Stereo audio track + Visually-Impared narrative stereo audio track

[Seppia]

Hello

I need to build a DCP with one video track and two stereo audio tracks : one is the main track and one is the Visually-Impaired Descriptive audio track.
As sources, I have one Quicktime ProRes file containing the video track and the main audio stereo track, andI have one WAV file containing only the VI audio tracks.
I don't know how/where I can set the audio matrix of the DCP so that it will be properly playback in theatre (that is :the main track in the theatre room, and the Descriptive track only in the visually-impaired headsets). I assume there is audio routing and meta-data to be set...

Any help ?

[carl]

Set the DCP audio channels to 8 (DCP/Audio tab) then add your ProRes and WAV files. Then select the ProRes file and map its channels to L and R. Select the VI file and map its channels to VI. Details on the audio map are in the manual.

[Carsten]

You should avoid stereo for HI/VI-N tracks. But you can select L or R or downmix to mono in DCP-o-matic.

- Carsten

[Seppia]

Set the DCP audio channels to 8 (DCP/Audio tab) then add your ProRes and WAV files. Then select the ProRes file and map its channels to L and R. Select the VI file and map its channels to VI. Details on the audio map are in the manual.

Hello Carl, thanks for the reply

I was considering doing a 4-channels DCP (since that's what I have, in the form of a dual stereo audio), but the audio matrix states that it is for L R C LFE, and not for dual stereo tracks.

You should avoid stereo for HI/VI-N tracks.

Hello Carsten, ans thanks for the reply.
Where did you find the information about mono HI/VI tracks ? I can't seem to find many informations about HI/VI tracks in the DCP implementation, even in the way the DCP should be named when containing several audio tracks.

Or should I do an OV with the video and just the "normal" stereo track, and then a VF with just the VI audio track ?

But you can select L or R or downmix to mono in DCP-o-matic.

Is mono a requirement for VI Descriptive tracks ? Is it not containing the full mix, but only an alternative Center tracks with narrative ?
If I downmix my stereo VI track, won't it end up too high in terms of audio levels (the sum of L + R) ?

Thanks again

[Carsten]

Both HI and VI-N are always mono, by definition. In order to use HI and/or VI-N, you need to create an 8ch DCP/audio file, with the 'public' 5.1 channels on 1-6, HI on 7, VI-N on 8. All unused channels (in your case, 3-6 ) will contain silence.

If you downmix your VI-N stereo file to mono, you can adjust each channel by e.g. -6dB in the matrix to avoid clipping (right click on matrix connection point).

VI-N should not contain any audio from the public channels (different to HI), but a dedicated spoken description of what is going on on the screen. The main/public audio will be heard through the screen speakers, not the VI-N headphone.

No need to create the VI-N version as a VF.

- Carsten

Bildschirmfoto 2020-03-03 um 00.11.55.png

Bildschirmfoto 2020-03-03 um 00.14.12.png

[Seppia]

Thanks a lot, Carsten, for your detailed explanation !

I ended up doing exactly what you wrote, and it seems to work

But I had to ask the producer for the narrative-voice-only mono track. At first I only had :
- the main stereo normal public full mix
- a stereo track with narrative voice already mixed inside the main mix (hence my initial questions about down-mixing).

So finally my DCP audio tracks are :
- channel 1 and 2 is main stereo mix
- channels 3 4 5 6 are empty (because it's a stereo-only audio main mix, otherwise if my source had been a true 5.1 mix I would have had much less doubts about the number of channels I was supposed to pick)
- channel 7 is also empty (as I don't have a "Hearing Impaired" audio track)
- channel 8 is the mono narrative-voice-only audio-description, my "Visually Impaired" track

From your experience, it "standard" procedure to do a 5.1 DCP (with 4 silent channels) when encoding from a 2.0 stereo audio mix source ?

[Carsten]

Yes. In general, all even numbers are technically allowed - 2,4,6,8,10,12,14,16. As the functional channel assignment is fixed (they were working on a dynamic channel routing scheme for SMPTE DCPs, but at some point withdrew from the idea by practical means), you need to fill unused channels with silence. In cinema, we mostly get 5.1 DCPs. A lot of advertising and many documentaries are done in 2ch. Most arthouse and mainstream features now come in 8ch (5.1+HI/VI-N, or 12ch (7.1+HI/VI-N). Sometimes I see silent channels in trailers for no apparent reason - e.g. I see 12 channels with 7-12 silent in a Disney trailers. I think that is because some companies simply use standard workflows or templates for 7.1+HI/VI-N and don't bother about the unused channels/storage space. Technically, that is completely okay, as long as the DCP is named properly - you can still name it '5.1', even if it contains 12ch.

As a matter of fact, if you actually want to distribute a single (mono) center feature, you have to use 4ch as a minimum (1,2,4 silent). As that is pretty rare, and the savings compared to 5.1 are small, often mono sound is packaged as stereo/2ch (bad), or a 5.1 file, the latter wasting 5 channels.

- Carsten