Advertisement
English transcript
- Hello everyone.
My name is Giovanni from the
Telestream closed caption team.
I came from CPC Closed Captioning and I am working
here with Telestream for a few months now and we have
some exciting things to show today.
The topic of the discussion today is called
"Verifying Closed Caption Data in a File Based Workflow" and
this particular presentation was the same one that was
presented at NAB conference in 2014,
a few months ago.
It was very well attended and I'm very grateful that
many of you have taken time out of your day to take a
look at what we have to present.
So, let's get on with it.
Let's, let's move forward and at the end I certainly
would appreciate any questions and for those who
know me and can reach out to me afterward as well.
So, let's get going.
So, quality control CC in file data so that as file
based workflows become more mature in our industry,
there's a need to take a look at those files and
analyze them, not only for audio and video quality,
for metadata quality, especially captioning data.
So, let's take a look at some of the topics we're
going to cover today.
The first one is where is the CEA-608 and 708 caption
data stored and what are the two protocols that are used
in the industry in Canada, the U.S.,
and in Mexico and we need to understand where these are
stored in the file formats and we'll take a look at
that in a second.
We're also going to talk about typical limitation of
software verification; there's software that can
just take a look at caption, whether it's there or not.
We'll take a look at some of those limitations there.
We'll talk about advanced 708 caption attributes and
what they are and where to look for them,
and how to check for them.
Data is stored alongside captions that's not
necessarily caption data.
So, we'll talk about other data that can be in the
video files in your server that you can check for.
And we can also check about verifying not only
captioning for a Television broadcast,
but also for Internet captioning.
And, of course, checking for common caption errors in a
file based workflow.
So, we'll talk about these topics and one by one and
let's get going with the first one.
Where is the captioning stored?
And also, we'll talk about emulating captioning viewing
experience the same way your viewers at home do and so
let's get going with that.
So, I have some sample video files that have captioning
and we're going to talk about where those captions
live and the interesting thing is that they could be
in different places in the file.
So, we'll take a look at some samples.
There's not every single sample,
but some of those most typical ones
we're going to cover right now.
So, let's talk about VBI captioning.
In this case, we have an XDCAM,
IMX30.
OK, it's got a specific MPEG 2 codecs,
but it-- VBI captioning refers to the actual
blanking lights otherwise known as line 21 in the
broadcast field.
And you can see in this particular example that the
captions are actually stored in a particular
vertical blanking interval.
So, that's one style of caption.
It's very challenging software verification 'cause
you have to be able to read that,
those particular lines, and extract the caption data and
make sure it's correct.
And it's not just limited to one caption channel,
there could be several languages in the VBI and so
we'll talk about how we can look at that.
And then, I guess another way to look at captioning is
inside of a track.
This case, we have an XDCAM,
MXF.
OK, it's 422 MPEG 2 codec.
It could be 1080i.
It could be 720.
So, it's high definition and the SMPTE Track is actually
in the MXF wrapper.
It's not part of the video essence inside of the file.
And it could contain a variety of different data.
So, we should be able to understand,
to look at that and, typically this is a little
easier to read than the standard def.
VBI.
Let's look at another sample here.
There's also something called user data.
So, let's say you have an MPEG 2 video file,
whether it's wrapped in an .MOV or whether it's wrapped
in MXF, but the actual data for captioning is stored in
the essence video, meaning in the actual MPEG 2
elementary stream inside of the data.
And this user data is a little harder to read
because it's actually inside,
you have to read it frame by frame within the
MPEG 2 structure.
So, this is very common and, in fact,
before MXF was really popular,
this seemed to be the standard and it's still the
standard for ATSC over the air captioning.
It's stored in the MPEG 2 user data.
OK, let's talk about something,
another file format that's more advanced and this is,
it's still SMPTE 436M, but this is called
an OP Atom, AS02.
And the way OP Atom is stored is,
is, is kind of, or structured rather,
is kind of different than other MXFs that are
self-contained.
You have different MXF files in different folders and the
reference MXF.
One of the MXFs is actually just the captioning.
It's an MXF that just contains a SMPTE 436M CC
Track and it links to the rest of the video.
And just like the XDCAM variety of MXF OP1A,
it can, it just has a variety of different data
inside of that track, including 608 data and
V-chip and XDS.
All right, let's look at a few more.
This is an, more of a legacy captioning standard made
popular by Panasonic, DVCAM,
or DV 25, DV 50.
It's a standard def.
It could be an MOV.
It could be an AVI or it could be a .DV file.
In those older days, that was really popular,
but still some broadcasters use this format.
And the way it's stored is, it's called a
Vertical Auxiliary Data, or VAUX data.
This is important because it's some,
somewhere still in the essence video.
So, it's stored within the actual DV codec
inside of the file.
So, I'm going to adjust my microphone so that it's
easier to hear.
If you do have trouble hearing,
please let us know.
All right, let's move on to the next.
So, user data CC essence.
This is actually inside of the MOV wrapper so let's say
you have a ProRes, but the ProRes actually,
you know, in the before closed caption tracks were popular,
inside ProRes MOV, there were some devices that
could capture video and put the captions in the ProRes
elementary stream and this becomes confusing to some
people because we have ProRes(es) with different
type of captions.
If you have a KiPro, for example,
AJA, you can actually capture the video right from
a caption source, via HD-SDI,
and the actual CC captions is not a caption track.
It's actually embedded into the ProRes codec.
This is considered a legacy standard.
Going forward, this is no longer going to be used by
many people, but it is something that's out there.
And there's got to be a way to verify it.
And now, the cousin of that is actually,
same thing, it's a ProRes, but it's actually in a
separate caption track.
If you look at the little graphic that I have on the
right, there's actually a closed caption track in the
structure of the MOV.
So, previously, let's go back one slide,
there's no closed caption track.
That means the captions are in the video track,
but you can also have a closed caption track in there
and that's a lot easier to read and if you
play it back with QuickTime, you'll be able to turn the
captions on and off so you can verify that they're there.
All right, let's, let's move on with that and,
one thing that I do want to talk about before I move on
with this is closed captions in ProRes can support V-chip
and XDS as long as they're in the closed caption track,
but you cannot support AFD,
AFD-- Active Format Description
is not supported in the closed caption track
for QuickTime.
OK?
So, let's talk about one other way to check for
captions and that's when you have a separate file.
Many times you have a file format called .SCC.
If you deal with captions, you know that this is a very
popular file format, but what's interesting about
that is, before you transcode it,
or before you deliver it, you have to make sure that
the .SCC matches your video.
So, there's ways to playback your video with .SCC and
just make sure it matches before you deliver it.
And many of you who deliver to say,
something like iTunes, you deliver an .SCC file along
with your ProRes.
You have to be able to verify it.
If you don't verify it, they could fail it and tell you,
OK, it's not in sync or, or just drifting,
or it's just not starting at the right time.
So, it's really important if the captions aren't baked
into the video, that you can check them with,
with a video player that can,
you know, kind of show you a preview of what the .SCC is
going to look like once it's married together
with the video.
All right, let's move on to some typical limitations of
software verification.
That's the second topic.
It does not detect multiple services.
So, if you have a more than one language in captioning.
Let's say you have English and French or English and
Spanish, which many broadcasters do,
if you check it with, say QuickTime and QuickTime can
show you the captions, there's no way to check for
other services.
It's going to give you the one language.
So, keep in mind that captioning is not limited to
one language and it needs to be verified for
multiple languages.
And, as I mentioned earlier, VBI,
the Vertical Blanking Intervals,
otherwise known as line 21, that's difficult to detect,
OK?
Usually, the only way to do it is you have buy
all this hardware.
You know, you have to play it out to a scope or some
sort of monitor that can decode captions.
With software, it's very difficult to read that data
right on your desktop.
OK, next is detects only data,
which could be incomplete.
So, there's, there's software out there
where you can open it up.
Let's say, like for an example there is MediaInfo.
You can open up a caption file,
or video file that has captions in it and it'll
just say yes, there's some captions there.
It's 608, it's got 708, and you say,
"OK, great.
It's ready to go.
The captions are there."
Well, they might not be complete.
Maybe half of the captions are missing.
Or maybe the captions are actually drifting or
not in sync.
That doesn't tell you whether the caption data is
complete.
There has to be a way to verify all the caption data
inside of the video file.
And, not all codecs and caption protocols are
supported.
Traditionally, I've had like 4 different video players,
and 4 different software to be able to check
caption files or caption video files because it seems like one
supports MXF, the other one supports MOV,
the other one supports AVI.
It's, it's kind of like convoluted and unfortunately
for broadcasters that deal with a lot of different file
formats, it can be very daunting to check for all
the different caption files.
And the same thing goes with hardware,
There's hardware that can play back QuickTime,
and there's hardware that can play back MPEG 2,
but it's seems like it's hard to just,
you know, be able to verify all the different codecs and
captions protocols, as I showed earlier.
OK, so let's move on to the next one.
Sync issues of captions may not be detected.
As I mentioned, there's, there's something called the
caption drift.
And with captioning, drift is basically,
the captions start on time and over an hour they'll
drift like, say by 3 seconds early or late.
That has to be detected prior to being able to be
delivered.
And many softwares doesn't, don't allow you to do that.
And positioning of caption texts onscreen are
not verified.
That's another thing.
So, let's say you have a caption.
It's in sync,
and what, well the problem is,
it's blocking half of lower thirds.
So, you have to be able to play it back and verify that
you know, the positioning is properly done and there's
new legislation that's coming next year,
from the Federal Communications Commission,
FCC in the U.S., that is going to check for these
kind of things.
So, it's important to verify that the positioning of your
captioning on the text is properly done for
accessibility.
And, of course, there's sidecar caption files for
Internet and TV transcoding.
You know, you've heard of SMPTE timed text,
WebVTT, DFXP, SCC.
So, these files you're going to deliver,
whether you're going to put it on your website or
or whether you're going to transcode it with,
say, Telestream Vantage.
Well, you have to be able to check those files to make
sure that they're going to match before you transcode,
and that's kind of tricky.
You have to have the right tools to do that.
OK, so let's, someone asked me a question here.
You know Extreme Reach.
Extreme Reach uses a variety of different things.
I would contact them about it.
I know that they were, they were using MPEG 2 a lot,
and now they're also supporting QuickTime.
So, it's kind of changes now,
but yeah they use a lot of different things.
OK, so let's talk about Advanced CEA-708 features.
OK?
And, 708 is the standard for high definition captioning.
High definition as it's broadcast in the U.S.
contains 608.
It's also contains 708.
And font styles and colors.
If you've ever been to a public place,
let's say it's a place where they serve food or they're
showing a sports game on the,
you know, the flat panel monitors.
After work, people, you know,
populate these places.
You'll see that sometimes the font is different.
It's not, you know, the text,
the white text on the back, black background with a
box on it.
It might be translucent.
It might have a different font.
It looks different than the regular
style of closed captions.
That's Advanced 708.
OK?
Font styles and colors.
They also can do translucency,
as I mentioned, and opacity parameters.
They support Asian languages.
Although, that's part of the 708 specification,
there's, there's only, there's not any TVs I've
seen that can support that, but TVs do support
translucency and opacity and they supports different
fonts and colors.
So, that brings me to my next point,
and that is, in your professional broadcast
facility, you have scopes.
You have different types.
Different companies make different HD-SDI monitoring
that can do, decode closed captioning,
but they do not show these Advanced 708 features.
I have not seen one that does,
and when you air that on TV, the people,
the folks at home-- they do have that feature.
So, really, these professional scopes may be
lacking in the ability to verify these font styles and
colors and translucencies that you can use with the
Advanced 708.
There's also 16x9 positioning.
There's no high definition 16x9 so you have more
real estate to move the captions from left to right.
And you can language tag.
So, if you have multiple languages,
you can tag language with English,
French, Spanish, whatever you need and that's also
part of the Advanced 708 features.
You should be able to verify that tagging with software only,
and of course, you can have up to 63 services of
multiple languages, although because of bandwidth
constraints, we've only been able to add 6 languages to
708 before you know it, it's too much bandwidth to
be able to actually air it.
And, of course, there's Drop Shadows.
These are all Advanced 708 features that are,
you should be able to verify.
OK?
So, verifying data stored with captioning.
Let's talk about that.
There's other data besides captioning and some of that
data could be, XDS and V-chip.
For example, in 608, you can have,
you know, a particular rating or
program identification and how long the program is,
total running time.
All these things you can verify as well,
and that's important to verify without having to use
some sort of hardware solution,
doing with software only.
Active format description, different aspect ratios,
that's important to verify before you air it.
Time code track, OK, that's important if
you're going to transcode.
If you're going to transcode,
you've got to make sure that your captions match your
time code track and your video file.
The next item that I want to cover is
verifying formatting for Internet captioning.
So, this is one legislation that's called the CVAA.
I know, Communication Video Accessibility Act of 2010 by
the FCC and it calls for being able to convert
captions from TV to captioning on the Internet.
And this kit example on the graphic,
I'm showing YouTube video player,
and you can see YouTube video player,
you can actually upload an SCC.
You can upload and actual caption file for Television
and YouTube will actually do its best to
keep the formatting.
For example, on TV, there's something called
split captions where, when two people are talking you have
captions on the one end and on the other end,
just like on the graphic.
So, that is maintained.
So, it's important that you verify that formatting.
So, that you know your web team is keeping the captions
the same way they were aired.
Roll up captions should be kept that way.
Pop on captions should be kept that way.
Positioning on the screen so they don't block a lower
third, they should be kept that way as well.
And there's different caption file formats
for the Internet.
So, verification of the positioning is critical as well
when you're uploading captioning to the Internet.
All right, the next step is checking for
common CC errors.
So, we talked about drifting out of sync captioning.
That happens, usually when you're converting
the frame rate.
Many online captioning videos are 2398,
but it was aired in 29.97, or 60i,
or 59.94, 720p, but when you upload the original file
that was shot in film that's 23.98 and often times that's
not going to match.
So, it's important that you check that drift.
As I mentioned, check to make sure that the text
formatting is following the standards of accessibility.
There's times also when 708 is missing and
only 608 is found.
So, you have an HD MPEG 2.
And let's say you have, you know,
some software that says, "OK,
well give an SCC file and we'll transcode
into MPEG 2."
Well, what happens is, you, you transcode it,
you get the MPEG2.
You send it to your TV station and they fail it.
They say that, "We're just not seeing 708.
Only 608 is found."
So, it's important that you check your,
your MPEG 2 or whatever file you're sending,
whether its to Extreme Reach or to a broadcaster,
to PBS to make sure that both 608 and 708 are there
and that applies to High Definition specifically.
Its also important that the syntax of the caption file
is compatible with the transcode engine so,
you know, you may get an SCC file that was made,
you know, someone, I've seen someone make an SCC file by
hand, manually, and it just doesn't work.
And, so just because you have an SCC,
you have got to make sure that it's actually compliant
and it matches the right syntax,
otherwise you're going to get a lot of problems.
And also, of course, the same thing applies for the
syntax of the file for Internet captioning players.
So, you have to check these things.
Spelling and transcription mistakes,
of course, how can you check to say,
see if the actual captioning is right,
unless you actually have some sort of player to,
to read the captions and then spot check them before
you air them.
This is an interesting one.
Let's say you air something with captioning,
but then you are going to put that on the Internet or
you're going to put it on VOD and that's mean that you
have to edit that video, and,
but the captioning is not necessarily
matching the edit.
Once you edit the video, the captions no longer match and
so, just, it's important to make sure that an edited
video is checked against the captioning.
Otherwise, you're going to get like a video you're
playing back on the Internet,
or on VOD that, halfway through the video,
all of a sudden the captions are not matching at all.
So, it's important to check that once you
edit your video.
All right, next let's talk about emulating Television
captioning viewing experience.
So, what I mean by viewing experience,
I'm talking about the viewers at home.
You should have the ability to view your video the same
way that the people at home are going to see it
prior to airing it, and you have to make sure
you support Advanced 708, like the fonts.
You have to make
…
My name is Giovanni from the
Telestream closed caption team.
I came from CPC Closed Captioning and I am working
here with Telestream for a few months now and we have
some exciting things to show today.
The topic of the discussion today is called
"Verifying Closed Caption Data in a File Based Workflow" and
this particular presentation was the same one that was
presented at NAB conference in 2014,
a few months ago.
It was very well attended and I'm very grateful that
many of you have taken time out of your day to take a
look at what we have to present.
So, let's get on with it.
Let's, let's move forward and at the end I certainly
would appreciate any questions and for those who
know me and can reach out to me afterward as well.
So, let's get going.
So, quality control CC in file data so that as file
based workflows become more mature in our industry,
there's a need to take a look at those files and
analyze them, not only for audio and video quality,
for metadata quality, especially captioning data.
So, let's take a look at some of the topics we're
going to cover today.
The first one is where is the CEA-608 and 708 caption
data stored and what are the two protocols that are used
in the industry in Canada, the U.S.,
and in Mexico and we need to understand where these are
stored in the file formats and we'll take a look at
that in a second.
We're also going to talk about typical limitation of
software verification; there's software that can
just take a look at caption, whether it's there or not.
We'll take a look at some of those limitations there.
We'll talk about advanced 708 caption attributes and
what they are and where to look for them,
and how to check for them.
Data is stored alongside captions that's not
necessarily caption data.
So, we'll talk about other data that can be in the
video files in your server that you can check for.
And we can also check about verifying not only
captioning for a Television broadcast,
but also for Internet captioning.
And, of course, checking for common caption errors in a
file based workflow.
So, we'll talk about these topics and one by one and
let's get going with the first one.
Where is the captioning stored?
And also, we'll talk about emulating captioning viewing
experience the same way your viewers at home do and so
let's get going with that.
So, I have some sample video files that have captioning
and we're going to talk about where those captions
live and the interesting thing is that they could be
in different places in the file.
So, we'll take a look at some samples.
There's not every single sample,
but some of those most typical ones
we're going to cover right now.
So, let's talk about VBI captioning.
In this case, we have an XDCAM,
IMX30.
OK, it's got a specific MPEG 2 codecs,
but it-- VBI captioning refers to the actual
blanking lights otherwise known as line 21 in the
broadcast field.
And you can see in this particular example that the
captions are actually stored in a particular
vertical blanking interval.
So, that's one style of caption.
It's very challenging software verification 'cause
you have to be able to read that,
those particular lines, and extract the caption data and
make sure it's correct.
And it's not just limited to one caption channel,
there could be several languages in the VBI and so
we'll talk about how we can look at that.
And then, I guess another way to look at captioning is
inside of a track.
This case, we have an XDCAM,
MXF.
OK, it's 422 MPEG 2 codec.
It could be 1080i.
It could be 720.
So, it's high definition and the SMPTE Track is actually
in the MXF wrapper.
It's not part of the video essence inside of the file.
And it could contain a variety of different data.
So, we should be able to understand,
to look at that and, typically this is a little
easier to read than the standard def.
VBI.
Let's look at another sample here.
There's also something called user data.
So, let's say you have an MPEG 2 video file,
whether it's wrapped in an .MOV or whether it's wrapped
in MXF, but the actual data for captioning is stored in
the essence video, meaning in the actual MPEG 2
elementary stream inside of the data.
And this user data is a little harder to read
because it's actually inside,
you have to read it frame by frame within the
MPEG 2 structure.
So, this is very common and, in fact,
before MXF was really popular,
this seemed to be the standard and it's still the
standard for ATSC over the air captioning.
It's stored in the MPEG 2 user data.
OK, let's talk about something,
another file format that's more advanced and this is,
it's still SMPTE 436M, but this is called
an OP Atom, AS02.
And the way OP Atom is stored is,
is, is kind of, or structured rather,
is kind of different than other MXFs that are
self-contained.
You have different MXF files in different folders and the
reference MXF.
One of the MXFs is actually just the captioning.
It's an MXF that just contains a SMPTE 436M CC
Track and it links to the rest of the video.
And just like the XDCAM variety of MXF OP1A,
it can, it just has a variety of different data
inside of that track, including 608 data and
V-chip and XDS.
All right, let's look at a few more.
This is an, more of a legacy captioning standard made
popular by Panasonic, DVCAM,
or DV 25, DV 50.
It's a standard def.
It could be an MOV.
It could be an AVI or it could be a .DV file.
In those older days, that was really popular,
but still some broadcasters use this format.
And the way it's stored is, it's called a
Vertical Auxiliary Data, or VAUX data.
This is important because it's some,
somewhere still in the essence video.
So, it's stored within the actual DV codec
inside of the file.
So, I'm going to adjust my microphone so that it's
easier to hear.
If you do have trouble hearing,
please let us know.
All right, let's move on to the next.
So, user data CC essence.
This is actually inside of the MOV wrapper so let's say
you have a ProRes, but the ProRes actually,
you know, in the before closed caption tracks were popular,
inside ProRes MOV, there were some devices that
could capture video and put the captions in the ProRes
elementary stream and this becomes confusing to some
people because we have ProRes(es) with different
type of captions.
If you have a KiPro, for example,
AJA, you can actually capture the video right from
a caption source, via HD-SDI,
and the actual CC captions is not a caption track.
It's actually embedded into the ProRes codec.
This is considered a legacy standard.
Going forward, this is no longer going to be used by
many people, but it is something that's out there.
And there's got to be a way to verify it.
And now, the cousin of that is actually,
same thing, it's a ProRes, but it's actually in a
separate caption track.
If you look at the little graphic that I have on the
right, there's actually a closed caption track in the
structure of the MOV.
So, previously, let's go back one slide,
there's no closed caption track.
That means the captions are in the video track,
but you can also have a closed caption track in there
and that's a lot easier to read and if you
play it back with QuickTime, you'll be able to turn the
captions on and off so you can verify that they're there.
All right, let's, let's move on with that and,
one thing that I do want to talk about before I move on
with this is closed captions in ProRes can support V-chip
and XDS as long as they're in the closed caption track,
but you cannot support AFD,
AFD-- Active Format Description
is not supported in the closed caption track
for QuickTime.
OK?
So, let's talk about one other way to check for
captions and that's when you have a separate file.
Many times you have a file format called .SCC.
If you deal with captions, you know that this is a very
popular file format, but what's interesting about
that is, before you transcode it,
or before you deliver it, you have to make sure that
the .SCC matches your video.
So, there's ways to playback your video with .SCC and
just make sure it matches before you deliver it.
And many of you who deliver to say,
something like iTunes, you deliver an .SCC file along
with your ProRes.
You have to be able to verify it.
If you don't verify it, they could fail it and tell you,
OK, it's not in sync or, or just drifting,
or it's just not starting at the right time.
So, it's really important if the captions aren't baked
into the video, that you can check them with,
with a video player that can,
you know, kind of show you a preview of what the .SCC is
going to look like once it's married together
with the video.
All right, let's move on to some typical limitations of
software verification.
That's the second topic.
It does not detect multiple services.
So, if you have a more than one language in captioning.
Let's say you have English and French or English and
Spanish, which many broadcasters do,
if you check it with, say QuickTime and QuickTime can
show you the captions, there's no way to check for
other services.
It's going to give you the one language.
So, keep in mind that captioning is not limited to
one language and it needs to be verified for
multiple languages.
And, as I mentioned earlier, VBI,
the Vertical Blanking Intervals,
otherwise known as line 21, that's difficult to detect,
OK?
Usually, the only way to do it is you have buy
all this hardware.
You know, you have to play it out to a scope or some
sort of monitor that can decode captions.
With software, it's very difficult to read that data
right on your desktop.
OK, next is detects only data,
which could be incomplete.
So, there's, there's software out there
where you can open it up.
Let's say, like for an example there is MediaInfo.
You can open up a caption file,
or video file that has captions in it and it'll
just say yes, there's some captions there.
It's 608, it's got 708, and you say,
"OK, great.
It's ready to go.
The captions are there."
Well, they might not be complete.
Maybe half of the captions are missing.
Or maybe the captions are actually drifting or
not in sync.
That doesn't tell you whether the caption data is
complete.
There has to be a way to verify all the caption data
inside of the video file.
And, not all codecs and caption protocols are
supported.
Traditionally, I've had like 4 different video players,
and 4 different software to be able to check
caption files or caption video files because it seems like one
supports MXF, the other one supports MOV,
the other one supports AVI.
It's, it's kind of like convoluted and unfortunately
for broadcasters that deal with a lot of different file
formats, it can be very daunting to check for all
the different caption files.
And the same thing goes with hardware,
There's hardware that can play back QuickTime,
and there's hardware that can play back MPEG 2,
but it's seems like it's hard to just,
you know, be able to verify all the different codecs and
captions protocols, as I showed earlier.
OK, so let's move on to the next one.
Sync issues of captions may not be detected.
As I mentioned, there's, there's something called the
caption drift.
And with captioning, drift is basically,
the captions start on time and over an hour they'll
drift like, say by 3 seconds early or late.
That has to be detected prior to being able to be
delivered.
And many softwares doesn't, don't allow you to do that.
And positioning of caption texts onscreen are
not verified.
That's another thing.
So, let's say you have a caption.
It's in sync,
and what, well the problem is,
it's blocking half of lower thirds.
So, you have to be able to play it back and verify that
you know, the positioning is properly done and there's
new legislation that's coming next year,
from the Federal Communications Commission,
FCC in the U.S., that is going to check for these
kind of things.
So, it's important to verify that the positioning of your
captioning on the text is properly done for
accessibility.
And, of course, there's sidecar caption files for
Internet and TV transcoding.
You know, you've heard of SMPTE timed text,
WebVTT, DFXP, SCC.
So, these files you're going to deliver,
whether you're going to put it on your website or
or whether you're going to transcode it with,
say, Telestream Vantage.
Well, you have to be able to check those files to make
sure that they're going to match before you transcode,
and that's kind of tricky.
You have to have the right tools to do that.
OK, so let's, someone asked me a question here.
You know Extreme Reach.
Extreme Reach uses a variety of different things.
I would contact them about it.
I know that they were, they were using MPEG 2 a lot,
and now they're also supporting QuickTime.
So, it's kind of changes now,
but yeah they use a lot of different things.
OK, so let's talk about Advanced CEA-708 features.
OK?
And, 708 is the standard for high definition captioning.
High definition as it's broadcast in the U.S.
contains 608.
It's also contains 708.
And font styles and colors.
If you've ever been to a public place,
let's say it's a place where they serve food or they're
showing a sports game on the,
you know, the flat panel monitors.
After work, people, you know,
populate these places.
You'll see that sometimes the font is different.
It's not, you know, the text,
the white text on the back, black background with a
box on it.
It might be translucent.
It might have a different font.
It looks different than the regular
style of closed captions.
That's Advanced 708.
OK?
Font styles and colors.
They also can do translucency,
as I mentioned, and opacity parameters.
They support Asian languages.
Although, that's part of the 708 specification,
there's, there's only, there's not any TVs I've
seen that can support that, but TVs do support
translucency and opacity and they supports different
fonts and colors.
So, that brings me to my next point,
and that is, in your professional broadcast
facility, you have scopes.
You have different types.
Different companies make different HD-SDI monitoring
that can do, decode closed captioning,
but they do not show these Advanced 708 features.
I have not seen one that does,
and when you air that on TV, the people,
the folks at home-- they do have that feature.
So, really, these professional scopes may be
lacking in the ability to verify these font styles and
colors and translucencies that you can use with the
Advanced 708.
There's also 16x9 positioning.
There's no high definition 16x9 so you have more
real estate to move the captions from left to right.
And you can language tag.
So, if you have multiple languages,
you can tag language with English,
French, Spanish, whatever you need and that's also
part of the Advanced 708 features.
You should be able to verify that tagging with software only,
and of course, you can have up to 63 services of
multiple languages, although because of bandwidth
constraints, we've only been able to add 6 languages to
708 before you know it, it's too much bandwidth to
be able to actually air it.
And, of course, there's Drop Shadows.
These are all Advanced 708 features that are,
you should be able to verify.
OK?
So, verifying data stored with captioning.
Let's talk about that.
There's other data besides captioning and some of that
data could be, XDS and V-chip.
For example, in 608, you can have,
you know, a particular rating or
program identification and how long the program is,
total running time.
All these things you can verify as well,
and that's important to verify without having to use
some sort of hardware solution,
doing with software only.
Active format description, different aspect ratios,
that's important to verify before you air it.
Time code track, OK, that's important if
you're going to transcode.
If you're going to transcode,
you've got to make sure that your captions match your
time code track and your video file.
The next item that I want to cover is
verifying formatting for Internet captioning.
So, this is one legislation that's called the CVAA.
I know, Communication Video Accessibility Act of 2010 by
the FCC and it calls for being able to convert
captions from TV to captioning on the Internet.
And this kit example on the graphic,
I'm showing YouTube video player,
and you can see YouTube video player,
you can actually upload an SCC.
You can upload and actual caption file for Television
and YouTube will actually do its best to
keep the formatting.
For example, on TV, there's something called
split captions where, when two people are talking you have
captions on the one end and on the other end,
just like on the graphic.
So, that is maintained.
So, it's important that you verify that formatting.
So, that you know your web team is keeping the captions
the same way they were aired.
Roll up captions should be kept that way.
Pop on captions should be kept that way.
Positioning on the screen so they don't block a lower
third, they should be kept that way as well.
And there's different caption file formats
for the Internet.
So, verification of the positioning is critical as well
when you're uploading captioning to the Internet.
All right, the next step is checking for
common CC errors.
So, we talked about drifting out of sync captioning.
That happens, usually when you're converting
the frame rate.
Many online captioning videos are 2398,
but it was aired in 29.97, or 60i,
or 59.94, 720p, but when you upload the original file
that was shot in film that's 23.98 and often times that's
not going to match.
So, it's important that you check that drift.
As I mentioned, check to make sure that the text
formatting is following the standards of accessibility.
There's times also when 708 is missing and
only 608 is found.
So, you have an HD MPEG 2.
And let's say you have, you know,
some software that says, "OK,
well give an SCC file and we'll transcode
into MPEG 2."
Well, what happens is, you, you transcode it,
you get the MPEG2.
You send it to your TV station and they fail it.
They say that, "We're just not seeing 708.
Only 608 is found."
So, it's important that you check your,
your MPEG 2 or whatever file you're sending,
whether its to Extreme Reach or to a broadcaster,
to PBS to make sure that both 608 and 708 are there
and that applies to High Definition specifically.
Its also important that the syntax of the caption file
is compatible with the transcode engine so,
you know, you may get an SCC file that was made,
you know, someone, I've seen someone make an SCC file by
hand, manually, and it just doesn't work.
And, so just because you have an SCC,
you have got to make sure that it's actually compliant
and it matches the right syntax,
otherwise you're going to get a lot of problems.
And also, of course, the same thing applies for the
syntax of the file for Internet captioning players.
So, you have to check these things.
Spelling and transcription mistakes,
of course, how can you check to say,
see if the actual captioning is right,
unless you actually have some sort of player to,
to read the captions and then spot check them before
you air them.
This is an interesting one.
Let's say you air something with captioning,
but then you are going to put that on the Internet or
you're going to put it on VOD and that's mean that you
have to edit that video, and,
but the captioning is not necessarily
matching the edit.
Once you edit the video, the captions no longer match and
so, just, it's important to make sure that an edited
video is checked against the captioning.
Otherwise, you're going to get like a video you're
playing back on the Internet,
or on VOD that, halfway through the video,
all of a sudden the captions are not matching at all.
So, it's important to check that once you
edit your video.
All right, next let's talk about emulating Television
captioning viewing experience.
So, what I mean by viewing experience,
I'm talking about the viewers at home.
You should have the ability to view your video the same
way that the people at home are going to see it
prior to airing it, and you have to make sure
you support Advanced 708, like the fonts.
You have to make
…