SMPTE / ST377-4-Revision-and-ST377-41

1 stars 0 forks source link

Audio Description with Control Track #5

Open ghost opened 4 years ago

ghost commented 4 years ago

In several European countries, AD is distributed as

We propose to add labels to the dictionary in 377-41 to support these channels.

Background:

BBC R&D registered MCA labels in smpte-ra as follows: 060e2b34.0401010d.0302010f.00000000 = ch Visually Impaired Narrative Channel 060e2b34.04010101.0d010801.01010100 = ch Audio Description Studio Data Channel 060e2b34.04010101.0d010801.01020000 = sg Audio Description Studio Signal Data Channel

Meanwhile ST 2067-8 also defined '060e2b34.0401010d.03020220.0c000000' = ch Audio Channel VIN

A question that rears its head in this context is whether the AD and the Studio Signal should form part of the same sound-field. An argument could be made that since the control signal should never be routed to a loudspeaker, it therefore should not be part of the sound-filed that contains the audio description. On the other hand, "groups of sound-field groups" suggest that at least in name, the control signal would still be in a "sound-field" so perhaps this is a rather philosophical debate?

dtatut commented 4 years ago

@Ulrich-DPP : if the control signal is transported in an audio channel as if it was PCM data, then it should be treated like an audio channel, even if it does not go directly to a speaker. Many control signals and other kinds of data are transported this way.

cseidel2 commented 4 years ago

Brian and I propose adding a MCA Content Value (Table 2 in 377-41) as follows:

Name: "Descriptive Video Control" Symbol: "DVC" Description: "Control signal associated with Descriptive Video to facilitate automated mixing of the program with DV audio."

Note that the term Descriptive Video is used in 377-41 for VIN and Audio Description.

To complete this, our intent is to add the channel and soundfield group names suggested in the comment above to 377-42.

thomasheritage commented 3 years ago

Thanks @Ulrich-DPP there is certainly a use case for this and it would be good for it to be supported.

FYI -- Note that the ULs given above are not entirely correct.

After discussion with Brian and @dtatut, to support this: