TrapVector's Guide to switching 2.0 songs to 5.1

[trapvector]

Hi Everyone. This guide is for everyone who wants to take a 2.0 stereo song and reencode it to a Dolby Digital 5.1 track. I've wanted to do this for a couple of months now, and thanks to this site, and the expertise of various members (and the required programs of course), I've finally figured out how.

First, this would not be possible without the guide by Eye of Horus:
http://forum.doom9.org/showthread.php?s=&threadid=60137

Second, you will need the six programs under section 04 of that link.
--Note--I use BeSweet version 1.5b22
You also need the HNM filter plugin for Plogue Bidule http://www.niederma.de/freeware/Audio/VST/HNM_Filter.zip

Next, use the guide by Eye of Horus to complete the installation of Plogue Bidule and its plugins.

Now, to get started, take a regular song from your computer--it has to be in WAV format--if it's in WMA format, you have to convert it to WAV format. A free program to do this with is CDEX http://download.com.com/3120-20-0.html?qt=cdex&tg=dl-2001
Although, personally I use Acoustica because it's faster and I'd already purchased it.

Now for the other programs you need:
VirtualDubMod - Freeware
BeSweet - Freeware
--Get both of these from the Gordian Knot Rippack here:
http://prdownloads.sourceforge.net/gordianknot/Gordian.Knot.Rip.Pack.0.28.7.Setup.exe?download

AC3Machinev0.41 - Freeware
http://www.doom9.org/Soft21/Audio/AC3Machinev0.41.zip

TMPGEnc - free 30 day trial after which you must purchase
http://www.tmpgenc.net/e_download.html

TMPGEnc DVD Author 1.5 and TMPGEnc Sound Plug-in AC3
-free trial on DVD Author but no trial on AC3 plugin-
http://www.pegasys-inc.com/en/index.html

OK, here we go. Open r8brain. Load your song in the INPUT section.
Select 48000 under Resample to r8. Choose a save destination and click Perform r8brain. Now you have a 2 channel WAV resampled at 48000 khz which you need for Plogue Bidule.

Now, open Plogue Bidule. Follow section 05 of Eye of Horus' guide to the letter. When you are finished, make the following changes: link the first 5 outputs of Emigrator straight down into the first 5 inputs of Audio Recorder. Now open the palette and import 5 HNM filters which are under VST Plugins. Download this preset for the HNM filter: http://downloada.in-berlin.de/hnm_lowpass80.fxb
Open each HNM filter and load the preset. At the end, Plogue Bidule should look like this: http://downloada.in-berlin.de/hnmfilter.gif

Now, inside Plogue Bidule, open the Audio File Player and load the song that r8brain converted to 48000khz (Planet Earth_r8) for example. Now open Audio File Recorder and choose a Save destination and rename the file (Planet Earth_6ch) for example. Now click play in the Audio File Player. The song will play in real time, and at the end you will have a 6 channel wav of that song. Do not delete this file, as you will need it later!

Now, open AC3Machine. Tell it where BeSweet.exe is by selecting the BeSweet.exe section under Locations. Now under the Input section, load the multichannel wav you made with Plogue Bidule. Under Output, select Save Destination and give it a name like Planet EarthAC3 for example. Now click Give me AC3!

Now after all this, you have a Dolby Digital AC3 file of your song.
--Finally!--But you're only halfway finished...

You need to make a DVD compliant video stream which is the exact length of your song. I do this with pictures because to me, the video doesn't matter. Open VirtualDubMod. Click File, Open Video File, then open a bmp picture. --Note--The title of the picture must have a number sequence like "Black Screen 01" for example.
Now click File, Save As and give it a name. What you get is a 0 second AVI file of your picture which you need for:

TMPGEnc--Open this program and the project wizard will start. Click Next. On Project Wizard page 2/5 click Browse Video File. Select the 0 second avi you just made with VirtualDubMod. Under Browse Audio File, open the multichannel wav you made with Plogue Bidule. The one that's titled Planet Earth_6ch for example. This is the file you got right before you used AC3Machine. Click Next, click Ok.
You should be on page 4/5 of the wizard. There's a section called Makes File Size. Click the Arrow down button until it stops. Your bitrate should read 2000kbits/sec. --Note this looks terrible for moving pictures but it doesn't matter for still pictures. This way you can fit more 5.1 songs on a disc.
Finally click next and choose where to save. You'll get an M2V file and a WAV file about 8 minutes later. Delete the WAV file when it's finished.

Now the next program is TMPGEnc DVD Author. You MUST have the AC3 Sound plugin for this to work. Click Source Setup. Load your newly created M2V file under video. Under Audio settings select all files from the drop-down bar. Navigate to the AC3 file you made with AC3Machine. Your Audio Input format should read: Dolby Digital AC3
48000hz 6 channel. Now click Output, choose a save destination, and you'll get an Audio TS folder and a Video TS folder. Now, use your favourite burning program, and Voila! Regular music in 5.1
--Note when you add a second song, TMPGEnc DVD Author will give you an error message basically saying that the second track has different audio than the first. Just click Cancel, and load your AC3 file for the second song just like you did for the first one.

I know that this guide is convoluted as all hell, but if you make it through successfully, the results are well worth it, believe me.
I reencoded Duran Duran's first album, and the results are unbelievable! And, if you use 2000kbits/sec for all the video, then you can fit about 5 albums on one DVD.

Also, you can use this method to reencode the audio on a DVD that only comes with 2.0 channel to 5.1 I tried that out on a Star Trek TOS DVD and it sounds pretty good.

Again, Thanks to Eye of Horus and DSPGuru and everyone else who wrote all the guides i studied and the programs I used. And Doom9 of course.:)

[daphy]

HiHo trapvector

very complex methode, but serious how long do have to work on one DVD?

some ideas:

1. I wouldn´t use the ac3enc.dll anymore because it is known to be buggy (-> there are some better AC3 encoder around ;) )
2. there are some apps like DVDlab, Maestro... which allow to import a simple BMP and an audio stream to make a DVD (but this is a commercial solution)
-> freeware alternative:
use the SlideShowMovieMaker (http://www.joern-thiemann.de/subpage/index.htm?/tools/ssmm/) for stills or slides as you like in combination with the YMPEG as codec (http://www.divxmania.it/index.php?showtopic=4525) -> result MPEG stream
(length can be tuned in the SlideShowMovieMaker - you´ll find the exact length by using an audio editor on the source material)
3. muxing
use ifoedit/rejig -> create/author new DVD -> load MV2 stream + AC3 stream; chaptors can be added manually with a chaptor-file.

Maybe someone has other (easier) solutions :rolleyes:

CYA Daphy

[trapvector]

Hi daphy, Thanks for the comments. I knew there must be alternate methods to use for this type of conversion. I've looked at DVD Maestro (haven't looked at DVDLab yet), and your comment is right. I wanted to be able to do this using the tools that I already had because one of the issues (for me) is money-in that I don't have a lot of it:) I'd never heard of SlideShowMaker or YMPEG, but I'll check them out also. And I found that TMPEG DVD Author has some neat options for menus. As for the time involved, it takes me about 2 hours to do the entire conversion for one album. And a 4.7GB DVD can hold about 5 albums. As you said, there are probably a lot of other ways, it's just that this is the way that works best for me with my current level of knowledge and skill. I will check out those other apps though. What I'd love to do now is be able to encode 2.0 into DTS - That would be really cool! But SurCode DTS is a little outside the monetary forecast. Thanks a lot for your thoughts:)

[raquete]

Excuse me after long time without post:

Can someone give me a working link to download HMN filter please?

Thanks in advance and best regards to old good friends! :-)

[filler56789]

Excuse me after long time without post:

Can someone give me a working link to download HMN filter please?

Thanks in advance and best regards to old good friends! :-)

Even though you would deserve an http://forum.videohelp.com/attachments/2671-1279232225/uglylol.gif :p ,
here it goes:

https://web.archive.org/web/20051028111143/http://www.niederma.de/freeware/Audio/VST/HNM_Filter.zip

[tebasuna51]

There are a lot of obsolete software in this guide.

To make a slide show in DVD there are many options but the audio conversion can be do with a simple way:

ffmpeg -i ANY_STEREO_AUDIO -af "aresample=48000,surround=lfe_out=0" -c:a ac3 -b:a 448k -center_mixlev:a 0.707 OUT.ac3

You can try with 'angle' and 'focus' parameters (https://www.ffmpeg.org/ffmpeg-filters.html#surround) or others.

My recommendation is create a empty lfe channel (lfe_out=0) preserving the low frequencies in original channels (lfe_mode=add default) to let the audio equipment send them to subwoofer without interferences.

[SallyDog]

To clarify for us dummies, ANY_STEREO_AUDIO = mysong.wav (or mp3 or Flac, or whatever) ?

[tebasuna51]

To clarify for us dummies, ANY_STEREO_AUDIO = mysong.wav (or mp3 or Flac, or whatever) ?

Yes, any stereo audio than can be decoded by ffmpeg (212 codecs: aac, ..., wma).

[raquete]

@filler56789
:p lol...and thanks for the link. :thanks:

@tebasuna51
ffmpeg really seems very very nice!
Thank you old friend for hints, link and, as always, extreme good will!
:thanks:

[raquete]

I have a simple but strange question, i'm old and with influenza in last days and it makes my memory mixed.

tebasuna51 gave me hints with links but i want only the LFE from audio.

I want only LFE from the audio loaded in Audacity.

Few months ago i used HNM filter in Audacity and was working.(don't remember if was in my notebook 32x that broke)

Today, with new PC 64x, i'm still using Audacity but HNM don't work.

If i install Audacity 32x in my computer 64x HNM can work?
or
how to use FFMPEG in Audacity to get LFE cos i really don't know how to use FFMPEG and don't know how to use command lines.

[tebasuna51]

Hi Raquete, I hope you get better soon.

I don't quite understand what you want to do. This thread is for converting a stereo to multichannel.

Therefore, a stereo doesn't have an LFE channel that you can extract.

In my old post, I recommended leaving the LFE channel empty because any audio system with a subwoofer already handles bass reproduction through it, regardless of which channel it comes from.

Extracting bass from other channels to create an LFE is a mistake that can only cause problems.

However, if your goal is something else, try explaining it to me.

I'd rather teach you how to use the command line than try to use outdated software.

[raquete]

OH Yes tebasuna51, i know about the objetive of this magnific thread.

I do upmixes for 5.1 and use in 6.1 discrete channels too then i need the LFE.

The HNM worked in the past but if i good remember, i was using an x86 notebook

With the new notebook that is 64x, HNM don't work and i can't deal with command lines then i ask:

Have another way to get LFE without command lines?
or
If i install Audacity 32x in my 64x system HNM will work?

[tebasuna51]

Create a folder C:\Portable\0 and put here the ffmpeg.exe from https://forum.doom9.org/showthread.php?p=2031037#post2031037
Put also the 2 bat scripts attached.

Drag and drop any source stereo audio over 20_to_51.bat, or any 6.1 audio over 61_to_51.bat and you obtain the desired OUT_51.ac3 at same folder than source (can be in other folder than C:\Portable\0 using two explorer windows)

Work also over the first audio in any container (mkv,mp4,m2ts,...)

EDIT: Added also a 71_to_51.bat for sources 7.1
Of course the bat files can be edited to modify the bitrate, now: -b:a 448k
And also the output codec (to eac3, opus, wav, ...), now: -c:a ac3 for ac3 output

[raquete]

Tebasuna51, you've always been courteous and very kind to me. :-)


Thank you very much.

[raquete]

Tebasuna51,

I always will load waves and want waves results too!

works perfectly but is removing the LFE and i want only the LFE in the result.

How can be done?

[tebasuna51]

I always will load waves and want waves results too!

Reload the bat's, now the default output is wav.

works perfectly but is removing the LFE and i want only the LFE in the result.
I can't imagine the purpose of extracting only the LFE channel from a multichannel audio file. However, I've attached an X1_to_LFE.bat file.

It would be a different matter if you wanted to extract the bass from any audio, but, as I said, any audio system with a subwoofer does that automatically. Don't confuse the LFE (Low Frequency Effects) channel with the bass in an audio file.

Let me know what you need it for.

[raquete]

Reload the bat's, now the default output is wav.


I can't imagine the purpose of extracting only the LFE channel from a multichannel audio file. However, I've attached an X1_to_LFE.bat file.

Let me know what you need it for.

OK let me explain:

I do upmixes from stereo to 5.1 only for musics, never for movies!
I extract L, R, C, SL and SR in many ways from source stereo and got very very good results.

One of my sytems have only discrete inputs(RCA) and need the LFE channel!

I extract the LFE too but using equalizers and don't like because have lots of 'BUMs' and few "TUMs"(subs) (i want the 'BUMBO' from drums.. i hope you understand what i mean.

I mix the basses (from bass) with the LFE from drums to use as LFE channel in the discrete LFE input.


-->X1_to_LFE giving LFE empty(no sound) loading sources stereo or dual mono.

--> Is possible to extract from 70 Hz to 200Hz (70Hz at -17dB decreasing the audio level to 200Hz at -90dB) ?

!

[tebasuna51]

One of my sytems have only discrete inputs(RCA) and need the LFE channel!
Let me know that system (manual or a web with info)
Audio equipment can usually be configured so that the subwoofer reproduces the bass from all channels, not just the LFE channel.
It shouldn't matter if the LFE channel is empty.

I extract the LFE too but using equalizers and don't like because have lots of 'BUMs' and few "TUMs"(subs) (i want the 'BUMBO' from drums.. i hope you understand what i mean.

I mix the basses (from bass) with the LFE from drums to use as LFE channel in the discrete LFE input.
It seems you like how the Bidule plugin works.
Unfortunately, I've never used it and can't help you with it.

-->X1_to_LFE giving LFE empty(no sound) loading sources stereo or dual mono.
Of course, it only extracts original LFE channels; it never generates new information.

--> Is possible to extract from 70 Hz to 200Hz (70Hz at -17dB decreasing the audio level to 200Hz at -90dB) ?
It's possible, but you must also remove that information from the original channels to avoid duplication or even interference that cancels or distorts the bass.

The filter used should match the filter used by the specific audio equipment to route the bass to the subwoofer, and this cannot be guaranteed in general.

Try using the ffmpeg filter, removing the one that leaves the LFE empty, and see if you like it. Or experiment with other surround filter parameters. (https://ffmpeg.org/ffmpeg-filters.html#toc-surround)

[raquete]

Let me know that system (manual or a web with info)
Audio equipment can usually be configured so that the subwoofer reproduces the bass from all channels, not just the LFE channel.
It shouldn't matter if the LFE channel is empty.

It seems you like how the Bidule plugin works.
Unfortunately, I've never used it and can't help you with it.

Of course, it only extracts original LFE channels; it never generates new information.

It's possible, but you must also remove that information from the original channels to avoid duplication or even interference that cancels or distorts the bass.

The filter used should match the filter used by the specific audio equipment to route the bass to the subwoofer, and this cannot be guaranteed in general.
As i wrote, using discrete 5.1 inputs in the receiver,

Try using the ffmpeg filter, removing the one that leaves the LFE empty, and see if you like it. Or experiment with other surround filter parameters. (https://ffmpeg.org/ffmpeg-filters.html#toc-surround)

SONY STR K 880 manual:
https://manualmachine.com/sony/strk880/2326250-service-manual/
https://archive.org/details/manual_STRK880_SM_SONY_EN/page/n3/mode/2up
ps: i was electronic technician for more than 40 years and the informations that i write are (always) corrects.

The system is a receiver Sony 6.1 channels that have coaxial input, optical input(etc) but i use only 6 dscrete RCA inputs for multichannel.
If i don't have the LFE discrete audio to the RCA discrete LFE channel, the LFE channel don't give sound.
This receiver is used only for discrete RCA inputs and don't redirect the others channels to LFE channel.
When you use the 'multichannel input' option, the RCA inputs are discrete: L, R, C, LFE, SL and SR
If you don't use the LFE input, the LFE channel don't give any sound!

(I have other systems that redirect the basses to LFE but using coaxial or HDMI)

I don't use bidule for more than 16 or 20 years and forgot how to use!

sorry, i don't understand what you mean.
X1_to_LFE giving LFE empty(one mono channel without any sound) and i want only the LFE mono from stereo source.
(gee, seems that i'm drivig you to be crazy) :-)


how to remove that information from the original channels to avoid duplication or even interference that cancels or distorts the bass? :confused:
The Receiver Sony 6.1 don't redirect the basses to LFE using discrete RCA inputs
Using the ffmeg filter, let the LFE empty but i want the LFE channel. lol
I will try the link and see if i learn something.

Cheers tebasuna51. :-)

[tebasuna51]

...
When you use the 'multichannel input' option, the RCA inputs are discrete: L, R, C, LFE, SL and SR
If you don't use the LFE input, the LFE channel don't give any sound!
Thanks for the comprehensive information.

I've never seen anything like it; your RCA inputs don't have an LFE channel, but rather a dedicated SubWoofer input.

It seems you need to extract the bass frequencies beforehand and remove them from the other channels (a feature that all the AVRs I'm familiar with have built-in settings).

I'm sorry I doubted your requirements.

I tried adding some parameters to ffmpeg's surround function:
surround=lfe_mode=sub:lfe_low=70:lfe_high=200
The default are:
lfe_low
Set LFE low cut off frequency. By default, this is 128 Hz.
lfe_high
Set LFE high cut off frequency. By default, this is 256 Hz.
lfe_mode
Set LFE mode, can be add or sub. Default is add. In add mode, LFE channel is created from input audio and added to output. In sub mode, LFE channel is created from input audio and added to output but also all non-LFE output channels are subtracted with output LFE channel.

You can try other values or parameters.

[raquete]

Perfect

The best is: i can deal with the values.

This is a fantastic lesson: :)
lfe_low
Set LFE low cut off frequency. By default, this is 128 Hz.
lfe_high
Set LFE high cut off frequency. By default, this is 256 Hz.
lfe_mode
Set LFE mode, can be add or sub. Default is add. In add mode, LFE channel is created from input audio and added to output.
In sub mode, LFE channel is created from input audio and added to output but also all non-LFE output channels are subtracted with output LFE channel.

Thank you very much tebasuna51, you're more than a teacher, your are very kind and full of patience.:thanks:
Best regards. :-)

[raquete]

Tebasuna51,

Is it possible to have only the CENTER channel (only what is common to the center channel) without the L and R channels (where L-C and R-C result in a Mono center channel) in the last command line you sent me?

last command line you send is called 2.0_To_5.1_SW

ffmpeg -drc_scale 0 -i %1 -af "aresample=48000,surround=lfe_mode=sub:lfe_low=30:lfe_high=90" -c:a pcm_s24le OUTPUT_5.1.wav

pause

rem ffmpeg -drc_scale 0 -i %1 -af "aresample=48000,surround=lfe_mode=sub:lfe_low=70:lfe_high=200" -c:a ac3 -b:a 448k -center_mixlev:a 0.707 OUT_51.ac3

PS i change LFE values from 30HZz to 90Hz, i don't like 'retumbantes' sosunds between 100Hz and 150 Hz

And thank you, I'm really surprised by the efficiency of ffmpeg and your usual helpfulness!

[tebasuna51]

I don't understand what you're trying to do:

a) Extract the mono center channel from the generated 5.1 surround sound.

b) Obtain a 5.1 surround sound with all channels empty except the center channel.

c)...

[raquete]

I don't understand what you're trying to do:

a) Extract the mono center channel from the generated 5.1 surround sound.

b) Obtain a 5.1 surround sound with all channels empty except the center channel.

c)...
Lol, i'm really hard to explain the things, tebasuna51. Please excuse me. :)

What i want is center only where don't have the sounds from L and R channels

if have voices, basses, L and R sounds in the center,
→ i have only the voices and basses in the center channel in the command line (to 5.1) that you gave me;

rem ffmpeg -drc_scale 0 -i %1 -af "aresample=48000,surround=lfe_mode=sub:lfe_low=70:lfe_high=200" -c:a ac3 -b:a 448k -center_mixlev:a 0.707 OUT_51.ac3

Is possible ?

[tebasuna51]

The software should already be extracting the common portion of the left and right channels to the center channel.

However, vocals aren't always extracted directly to the center channel, as they can have some directionality.

And bass should never be in the center channel.

If you intend to use it for karaoke-type functions, it may not work well.

However, the surround function (https://ffmpeg.org/ffmpeg-filters.html#surround) has many configurable parameters.
But I've never used them and I don't know how to help you with your goal.

[raquete]

I thank you so much tebasuna51. :helpful:

Is not for use in karaoke-type functions, is to hear musics in 5.1.

In the end i have to learn how to deal with ffmpeg command lines.

:thanks:

[raquete]

Magnifics and lovely results here tebasuna51,
i have only more 2 despicables doubts:

When i load a music with -1dB and run the "20_to_51_SW" the general volume down ~ -5.5dBs. Can i encrease the general volume to -1dB or let as the script result give?

Another question is:
Can i change the 48KHz to 96KHz in the script?
Later i'll back to 48k, it's will used only to mix upmixes
(i'm doing mad things here)

Thanks in advance. :thanks:

[tebasuna51]

When i load a music with -1dB and run the "20_to_51_SW" the general volume down ~ -5.5dBs. Can i encrease the general volume to -1dB or let as the script result give?

It is your option, you can always amplify the output volume:

...,surround=lfe_mode=sub:lfe_low=70:lfe_high=200,volume=volume=4.5dB"

(I don't know if always need 4.5dB)

Can i change the 48KHz to 96KHz in the script?
Later i'll back to 48k, it's will used only to mix upmixes
(i'm doing mad things here)

I don't know for what, but of course you can do it with output wav for subsequent manipulation, not with AC3 output.

[raquete]

It is your option, you can always amplify the output volume:

...,surround=lfe_mode=sub:lfe_low=70:lfe_high=200,volume=volume=4.5dB"

(I don't know if always need 4.5dB)

I don't know for what, but of course you can do it with output wav for subsequent manipulation, not with AC3 output.

96-24 is to play waves using foobar.
i 'do the job' with 96k and later back to 48K for AC3.

Thank you for all answers and the lesson about what can be change in the script.
:thanks: