Alternative Audio Decoding Methods

Apparently a lot of people are not happy with their audio tracks they get using "standard" tools suggested in my guides. This page is for you. It's also for all of you who have a special DVD which has a soundtrack on it, that ac3dec or flask will choke upon.

Total Recorder

First you need to get a copy of Total Recorder 2.2. You'll need a registration key to record longer than 40 secs. I can't give you one but there are good sites like astalavista.box.sk, aren't there? Then insert the DVD you want to get the soundtrack from, start up the DVD player, select the right soundtrack, but don't start playback yet. Then start Total Recorder.

Select Options, Recording source and parameters..



Select Software accelerated recording/converting. Make sure you check Max speed, too. If you want to convert the audio you can do it right here, or you can do it later in VirtualDub - the choice is up to you.

Then press the Record button in the main Total Recorder. Don't worry.. recording won't start yet. Now go back to your software DVD player and start playing. Since there are often Trailers, Ads, FBI warnings and other crap before the main movie, you should go the chapter selection and select chapter 1. That should start the main movie right away. You can see that playback is greatly accelerated. After some time recording will finish. The press Control-S to save the recorded stuff to a wav file. 

>> JOIN AVI & WAV

Graphedit

Here we go with the ugly one. Let me just say that I really hate that program... I've never had that so trouble converting an audio track than with Graphedit. So since I consider this to be an expert method I won't explain again how to rip, merge and all the other basic stuff. 

Graphedit is a part of the Microsoft Media SDK. It'll let you connect several direct show filters together. You can use the vob splitter, ac3 parser and ac3 decoder if your favorite software player. For your help I've included the Xing direct show filter which is the only one that can handle ac3 files. Download the package, and unzip it. Then run register.bat to register the necessary filters. 

First we need to get the ac3 file out of our ripped vob files. To do so you can use vstrip and a streamlist. After you have got the right ac3 file start up Graphedit. Press File - Open:

>> EXTRACTING AC3

Make sure you select All Files in the File Type selection. Then select the ac3 file you've just stripped from the VOBs.




After that if you have installed all filters, you should get something like that graph:

If you press the play button you should hear the audio file playing. Now press Graph - Insert Filters:

Go to DirectShow Filters. Insert the following Filters: Wav Dest and Dump. If those two don't show up you didn't register it. Go back to the start to see how to register the DirectShow filters provided. When you insert the dump filter you're being asked to provide a media filename. Enter whatever name you want but the extension has to be .wav .



After that you have to delete the Default DirectSound Device and to reconnect the filters. You can connect two filters by clicking on the Output rectangle of a filter (on the right side), keep pressing the mouse button and draw a line to the input of the next filter (on the left side). Here's how it looks:

Connect the filters in the following order: XingDVD Audio Decoder to WAV Dest, WAV Dest to Dump. In the end it should look like this:

Now you just have to click on the play button and after some time the ac3 will be decoded to wav. To set the options of the Audio decoder, right click on the XingDVD Audio Decoder, and chose properties. You can select your sampling rate, for instance you can downsample to 44.1khz if you need it.



There are other DirectShow filters that allow you to decode audio in Graphedit. However, none of the other filters allows for direct ac3 parsing and that would mean you have to do each vob file separately, and since starting from the 2nd vob those files don't have a valid ac3 header, you have to copy the header from the 1st vob, then append the 2nd file to this header, and repeat the procedure for each vob. And in the end you have to merge the wav files.

[UPDATE] Using the latest vstrip, it's possible to strip the soundtrack from the VOBs and still getting vob files as result - but they won't contain the video, just the GOPs - and to load these into Graphedit which allows you to use other filters than the one from Xing. The command line is: vstrip @streamlist.txt output.vob g.

The following method has been outdated since the release of the I-media multiple MPEG-2 source filter. I strongly suggest you use that filter since processing VOBs one at a time is really a STUPID idea and will only lead to unnecessary trouble. The I-media Filters contain a really good how-to including screen-shots so make sure you read the readMe.Html file in the package!!!!!

Still not scared? Ok here we go. But not the usual detailed stuff. If you want to do this you must be a real hardcore buff and not afraid of anything. First we need to find out which ac3 track we want go get. Connect the filters like that and press play - you'll hear what soundtrack you've selected:

After selecting the right soundtrack, connect the filters like that:

Just for your information: All filters are DirectShow filters, just the Default DirectSound Device can be found in AudioRenderers. The filters have the following names - starting with vts_01_1.vob : File Source (Asynch) - Ligos MPEG Splitter - InterVideo Audio Decoder - WAV Dest - Dump. The last filter will pop up a file selection dialog. Just select any path and filename, but it has to be a wav file.

Repeat this step for each vob file. Unfortunately only the 1st vob has a valid header, therefore you need to copy headers. Run the headers.exe file provided in the Graphedit package to extract the vob headers from the first vob. Sample command line, assuming all files are in the same folder:
header vts_01_1.vob header.vob
Make as many copies of the header.vob as you need (number of VOBs minus one), then use FileMerger to merge the header and the vts_01_x.vob files. You don't have to do this step for the first vob file obviously. Another method would be to rip the first vob, then extract the header to vts_01_2.vob, then rip the 2nd vob and append to the extracted header. Yeah.. I told you it was complicated, didn't' I? Don't forget to right click on the InterVideo Audio Decoder and set it's properties.


Also make sure that you don't overwrite your wave files. Name them vts_01_1.wav, vts_01_2.wav, etc. In the end you have to merge these wave files together using a wave editor. I had a nice small wav merger but it doesn't work in Windows2000 so there's no point in releasing it - since Windows2000 is my suggested ripping system. Anyways.. there are enough good audio editors that allow you to merge wav files. You probably already have one on the installation CD of your soundcard. Make sure you merge the files in the right order, then get rid of the partial wav files.

>> MULTIPLEXING

 

Сайт создан в системе uCoz