Sync Guide - How to sync

[jonjonjon]

tks to FRITOLAY, who made this unique post some time ago
That's his guide, not mine :

"Who Are The Syncing Masters Up In Here?

I'm writing this for anyone; not assume anything... As well as only covering the basics, but it is a complete picture.


The Big Idea
Videos come in containers. Some may be familiar already, such as mp4, mkv, avi and so on. These containers have different capabilities, and capabilities.

Inside each of these you will most likely find a video track (avc, mpeg, h.264...) plus one or more audio (aac, mp3, ac3...) or subtitle (srt) tracks. Rarely you will find extra things like cover images, chapter headings, or encoder tags. Though these are not as common as it reduces compatibility quickly.

So the idea is to take out (demux) the main audio track from the source material, do some editing magic, sync the riff and then add it back (mux) into the source material either replacing or adding to the main audio track. Easy ? Oh ya, wear headphones when your editing audio....


The Tools of This Trade
When it comes down to it, you only really need a few programs to accomplish this task.


First no matter what you are dealing with you will need Audacity.
https://sourceforge.net/projects/audacity/
Grab & install - Audacity 2.0.5 installer


And you will need to add 3 plugins.
LAME & FFmpeg will take care of importing just about any audio track into Audacity.
https://lame.buanzo.org/
Grab & install - Lame_v3.xx.x_for_Windows.exe & FFmpeg_v0.x.x_for_Audacity_on_Windows.exe


Chris’s Dynamic Compressor plugin for Audacity will be the editing magic I spoke of.
https://theaudacitytopodcast.com/chriss-...-audacity/
Grab & install - Chris’s Dynamic Compressor plugin for Audacity


As for adding the riff to the source material, this will require 1 of 3 programs; mp4box for mp4's, mkvtoolnix for mkv's, and avidemux for avi's. I will only be covering mkvtoolnix as I only produce mkv's for this type of thing. I find mkv's are just easier to work with; though you can jump between mkv and mp4 without recoding. They support similar formats so its a matter of demuxing the tracks from one and muxing it as another.


Mkvtoolnix
https://mkvtoolnix.download/downloads.html


MKVExtractGUI-2 - a handy gui for extracting tracks from mkv files
https://sourceforge.net/projects/mkvextractgui-2
Grab & extract into the mkvtoolnix folder 


The Good Stuff
Deciding on good source material is a purely subjective thing.

Myself, for example, aim for at least 720p, x264 video codec, and a less than 2GB final product. Bonus stuff are things like 1080p, subtitles, and 2ch audio.

From experience people will complain if its too big, and if it has multiple audio tracks because for whatever reason people still use windows media player. I still like to include the original audio track in the final product; one stone, two audio tracks. Just tell them to get VLC and its fixed.


The Nitty
Now that you have the tools, and your source material, its time prep the audio.

Start audacity, drag and drop your source material into the main window.




Once that's done, you will be presented with a nice waveform (the earthquake graph looking thing). The higher the spike, the louder something is. Now you have 2 channels (stereo), so 2 waveforms.




Our goal is to make your waveform look like this once done. To do this you will use the Compress dynamics plugin you added.




First find a small section of the waveform that has both loud parts with people talking, and quiet parts, again with talking. The reason for the talking is because that's what you want to compress to the same volume; its what people want to be able to hear clearly.




Now comes what I call the fudge, as its a little different for every file. Keeping your section highlighted goto the Effect menu and find Compress dynamics 1.2.6. A window will open; read the little howto blurb as its actually quite concise. Below are the settings I used for The Wizard of Oz. You will need to play around with the settings to find what works best with your source material. This is where you should spend the most time and care as this will decide whether it is a good or bad sync.




Once you find something that works, select the whole waveform and apply those settings. It may take some time depending on your hardware; that's why you tested it on a small section.




After all that time producing some nice even audio, export a copy of your work. You may want to replace the quiet source audio with your even audio track. As for the type of audio AAC probably is the more common, its what I use; the quality setting is up to you, higher number, bigger filesize.





The Break
Watch some of your source material, since you should be using vlc; play your even audio too. Switching between the two, pausing one and listening to the same part on the other. It should sound as if you just turned up the volume without actually doing so. Compared to the source, there still should be enough difference between voices and background noise's to be pleasant and not drowned out.

If your not satisfied go back and rework your compression settings. Otherwise take a break. I'm serious give your ear a minute to relax. In the same spirit you would rinse your mouth between tasting different things. Come back in 30mins and finish the job.


The Gritty
Now that your source material is ready lets move on to syncing the Rifftrax MP3 (which I hope you payed for and downloaded already). Sometimes there is enough space to cut the intro and paste it so it plays during the credits. Like I did with Harry Potter, but that's not the case here.

Drag and drop the Rifftrax mp3 into Audacity. You will now have 2 tracks.




Form a selection of the intro on the riff track (get it now?), all the up to where Mike says "1, 2, 3, Pause." (you can see the 123 pause on the right side). Click the Zoom button a few times to see some details if needs be. When your cursor is close to the selection borders you get a finger, it will let you expand the selection in that direction. Once selected, click File->Export Selection, save it with the same settings as before. Make sure to either mute the main track or click solo on the riff track. Once done hit Delete to remove the selection.




Open the text file that came with the riff as it has the Disembaudio timings you need. Times on the left are the ones that we care about.

Code:

-= DISEMBAUDIO SYNCH LINES =-

DVD | Line
02:08    Did she hurt you?

03:07    Take it easy

05:15    You always get yourself into a fret over nothing

14:58    There’s a storm blowing up, a whopper

24:40    We thank you very sweetly

34:41    Are you doing that on purpose?

45:49    He must

55:27    Here, give us your hands and we’ll pull you along

1:06:26    How bout a hippopotamus

1:16:48    But give me back Toto

1:27:29    Can I believe my eyes

Back in Audacity key in the first timing, 00h 02m 08.000s in this case. Zoom in a bit, hit the space bar to play from that time point. Should hear the line from the text, "Did she hurt you?", fairly quickly. If not you may have gotten a PAL source, or is an extended cut and the riff's usually are the theatrical cuts.




Next to get a rough idea of how much time we need to add at the front of the riff to sync it form a selection from the beginning of "Did she hurt you?" to the beginning of Disembaudio's line. Note the selection length, 10.722s.




Move your selection cursor to the very begining of the riff track. Click Generate->Silence, round your time 10.7 here.




Go back to 00h 02m 08.000s, take a listen. How close are you ? Zoom in and see. This was the first adjustment I had to make, about 00.017s more silence (I round to 00.015s). So add or remove silence at the beginning to get it right.




Once it sounds right, that they are in sync click the Silence Audio button to mute Disembaudio. Try to include any "Tones" you see too.





The Shampoo
At this point its just a matter of rinse and repeat muting all the Disembaudio timings.

If you source was a good match you really shouldn't need to find a "middle ground" between the first timing and the last one. I had to remove an extra 00.060s, just a little fudge. Sometimes without having or renting the actual DVD you just cant get a good match online (some of the star trek ones stink).




One issue in this case is that the Rifftrax outro is making the riff track longer than the main audio. Just remove it; its only more guitar and the copyright.





The Quack
Now you have even audio, a synced Rifftrax, and a muted Disembaudio. Yet something is missing, yes the audio ducking. This is where the main audio will lower in volume while the guys talk.

To do this first select the top track by clicking in the empty space to the left of the second waveform. Once selected use the Effect->Auto Duck tool. You can test different timings yourself, but this is what I used for this one. Again time and care matter here, the trick is if you can listen to just the audio and understand both the riff and main audio without closing your eyes to focus on it.




That's it for the audio, File->Export same as before. I usually export a copy of just the riff track as well seeing as all the work is done. Maybe I messed up ?


The Big Mux
Only thing left to do is bring all this together into one lovely product.

Run mmg.exe from the mkvtoolnix folder. Drag your source material into this program.




My source has 2 audio tracks, one is 5.1ch the other 2ch. I don't care about the 5.1ch, and I want to use my even audio so I unchecked both. Make sure you set the Display width/height for the video track. I also added an srt file. Order should be video, audio 1, audio 2, subs.




Everything in order ? Set the output filename and click Start muxing.





The End
Well you made it, hope this helped. I'm f*****g tired. "

[TheDarkPrince2018]

tks to FRITOLAY, who made this unique post some time ago
That's his guide, not mine :

"Who Are The Syncing Masters Up In Here?

I'm writing this for anyone; not assume anything... As well as only covering the basics, but it is a complete picture.


The Big Idea
Videos come in containers. Some may be familiar already, such as mp4, mkv, avi and so on. These containers have different capabilities, and capabilities.

Inside each of these you will most likely find a video track (avc, mpeg, h.264...) plus one or more audio (aac, mp3, ac3...) or subtitle (srt) tracks. Rarely you will find extra things like cover images, chapter headings, or encoder tags. Though these are not as common as it reduces compatibility quickly.

So the idea is to take out (demux) the main audio track from the source material, do some editing magic, sync the riff and then add it back (mux) into the source material either replacing or adding to the main audio track. Easy ? Oh ya, wear headphones when your editing audio....


The Tools of This Trade
When it comes down to it, you only really need a few programs to accomplish this task.


First no matter what you are dealing with you will need Audacity.
https://sourceforge.net/projects/audacity/
Grab & install - Audacity 2.0.5 installer


And you will need to add 3 plugins.
LAME & FFmpeg will take care of importing just about any audio track into Audacity.
https://lame.buanzo.org/
Grab & install - Lame_v3.xx.x_for_Windows.exe & FFmpeg_v0.x.x_for_Audacity_on_Windows.exe


Chris’s Dynamic Compressor plugin for Audacity will be the editing magic I spoke of.
https://theaudacitytopodcast.com/chriss-...-audacity/
Grab & install - Chris’s Dynamic Compressor plugin for Audacity


As for adding the riff to the source material, this will require 1 of 3 programs; mp4box for mp4's, mkvtoolnix for mkv's, and avidemux for avi's. I will only be covering mkvtoolnix as I only produce mkv's for this type of thing. I find mkv's are just easier to work with; though you can jump between mkv and mp4 without recoding. They support similar formats so its a matter of demuxing the tracks from one and muxing it as another.


Mkvtoolnix
https://mkvtoolnix.download/downloads.html


MKVExtractGUI-2 - a handy gui for extracting tracks from mkv files
https://sourceforge.net/projects/mkvextractgui-2
Grab & extract into the mkvtoolnix folder 


The Good Stuff
Deciding on good source material is a purely subjective thing.

Myself, for example, aim for at least 720p, x264 video codec, and a less than 2GB final product. Bonus stuff are things like 1080p, subtitles, and 2ch audio.

From experience people will complain if its too big, and if it has multiple audio tracks because for whatever reason people still use windows media player. I still like to include the original audio track in the final product; one stone, two audio tracks. Just tell them to get VLC and its fixed.


The Nitty
Now that you have the tools, and your source material, its time prep the audio.

Start audacity, drag and drop your source material into the main window.




Once that's done, you will be presented with a nice waveform (the earthquake graph looking thing). The higher the spike, the louder something is. Now you have 2 channels (stereo), so 2 waveforms.




Our goal is to make your waveform look like this once done. To do this you will use the Compress dynamics plugin you added.




First find a small section of the waveform that has both loud parts with people talking, and quiet parts, again with talking. The reason for the talking is because that's what you want to compress to the same volume; its what people want to be able to hear clearly.




Now comes what I call the fudge, as its a little different for every file. Keeping your section highlighted goto the Effect menu and find Compress dynamics 1.2.6. A window will open; read the little howto blurb as its actually quite concise. Below are the settings I used for The Wizard of Oz. You will need to play around with the settings to find what works best with your source material. This is where you should spend the most time and care as this will decide whether it is a good or bad sync.




Once you find something that works, select the whole waveform and apply those settings. It may take some time depending on your hardware; that's why you tested it on a small section.




After all that time producing some nice even audio, export a copy of your work. You may want to replace the quiet source audio with your even audio track. As for the type of audio AAC probably is the more common, its what I use; the quality setting is up to you, higher number, bigger filesize.





The Break
Watch some of your source material, since you should be using vlc; play your even audio too. Switching between the two, pausing one and listening to the same part on the other. It should sound as if you just turned up the volume without actually doing so. Compared to the source, there still should be enough difference between voices and background noise's to be pleasant and not drowned out.

If your not satisfied go back and rework your compression settings. Otherwise take a break. I'm serious give your ear a minute to relax. In the same spirit you would rinse your mouth between tasting different things. Come back in 30mins and finish the job.


The Gritty
Now that your source material is ready lets move on to syncing the Rifftrax MP3 (which I hope you payed for and downloaded already). Sometimes there is enough space to cut the intro and paste it so it plays during the credits. Like I did with Harry Potter, but that's not the case here.

Drag and drop the Rifftrax mp3 into Audacity. You will now have 2 tracks.




Form a selection of the intro on the riff track (get it now?), all the up to where Mike says "1, 2, 3, Pause." (you can see the 123 pause on the right side). Click the Zoom button a few times to see some details if needs be. When your cursor is close to the selection borders you get a finger, it will let you expand the selection in that direction. Once selected, click File->Export Selection, save it with the same settings as before. Make sure to either mute the main track or click solo on the riff track. Once done hit Delete to remove the selection.




Open the text file that came with the riff as it has the Disembaudio timings you need. Times on the left are the ones that we care about.

Code:

-= DISEMBAUDIO SYNCH LINES =-

DVD | Line
02:08    Did she hurt you?

03:07    Take it easy

05:15    You always get yourself into a fret over nothing

14:58    There’s a storm blowing up, a whopper

24:40    We thank you very sweetly

34:41    Are you doing that on purpose?

45:49    He must

55:27    Here, give us your hands and we’ll pull you along

1:06:26    How bout a hippopotamus

1:16:48    But give me back Toto

1:27:29    Can I believe my eyes

Back in Audacity key in the first timing, 00h 02m 08.000s in this case. Zoom in a bit, hit the space bar to play from that time point. Should hear the line from the text, "Did she hurt you?", fairly quickly. If not you may have gotten a PAL source, or is an extended cut and the riff's usually are the theatrical cuts.




Next to get a rough idea of how much time we need to add at the front of the riff to sync it form a selection from the beginning of "Did she hurt you?" to the beginning of Disembaudio's line. Note the selection length, 10.722s.




Move your selection cursor to the very begining of the riff track. Click Generate->Silence, round your time 10.7 here.




Go back to 00h 02m 08.000s, take a listen. How close are you ? Zoom in and see. This was the first adjustment I had to make, about 00.017s more silence (I round to 00.015s). So add or remove silence at the beginning to get it right.




Once it sounds right, that they are in sync click the Silence Audio button to mute Disembaudio. Try to include any "Tones" you see too.





The Shampoo
At this point its just a matter of rinse and repeat muting all the Disembaudio timings.

If you source was a good match you really shouldn't need to find a "middle ground" between the first timing and the last one. I had to remove an extra 00.060s, just a little fudge. Sometimes without having or renting the actual DVD you just cant get a good match online (some of the star trek ones stink).




One issue in this case is that the Rifftrax outro is making the riff track longer than the main audio. Just remove it; its only more guitar and the copyright.





The Quack
Now you have even audio, a synced Rifftrax, and a muted Disembaudio. Yet something is missing, yes the audio ducking. This is where the main audio will lower in volume while the guys talk.

To do this first select the top track by clicking in the empty space to the left of the second waveform. Once selected use the Effect->Auto Duck tool. You can test different timings yourself, but this is what I used for this one. Again time and care matter here, the trick is if you can listen to just the audio and understand both the riff and main audio without closing your eyes to focus on it.




That's it for the audio, File->Export same as before. I usually export a copy of just the riff track as well seeing as all the work is done. Maybe I messed up ?


The Big Mux
Only thing left to do is bring all this together into one lovely product.

Run mmg.exe from the mkvtoolnix folder. Drag your source material into this program.




My source has 2 audio tracks, one is 5.1ch the other 2ch. I don't care about the 5.1ch, and I want to use my even audio so I unchecked both. Make sure you set the Display width/height for the video track. I also added an srt file. Order should be video, audio 1, audio 2, subs.




Everything in order ? Set the output filename and click Start muxing.





The End
Well you made it, hope this helped. I'm f*****g tired. "

it's all Greek to me lol ... I have tried this and it never turns out right for me lol

[jonjonjon]

here's an old guide i did a while ago. note, i use Audacity 2.1.0 since the newer versions removed/changed some effects i use. here are the 2.1.0 links for windows:
portableapps audacity: https://sourceforge.net/projects/portabl...0Portable/
direct 2.1.0 download: https://sourceforge.net/projects/portabl...e/download

***

The key to my volume balancing is:
* movie compression with Chris’s Dynamic Compressor 1.2.6 plugin
* rifftrax compression with Level Speech plugin
* movie auto-duck -18db
* rifftrax amp ~ -12.3db (I tweak this depending on the riff and movie, -12.5, -12.20, etc)

Chris’s Dynamic Compressor 1.2.6 plugin for Audacity:
>> appears in Audacity Effect menu as "Compress dynamics 1.2.6..."
http://manual.audacityteam.org/man/chris...essor.html
https://theaudacitytopodcast.com/chriss-...-audacity/
http://forum.audacityteam.org/viewtopic.php?p=149994
DOWNLOAD: https://theaudacitytopodcast.com/apps/ch...-1_2_6.zip

Level Speech plugin for Audacity:
http://forum.audacityteam.org/viewtopic....46&t=91671
DOWNLOAD: http://forum.audacityteam.org/download/file.php?id=9658

bsg4you detailed synch steps: (sorry, it's kind of rough)
* riff: mute opening song
* riff: copy intro comments to end
* riff: delete tagline and legal statement at end, create 1sec gap between mkb farewell and intro comments
* riff: mute intro comments
* riff: use text file to approximate synch with first dis line, delete start of riff to match
* riff: check first dis to fine tune synch, delete start of riff to match >> sometimes first dis line isn't in time with movie or dis says it at the wrong pace
* riff: duplicate riff track and mute as a backup and for dis line qa
* riff: check synch of each dis line, mute, then check again against muted riff backup, ALSO note any curse beep times
* riff: 2nd pass, check for and remove stray mumbling and chuckles, also mute studio pops/ticks
* all: reset all track volumes to 0.0 (i adjust these initially when doing the dis synch stuff)
* riff export: 155 quality (~96kbps) aac just synch no dis line riff mono backup
* delete backup riff at bottom of audacity window
* movie: noclip.amp to 0.0
* movie: chris compress dynamics 1.2.6: comp.ratio+0.600(custom) comp.hard+0.500(default) floor-20.0(custom) noise.gate+0.00(default) max.amp+1.00(custom)
* TOURBLESHOOTING: if selecting the whole movie and doing chris results in nothing happening, select from 0.100s to end and apply chris, it should then work
* movie: hard limiter: db.limit-1.0(custom) wet+1.0(default) residue+0.0(default)
* riff: level.speech-95 << selected whole riff, selected level speech, got an error pop up, dismissed it, got the options, set 95 and hit apply << no pop up the 2nd time
* movie: auto-duck: duck-18.0(custom) defaults: 1 0.5 0.5 0 0 -30
* riff: amp new peak = -12.3
* riff: check if intro sounds too loud at end (it depends if the movie is loud or quiet during intro); also lower dis -1.5 or so depending on how it sounds
* riff: amp bit of level countdown at end of intro comments to riff.max (sometimes mike's voice gets really quiet during his "3, 2, 1, pause")
* amp.net(-9) on credits music (chris can make a loud credits music sequence super loud, this fixes it)
* curse beeps: do amp -9 to -15 (i don't like loud beeps)
* export of 3 channels (2 movie and 1 rifftrax) to 2: WAVE and use external audio encoder -or- do audacity aac export at 155 quality
* add subs, mux
* add named chapters
* mediainfo and gmkvextractgui checks
* qa listen for good volume balance, visually check audacity for curse beeps and missed synch lines in waveform while listening

***

I haven't seen that
The precise values are much of use

Tks, bsg !