the Filmora reading seems to be in PAL frames
(count 25 frames for each second in timeline display)
00:15:03:10 in PAL frames 25 fps
when I switch units in Audacity to PAL frames
it reads 00:14:49:01
still no match up
and not much different that the
00:14:49:00 reading in NTSC drop frames (29.97? frames per second)
seems like the PAL reading should be more different
NTSC intentionally doesn’t come out even to hide the analog color signals on the TV screen. Nothing lines up so you see a fine grain pattern instead of stripes.
Why do you think the system is converting with Drop Frame? It doesn’t have to be. The error would be…ummmm…about a frame per minute, I think. Something like that.
A word on Drop Frame. It doesn’t drop frames. It drops time. Perfect, sequential video frames might be labeled 1, 2, 3, 4, 6, 7, 8… for example. Video editors know how to deal with the missing numbers.
Welcome to NTSC, where an hour show doesn’t fit in 60 minutes.
The fly in the ointment is that the original tape has been “dubbed” to a DVD
and either the tape has bad motor noise in the background
or the dubbing adds the motor noise to the DVD audio.
I want to split the VOB into audio and video
copy the audio
de-noise the audio
and then merge the audio and video back together again.
easy peasy right?
Audacity can edit the VOB
Filmora can detach the audio, delete the audio, add new audio files and export to DVD files
BUT
the fixed files are not in sync.
with some coaxing the files can be brought back in better sync
but its driving me nuts.
The tape appears to be NTSB, it plays well for there first hour
then tears up frequently in the last 40 minutes
so I do not think it is somehow a PAL tape.
But Filmora must think so because it has the frames at 25 per second, you can count them.
I just want to get the audio and video on the same timeline
so they play together
or I can fudge the audio to match the video.
the original tape has been “dubbed” to a DVD
and either the tape has bad motor noise in the background
or the dubbing adds the motor noise to the DVD audio.
So this isn’t a conversion or dub job. It’s disaster recovery.
You have no idea of the show duration or frame rate at the time of the dub. Home tape players don’t care.
If the player speed wanders, then Noise Reduction may fail. Noise Reduction makes you sample the noises at the start of the process (the Profile step) and then uses that sample through the show to know what to remove. That’s like letting the dog sniff the shirt to know what to do.
If the noise is moving, the profile falls out of alignment.
Maybe the other elves will have ideas.
Oh, one other point.
and then merge the audio and video back together again.
Audacity can open up the audio portion of many video files, but not put them back together. For that you need a video editor or manager.
Many errors like this may seem to exotic file shifts, but turn out to be simple speed changes. Effect > Change Speed can be as simple as telling the tool what the two show duration values are and it will figure out everything else.
It is an old commercial VHS tape
the LG DVD player calls playing a VHS tape and recording it to DVD at the same time as
Dubbing
that’s not what I would call it
So not quite disaster recovery but an attempt to conserve a diminishing resource.
I am using Filmora as the video editor, am open to suggestion.
tried remux but did not seem to do the trick.
Working with Filmora but most seem unfamiliar with the product except for what they can read on their screens
and I am so ignorant in this area, it has to be a challenge.
I think we can make the duration of the tracks come out right and maybe even help a bit with noise reduction, but reuniting the patched track to its video is not an Audacity job.
Audacity uses FFmpeg (if available) to import audio from VOB files. FFmpeg extracts the audio stream and passes it to Audacity.
Is there any guarantee that the audio stream starts at the same time as the video stream in a VOB? I don’t think there is.
Ok I have the video in Filmora
1 second by their clock, marker at every 5 frames
I can count 25 frames in the 1 second
So I am thinking
this was shot by bbc in PAL 25 frames per second
and “transformed” into NTSB at a speed of 30 frames per second for the yanks
the frames seem to be fixed so the other variable is time
the “transformation” played 25 frames in .83 seconds or 1.2x faster
and “shifted” the audio along and rolled up in the vob
when I open the vob the video is new time
and the audio is old time ???
maybe trial and error “snoop a little off here, add a little bit there”
may be the best way to approach
but would still like to figure out how to calculate how far to shift the audio in audacity
the audio files have different and mysterious starts
vob 1 0:32 seconds of silence ok intro
vob 2 2:43 minutes silence
vob 3 5:15 minutes silence
vob 4 7:43 minutes silence
all jobs are 1.07 Gb in size
maybe audacity can only see the “real” audio
and the silence is the time where the magic lip synch took place
“played” 1.2x faster to match the new (NTSB) rate
I just looked at Daffy and Bugs VHS to DVD
and Filmora said 25 frames per second
NO WAY
homemade VHS copy direct from the antennae in the U S of A 30 years ago
got to be 30 frames per second
Filmora edits on 25 fps scale
can output about whatever you want.
I am guessing if the lip/action is at 00:00:25:05:24 on the video,
the audio, as seen in audacity, using 25 fps scale
should be there too.
Current plan:
the speed of the audio seems to have been altered in production?
so that what is in synch at beginning of VOB_01
is out synch at beginning of VOB_2.
Just going to insert blip
by re-synching at the beginning of each VOB.