Use Auto GK or DGIndex to Demux to MP3 or WAV

[loninappleton]

Trying to read old threads it just got too messy and so from an older how- to repair an avi thread I have process questions and will possibly make a guide.

The goal, as before, is to get audio from the original dvd disk as an AVI using AGK and the preparation described by HH.

The question is: Use Auto GK which contains DGIndex or DGIndex stand alone to normalize (that is bring up low volume and decrease loud volume to that which is more level) with Goldwave which uses WAV files.

My use of AGK shows that it is normalizing audio to the AVI encode from it's screen prompts showing process of the job. But playback has not given an adequate level audio. This is why my continuing questions on the same subject. Also a previous post said that AGK automatically raises the volume. I would not be doing this if that were the case.

Have I missed a setting in setup of AGK?

Goldwave has given me success in taking the MP3 then converting it to .WAV for processing (normalizing.) It then offers the option to make an mp3 from the completed coding job.

At that point the new mp3 has to be remuxed replacing the original audio.

Nevermind that for now.

Do I use AGK or DGIndex in native mode to get the mp3 out? What are the steps?

I have read the DGIndex manual as much as I can understand. DGIndex has the option to make a WAV file in it but I have not been able to understand the instruction.

What I have from DGIndex is a file called VTS 01 1.d2v.
After the Save Project operation I'm lost. DGIndex will not show me how to make the WAV file for Goldwave.

[manono]

Your post is all over the place so I don't know if I'll be answering what you want to know. Neither AutoGK nor DGIndex can demux MP3 audio as such. AutoGK demuxes it if it's working on an AVI with it. DGIndex won't and can't. When I want to extract MP3 audio from an AVI for normalizing, I use VDubMod followed by MP3Gain for the normalizing without reencoding. You don't need a WAV Editor such as Goldwave or Audacity just to adjust the volume in an MP3. Alternatively, if you're using AutoGK to reencode an AVI and you don't like the resulting volume level, get the MP3 from the temporary folder, run it through MP3Gain, and then mux it into the AVI .

DGIndex is for indexing DVDs for use in AviSynth scripts. It will, by default, demux whatever audio is in the DVD (and MP3 isn't one of them) and can be set up to decode one of the audio tracks to WAV audio. However, in my experience it gives you a WAV file at lower volume than in the DVD so I demux the audio (AC3 usually) and then convert it to WAV audio in BeSweet, although there are other ways.

[loninappleton]

manono thanks for answering.

But what you say about wave editors simply contradicts what was said previously (I didn't want to make that other thread longer for pages and pages.) I've used Goldwave effectively to do the normalizing routine.

What I'd like to do is simpify the routine and make it understandable. If these tools mentioned should be bypassed then can we procede.

Demux audio from the TS folder made by Auto Gordian Knot.

Decode the audio track to WAV audio.

I can start over completely from the original DVD if that will make things more clear.

BeSweet is another new program name to me.

[manono]

But what you say about wave editors simply contradicts what was said previously ... I've used Goldwave effectively to do the normalizing routine.

Of course it contradicts what was said. Because it's not necessary to use a WAV editor just for normalizing MP3 audio and degrades the audio by converting it first to WAV audio before normalizing it and then reencoding it for MP3. It's slower too. I open the MP3 in MP3Gain, run the track analysis and then track gain to 92 (or whatever volume you like). Of course Goldwave can do the job, and if you have other work you want to do on the audio then you'll want to use it. But for simply adjusting the volume it's not necessary.

If you're unhappy with the volume of the audio created by AutoGK, then take the MP3 AutoGK created for you, run it through MP3Gain, and mux it into the AVI, replacing the audio already there. I should also say I've not found that AutoGK gives you audio at too low of a volume.

Or, if you prefer, decode the AC3 from the DVD to WAV audio, work on it in Goldwave, convert it to MP3 and then stick it in the AVI.

[loninappleton]

The thread is getting long again but I suppose it's unavoidable. MP3Gain makes everything louder. Normalization and compression often become confused in these descriptions. When I speak of normalization-- Goldwave has an option to do this-- I mean bringing up lows and bringing down big booms. In this piece the big booms come from tympany style drum intros to some scenes, plus some celebration dances. The audio on the original dvd by loading the TS folder into AGK to inspect is:

Open Auto Gordian Knot
Load DVD
Find TS Folder on DVD drive
Load the TS folder into AGK

in the Audio box, find

Audio
Audio Stream 0 AC3

For making a guide do you not agree this is a good place to start?

Alternatively, if starting with an AVI previously made, GSpot will reveal the audio format
for that AVI.


An aside which may be interesting to others is a thread I found from Baldrick on using Goldwave to process an AVI file directly to WAV.

It is here and included for those with interest:

https://forum.videohelp.com/threads/106981-How-to-convert-AVI%28DivX-XviD-AC3%29-to-WAV...using-GoldWave

(I actually found this at the Goldwave site. It includes screen shots.)


back on point....


For varifiaction, I loaded my TS folder made from AGK into AGK. Here the Audio shows

Audio 0 PCM for Pulse Code Modulation.

This is another audio file format.

The question for this post since we are beginning a fresh analysis is:

Why are the two audio formats different- one from DVD the other from AGK preparation?
I followed the HH procedure but cannot copy it in here unless perhaps a separate post.

For completeness, I could bring that in here in a separate post.

Second question is, which source-- DVD TS folder or prepared AGK TS folder is better to work on?

A program called BeSweet was mentioned earlier and even in that Goldwave How-To from Baldrick was mentioned back then in 2003.

After this analysis which is the best way to proceed?

[jagabo]

When I speak of normalization-- Goldwave has an option to do this-- I mean bringing up lows and bringing down big booms.

That's dynamic range compression, not normalization.

http://en.wikipedia.org/wiki/Dynamic_range_compression
http://en.wikipedia.org/wiki/Audio_normalization

[loninappleton]

When I speak of normalization-- Goldwave has an option to do this-- I mean bringing up lows and bringing down big booms.

That's dynamic range compression, not normalization.

http://en.wikipedia.org/wiki/Dynamic_range_compression
http://en.wikipedia.org/wiki/Audio_normalization

This is a helpful distinction. And here is something at Hydrogen Audio describing how to do it in Goldwave:



I dislike posting links and again for completelness here it is, briefly in text:

QUOTE
but I don't want to spend time experimenting -- I just want settings that will give reasonable, 'average' results.)
Most audio editing requires some listening and some trial-and-error! Once you find settings that work for you and your program material, you might be satisfied using the same settings for all of your files.

I don't use Audacity regularly, but I can give you some settings from GoldWave.

GoldWave's compressor doesn't have a gain control (maybe your "peak/RMS setting"), so with GoldWave, it's a 3-step process like this:

1. Normalize (set the peaks to 0dB... Goldwave calls it "maximize")
2. Use the compressor to "reduce loud parts"
3. Normalize again to bring-up the average level.

Here are a couple of presets from GoldWave:

Reduce Peaks:
Multiplier -6dB (your "ratio")
Threshold -6dB
Attack 0.001 seconds
Release 0.001 seconds

Reduce Loud Parts:
Multiplier -6dB (your "ratio")
Threshold -15dB
Attack 0.1 seconds
Release 0.1 seconds

-- Hydrogen Audio thread attribution:
http://www.hydrogenaudio.org/forums/index.php?showtopic=70743


Progress.

Goldwave works on WAV files.

Can we now return to getting a WAV file from either/ or TS folder on DVD or made from AutoGK?

[manono]

Can we now return to getting a WAV file from either/ or TS folder on DVD or made from AutoGK?

It won't be from AutoGK as it only has the choices of the original audio or a couple of kinds of MP3. I usually get the AC3 from the DVD when creating the D2V file (although you can use PGCDemux or a few other ways), and then convert it to WAV in BeSweet. Afterwards I load it into Audacity to do whatever I want with it.

I'm sure others have their own methods.

[jagabo]

Audacity has similar normalization and DRC filters. And with the ffmpeg source plugin it can import audio directly from AC3 and VOB files (and many other types). It won't automatically append the audio from several VOB files though.

[nic2k4]

Even after reading the whole thread I still can't figure out what's your objective. As Manono said, I have never found AGK to produce files with the volume too low; is your issue with some specific DVD's?

One issue I have noticed at times is with 5.1 downmix to stereo which produces an effect that sounds like the dynamic range is too large. When that happens I use the DVD stereo soundtrack, the studio sound engineers are better than me at downmixing.

FYI, Besweet can downmix AC3 surround audio to stereo MP3 as well as adjust the DRC.

[loninappleton]

Can we now return to getting a WAV file from either/ or TS folder on DVD or made from AutoGK?

It won't be from AutoGK as it only has the choices of the original audio or a couple of kinds of MP3. I usually get the AC3 from the DVD when creating the D2V file (although you can use PGCDemux or a few other ways), and then convert it to WAV in BeSweet. Afterwards I load it into Audacity to do whatever I want with it.

I'm sure others have their own methods.

D2V files are in DGIndex. I'm not adept at using that. The one experiment I did
I only highlighted the first VOB. And yes I had seen the manual but really need some good screen shots.

Here is a VOB extraction process using DGIndex:

https://forum.videohelp.com/threads/278086-Edit-a-VOB-without-Intermediate-Files

Does this look like the right path forward?

To this point we have used discovery tools (those which identify file types of audio.)

Next after discovering AC3 audio how will DGIndex extract its D2V file?

[manono]

Does this look like the right path forward?

No. He had the whole thing in a single VOB to begin with so he either decrypted the whole movie to a single VOB or he used VobMerge on them afterwards. Just load all the individual VOBs into DGIndex. Either highlight the first one and use Shift/Down Arrow to 'corral' them all, or add them one at a time until you get them all. Plus, that guide is over six years old. You use Force Film only if it's NTSC and a very high percentage of film (100% is best). Otherwise use 'Honor Pulldown Flags'. Either way you'll get a D2V file and the audio (Audio->Demux All Tracks). I guess the audio is all you're interested in. As nic2k4 suggested, if you're going to have AutoGK make an MP3 for you, you'll want to use the stereo track, rather than the DD5.1.

Next after discovering AC3 audio how will DGIndex extract its D2V file?

File->Save Project.

[loninappleton]

From the videohelp download page, here is a guide from Afterdawn on the topic of DGIndex

I do not see a date on it:

http://www.afterdawn.com/guides/archive/using_dgindex.cfm

Is there any erroneous information on this one?

Can you give the steps before File > Save project to include the pull downs and soforth?

In DGIdex there is a menu option to save to WAV which is our goal right now. How is that retreived after saving? Once the WAV is made the earlier quote on using Goldwave should help complete the process of Compression.

I looked at the Afterdawn guide earlier but some direct steps describing how to execute would help.

[manono]

I thought you were only interested in the audio? Whatever you do will give you the audio by default, unless you fool with the default setting. The best guide to DGIndex is the one included in the DGMPEGDec package. Also included are the excellent DGDecode Quickstart and DGDecode Manual if you have any interest in using the D2V project files in AviSynth scripts.

In DGIndex there is a menu option to save to WAV which is our goal right now.

I already told you it lowers the volume. However, if you want DGIndex to give you a WAV file, go Audio->Output Method->Decode AC3 Track to WAV. If there's more than one AC3 track, you have to choose one. Once you choose to decode to WAV audio, you can then also choose to apply Dynamic Range Control. This is also clearly explained in the DGIndex Manual.

[loninappleton]

I will practice with it and report any difficulties (which I expect.) I will use the TS folder direct from the DVD and see if I can get a D2V of the content.

[poisondeathray]

Add the vob's to the list, as in that guide (see the screenshot in that guide)

You don't need pulldown mode or anything since you're only dealing with audio . You're only using DGIndex to join the vob's and demux the audio

Audio=> output method should be set to "demux all tracks" by default

File=>save project will demux the original audio . You don't need the .d2v since you're only dealing with audio. The .d2v is for avisynth video frameserving

If goldwave cannot use the original audio, and you dont' like audacity for some reason, you can still use audacity to convert to uncompressed PCM wave , then use that as input into goldwave

Audacity requires the ffmpeg plugin to open different types of audio, follow these steps
http://manual.audacityteam.org/man/FAQ:Installation_and_Plug-Ins#How_do_I_download_and...ort_Library.3F

[loninappleton]

I have read the above. Tried a sample run and got some errors using the TS folder on the DVD which I think included titles not part of this job. Next I loaded the TS folder made from the DVD according to the HH instruction formula which I will put in a separate post.

To avoid the errors I'm going to see if I can get a response in real time here.

I've loaded the the Video TS files one at a time and pressed ADD.

Next should be to OK the set.

What is next?



Will keep that machine as- is until a response comes through

[manono]

...using the TS folder on the DVD

You don't load the entire VIDEO_TS folder. You load the VOBs that are part of the movie. These would be:

VTS_01_1.VOB
VTS_01_2.VOB
VTS_01_3.VOB
VTS_01_4.VOB
VTS_01_5.VOB
etc.

The VTS_01_0.VOB is not a movie VOB. And sometimes that first '1' is a '2' or more. It's the VTS with the movie. Play them if you can't tell.

[loninappleton]

Ok. This is helpful. My latest practice contains:


C: Documents & Settings....\Desktop\VIDEO_TS\VTS_01_1.VOB
thru
\VIDEO_TS\VTS_01_7.VOB


and clicked OK

Back at the main screen.

That shows nothing. I'll continue to

Audio > Output Method > Demux All TRacks


Did that.
Nothing happens. I'm not trying to be thick. Nothing happens, there's nothing to save. What's missing?

Is there a save step in between here that I missed? This is the reason I wanted to do this in real time.

[poisondeathray]

Go back and re-read the post, follow the instructions step by step

Audio=> output method should be set to "demux all tracks" by default

File=>save project will demux the original audio

In other words, nothing will happen until you do file=>save project

[manono]

I'm not trying to be thick.

So you just come by it naturally? You've already been told twice how to do it. pdr showed you his. Here's mine:

Next after discovering AC3 audio how will DGIndex extract its D2V file?

File->Save Project.

[loninappleton]

No need to pile on.


I did do a Save eventually and the process index took a better amount of time
to complete. The runtime of the audio is 2 hrs 26 min 44 sec. I didn't compare that with the original at this time but it is very close.

The progress showed about 1 mb per second so 2 and half minutes to make the d2v file. The file size looks too small though:

Mymoviefile.d2v is 1.8 mb



But I took what I had just to keep going. I renamed it from TS Video whatevs to Mymoviefile and reloaded that file name to do the convert to WAV.

Though the process could not be detected, I checked the file name and a .wav file is added to the directory.

Mymoviefile.wav is 1.57 Gb

One file really small and the other really big.

Sample playback of the .wav was pretty low as manono indicated.

What is the alternative? The option still remains of doing the Goldwave routine given earlier.

But here I promised to write in the routine for 'preparing' the TS folder from a dvd using DVD Shrink by HH at videohelp. That will be in the next post for those still reading this.

[loninappleton]

HH gave a good way to prepare a DVD in DVD Shrink and here it is quoted and unchanged:


" Many conversion programs will let you open a full set of ripped DVD files and select what you want to encode (just the movie or an individual episode etc) but many require you to prepare the DVD properly for encoding first. AutoGK and MeGUI both fit into the latter category. Here's how I do it:

Open the DVD (or the ripped DVD files on you hard drive) using DVDShrink.
Select the Re-author button.
If it's a movie DVD you'll see a large title in the right pane (there might be several titles but generally only one will be long enough to be the movie).
Drag the movie title from the right pane to the left.
Select the title in the left pane, then select the Compression tab at the top of the right pane. You'll see a list of audio streams and subtitles. De-select the ones you don't want.
In DVDShrink's preferences set the output DVD size to something large (I set it to 50,000MB) so DVDShrink won't try to "shrink" the video (you want it to just copy the video). You only need to do that once.
Use the backup function to "backup" the DVD to a new location.
When it's done, the newly saved set of DVD files should be just the movie (or you can backup individual episodes in a similar manner).

It's fine to let DVDShrink create Video and Audio TS folders when backing up, but it's not necessary. It makes no difference. They're only needed for burning a compliant DVD video disc. Just "backup" each ripped DVD to it's own location (folder) and give it an appropriate name.

The above is how I rip DVDs, although to do it in a single step (rip and re-author the DVD as a single movie in one step). I open the disc directly with DVDShrink. If it can't open the disc due to copy protection I run AnyDVD in the background to do the decrypting. If you can only rip the whole disc though (menus and extras etc included) then I find using DVDShrink to open the ripped files and re-author them as I described above the easiest way to prepare the DVD properly for encoding.

The ripped/re-authored set of vob files will contain the selected subtitles even though you can't see them in the preview while DVDShrink is running the backup.

Sometimes when you open ripped files with DVDShrink it'll display audio and subtitle streams which have a size of 0MB. There's no point selecting those as they're listed in the DVD menu but they aren't actually there (chances are they weren't included when ripping the disc).

** If you open the correct IFO file using AutoGK (after re-authoring the DVD) it'll display any audio streams and subtitles it finds (I'm not sure it displays the subtitles if you open the first vob file directly though).

AutoGK always encodes the subtitles you select after opening the IFO file as part of the video (they're encoded "into" the video). If you don't want them encoded, make sure none are selected. If you select more than one subtitle stream it might save the second subtitles as an individual file but it's been so long since I've used more than one set of subtitles I can't remember. Normally you'd just select the single subtitles you want (if there's more than one) and AutoGK will encode them into the video. If you want the subtitles as a separate file you'd need to use a different program to extract them from the vob files (and maybe convert them to a different format), but that's a whole other story......

Episodic DVDs can be re-authored using the same method as above (drag each episode title from the right pane to the left), but instead of a single set of vob files for the movie you'll have a set of vob files for each episode. Each will be sequentially numbered and have a corresponding IFO file AutoGK can open.

A set of vob files re-authored to contain just the movie should like this (although there could be more or less actual vob files depending on the movie size/length). You'd open the second IFO file with AutoGK (VTS_01_0.IFO)"

[manono]

If you want the subtitles as a separate file you'd need to use a different program to extract them from the vob files

No. You just go into the Advanced Settings of AutoGK and check 'Use external subtitles'.

[loninappleton]

I am re-reading the thread to get the fine points. So far DGIndex has made a d2v file and a WAV which should both be audio.

I copied out by hand the "information box" which has the exact fps and such.

The WAV is the only audio file available in the work so far that I have the software (Goldwave and possibly Audacity after setting that up) which can do Dynamic Range Compression (DRC). It is dynmamic range compression that brings up the low volume and brings down the loud parts, and not normalization as I now understand it. Above and printed out is the Goldwave routine to accomplish compression on a WAV file. GW can then convert to MP3 which will be the new audio remuxed by-- this time -- Virtual Dub Mod - to replace the original audio with the new.

Tomorrow or so I am going to do the DRC routines described for Goldwave.

For the group:

I am nlooking at the "Information Box" report which DGIndex gives How is that retrieved? As precaution, I copied it out by hand. Lengths of files can be critical in doing all this.

The time shown on the DGI info box ias 2:26:44

If I load the prepared AVI from Auto Gordian Knot into GSpot I get.... see next post. Flipping around I do not want to lose what I have written.

[loninappleton]

.... from previous

My double check of things shows that GSpot reports 2:26:45 as total time of the AVI
and 29.970 Fps

The frame rate looks very close at 29.970 vs 29.97003

But will a second off be noticeable across the run of the audio? Or, as I think the guide said for DGindex does it do a fill routine?

[manono]

The WAV is the only audio file available in the work so far that I have the software (Goldwave and possibly Audacity after setting that up) which can do Dynamic Range Compression (DRC).

As I mentioned earlier, DGIndex can also perform a DRC as it's getting the WAV file.

[loninappleton]

Well at least we are using the same acronyms now. Progress.
I need more practice with doing this and so will be doing the whole operation from scratch until it becomes familiar. I've seen where DGindex can make the WAV after file > save.

I see where it says "normalize" in the same menu as Audio> Output Method. Are we talking about the same thing yet? Or is there script writing involved? I have seen the manual in DGIndex. How do I then perform a DRC as it is getting the WAV file?

And, if wave form editing is not advised and there is lower audio as you said, how is this a better technique for a WAV?

[loninappleton]

Found it:

"Once you choose to decode to WAV audio, you can then also choose to apply Dynamic Range Control. This is also clearly explained in the DGIndex Manual."-- manono

I'll find that manual page.

[manono]

And, if wave form editing is not advised and there is lower audio as you said...

And as you discovered yourself.

...how is this a better technique for a WAV?

I've already told you I don't use DGIndex for getting a WAV file:

I usually get the AC3 from the DVD when creating the D2V file (although you can use PGCDemux or a few other ways), and then convert it to WAV in BeSweet. Afterwards I load it into Audacity to do whatever I want with it.