necessary, export the results afterwards.

Do any other DAWs have spotting from an EDL in the same way PT does? Genuinely asking. That's the quickest way to set up these types of sessions,

As has been said by others above, most decent DAWs can spot to video. This idea of PT being the only one to do these things is a stubbornly persistent myth!

First off, thank you guys so much for the quick responses. To be frank, this is all very new to me. However, because of past gigs I've done with this contractor, they have put their trust in me for this project, which I am confident I can complete once I learn the different aspects. One of my past gigs consisted of mixing live performance stems of multiple songs for an artist, for a TV broadcast. It turned out great, sounded wonderful on TV, everyone was please - but, in that case there was an editor that matched my mix to TV broadcast specs.

Some relieving news - the client wants to get all the video editing approved before audio is touched, so I have about a week.

prior to mastering the entire audio for the TV special, I will be mixing a bunch of the performances for the special. Is there any beforehand Loudness monitoring (True Peak, LKFS, etc.) I should be doing while I'm conducting each of the mixes?

If so, any advice on how to generally approach it would be great.

Kwackman wrote:If you understand loudness specs, and if Pro-tools has a loudness meter, then that covers most of the above.

For a background to loudness specs and pointers to other resources, this is a great read.
https://www.soundonsound.com/techniques ... udness-war

If PT doesn't have a loudness meter, there are several out there that should work with-in PT.

You probably already have, but you'll need to check through the ATSC A/85RP document.
This one is a few years old, so there may be a more recent one out there?
https://www.atsc.org/wp-content/uploads ... ness-1.pdf

Thank you Kwackman, I will look into the PDF.

ken long wrote:

Kwackman wrote:

ken long wrote:I don't know the term "spotting from an EDL", is it a ProTools thing?

One of Pro Tools editing modes which allows fast and accurate placement of files based on decision lists. It is part of a standard workflow in post.

Thanks, I will look into spotting from an EDL and come back with any questions I may have.

Hugh Robjohns wrote:Most decent DAWs support placing of clips by embedded timecode (when using BWAV files).

Unfortunately, my DAW doesn't have an SMPTE Time Code mode. The DAW is great for all other music purposes for the most part, but in this situation I'm almost certain it just simply can't do the job.

Hugh Robjohns wrote:

NewtoPT wrote:Hey guys, new here.

Welcome to the SOS forums!

2. I will be using Pro Tools for the first time rather than the DAW I am currently on.

Why? That sounds like stress you really don't need!

The broadcast specs may sound complicated, but they really aren't.

In your situation, with an off-line mix, I'd just assemble and mix the show in your normal way, and then check the Integrated Loudness of the whole thing once you've finished. You can then dial in an overall level shift to make it conform with the -24LKFS target and export... Much easier than trying to mix while simultaneously hitting the loudness target.

Obviously, you'll need to be aware of the (slightly) restricted dynamic range requirements for TV, and of the appropriate relative balance between interview voices and music peak levels. You can figure that out from using a previous similar programme as a reference easily enough.... but in general I'd aim for average voice levels to be no more than 8dB lower than the music peaks, and often 4dB lower is more acceptable -- it depends on the kind of show and viewer expectations.

Are you mixing this in stereo or surround? If the latter, be aware that the LFE channel is not included in stereo/mono downmixes, so DON'T stick you kick drums and basses in there!

In fact, it's best not to stick anything at all in the LFE channel. It's simply not necessary and only causes confusion and incompatibilities. The five main channels all have a full bass responses and all get included in stereo mix-downs AND any bass in them will be routed to people's subwoofers at home.

Hey Hugh, really appreciate you welcoming me. My DAW doesn't have an SMPTE Time Code mode unfortunately. I feel like PT is the most logical option and have already begun the learning process for it.

I will be mixing in Stereo for this project.

Re:
"You can then dial in an overall level shift to make it conform with the -24LKFS target and export... Much easier than trying to mix while simultaneously hitting the loudness target".

Hugh, would what you're referring to by any chance be something like what NUGEN LM Correct 2 does? I did a quick test on a .wav file of a song with - Analyze, and then Render.
The new Rendered Audio file is hitting around -23 Integrated, but only -12.3 True Peak. With an LRA of 4.3 - 7 once activated when playing the new file.


Thanks again guys. I have demo's so far of NUGEN VisLM, LM Correct 2, and ISL2 and about to purchase as based on my research this seems like the way to go. If anyone has advice on the optimal way, in my situation, to be using these 3 in tandem that would be great.

We give our stuff (audio and video) to a mastering house who are used to dealing with this specific broadcaster.

If you are dealing with someone like the BBC, they are going to reject, reject and reject again for the most absurd reasons, so it's best to give that to people who are used to dealing with the new generation of jobsworths that seem to have infested many broadcasters.

Updated Test numbers

2 minute song, Rendered LM Correct 2 file: -23.8 Integrated, -12.3 Max True Peak, 12.3 LRA

and then tested separately after

12 minute Dialogue, Rendered LM Correct2 file: -24.0 Integrated, - 2 Max True Peak, 12.1 LRA

NewtoPT wrote:Hugh, would what you're referring to by any chance be something like what NUGEN LM Correct 2 does?

Hugh will certainly reply, but he most likely simply meant "lower the fader" .

Loudness correction is simply attenuation, so if your mastered stero track is hitting -10LUFS average, you simply lower the master fader (or the mix fader, depending on how your mastering chain is set up) and replay the whole track (or re-run the loudness measurement tool) until you hit within -22 to -26 LUFS average (that is, -24 with the +/- 2dB slack).

A common free plugin for checking integrated LUFS is the Youlean Loudness Meter

You can surely use the LM correct as well.

NewtoPT wrote:Is there any beforehand Loudness monitoring (True Peak, LKFS, etc.) I should be doing while I'm conducting each of the mixes?

No, just mix each piece as you normally would -- that's why they've asked you to do it, after all. Then when each mix is finished you can run it through a loudness meter and adjust the overall mix level up or down as necessary to comply with the ATSC requirements for Integrated loudness and True Peak.

Hugh, would what you're referring to by any chance be something like what NUGEN LM Correct 2 does?

Yes, Nugen LM correct is a good solution.

I did a quick test on a .wav file of a song with - Analyze, and then Render. The new Rendered Audio file is hitting around -23 Integrated, but only -12.3 True Peak. With an LRA of 4.3 - 7 once activated when playing the new file.

The only figure that really matters for the broadcaster is the Integrated loudness -- provided the True Peak is under -1dBTP. The fact that your True Peak is -12dBTP implies you've applied a lot of compression/limiting, which is also confirmed by the relatively small LRA (dynamic range) figure... But I'm guessing 'squashed' is the style of the music you're producing.