BeHappy - AviSynth based audio transcoding tool (UPD 19-07-2006)

[Chumbo]

Discovered a small bug:
2007-03-20 (Chumbo)
+ fixed a small UI bug to make sure when a job is finished, the item selected doesn't get deselected in the jobs list

BeHappyCAT (http://www.mytempdir.com/1262803) update.

[tebasuna51]

[EDIT] Here's the ac3 file I'm experimenting with. test file (http://www.mytempdir.com/1262380). btw, I also tried using DirectShow with the sonic audio decoder instead of ac3filter and got the same bloated output. Ugh...

Your test file is correct. And I can't reproduce your problem.

Really is not recommended use the DirectShowSource method because config differences between systems, player/convert options and difficult debug.

In my system I have ffdshow DSF to decode ac3 and work with BeHappy without this problem. Most sure is use the dedicated decoder NicAc3Source.

Sorry but I can't help you with your DirectShow configuration (is a mystery for me).

[Chumbo]

Your test file is correct. And I can't reproduce your problem.

Really is not recommended use the DirectShowSource method because config differences between systems, player/convert options and difficult debug.

In my system I have ffdshow DSF to decode ac3 and work with BeHappy without this problem. Most sure is use the dedicated decoder NicAc3Source.

Sorry but I can help you with your DirectShow configuration (is a mystery for me).
No worries, I really appreciate you trying to duplicate. I kinda' figured it was a filter issue and, I think, you validated that and am grateful to you for this. :) Many thanks.

[shon3i]

One more thing to do:

When is [..] for destination pressed, and common dialog show, should filename filed in common dlg, be filed with filename from destination textbox without full path, only filename.

[Chumbo]

Let me take a look at that for you shon3i.

[EDIT] here you go shon3i:
2007-03-21 (Chumbo)
+ Destination/output dialog now prefills the file name

Latest BeHappyCAT (http://www.mytempdir.com/1265030) build.

[shon3i]

Not working in my type.

http://img208.imageshack.us/img208/8388/untitledgi5.jpg

Maybe you didn't understand me.

[Chumbo]

Not working in my type.
...
Maybe you didn't understand me.
Oh no, I understood perfectly. I may have copied the wrong BeHappy.exe in my haste. Sorry, let me double check it.

[EDIT] I'm an idiot. I uploaded the wrong file. It should be the 3/21 build. Here's the right link. BeHappyCAT (http://www.mytempdir.com/1265030).

[shon3i]

Thanks Chumbo, work like charm :)

@Chumbo, tebasuna51, Alwa, can we stop development now, and sum what we have here

BeHappy.exe 03/21/2007 version 0.1.9.35190
AvisynthWrapper.dll ? 16 or 32, you tebasuna51 said that all encoders support 32bit output except aud-x and CT. Did i were have problems using 32bit and CT encoder on 16bit audio?
And can somebody upload lastest versions of both thanks.

Aslo to upload lastest extensions.

What to use for ogg encoding and link?
What is last aften version and link?
What is good for wav2mono & wav2stereo extension?
What is last NicAC3?
What encoder for musepack?

[Chumbo]

Thanks Chumbo, work like charm :)

@Chumbo, tebasuna51, Alwa, can we stop development now, and sum what we have here

BeHappy.exe 03/21/2007 version 0.1.9.35190
AvisynthWrapper.dll ? 16 or 32, you tebasuna51 said that all encoders support 32bit output except aud-x and CT. Did i were have problems using 32bit and CT encoder on 16bit audio?
And can somebody upload lastest versions of both thanks.

Aslo to upload lastest extensions.

What to use for ogg encoding and link?
What is last aften version and link?
What is good for wav2mono & wav2stereo extension?
What is last NicAC3?
What encoder for musepack?
Glad it worked. :)

btw, my project contains all the changes made by CAT crew. ;) I also have the latest utilities. I think I have twolame.exe too since that wasn't there initially. I can pack up all the parts including the two versions of avisynthwrapper some time tonight.

[EDIT] I packed my BeHappy program folder into this package (http://www.mytempdir.com/1265592). btw, I used your most recent install, but I updated all the files that needed updating, i.e., aften, extensions, etc. I included the original/old avisynthwrapper dll but the new one is the active one.

[tebasuna51]

AvisynthWrapper.dll ? 16 or 32, you tebasuna51 said that all encoders support 32bit output except aud-x and CT. Did i were have problems using 32bit and CT encoder on 16bit audio?
AFAIK all encoders work with new AvisynthWrapper.dll and the added functions than work only when is necessary and not always like with old AvisynthWrapper.dll.
And can somebody upload lastest versions of both thanks.
Aslo to upload lastest extensions.

What to use for ogg encoding and link?
What is last aften version and link?
What is last NicAC3?
What encoder for musepack?
I agree with Chumbo package but :

BassAudio.extension (added .cda extension)
NicAudio.extension (added Normalize in mp123)

There are some files I don't know if are needed:
lame_enc.dll
wvgain.exe
wvselfx.exe
wvunpack.exe

And other I have new versions:
enc_aacPlus.exe 45.056 28/09/2006 12:03
enc_aacplus.dll 529.408 13/02/2007 19:28 Winamp v5.33
flac.exe 237.568 15/02/2007 17:38 v1.14
neroAacEnc.exe 843.776 12/02/2007 09:49 and SS2
oggenc2.exe 403.968 25/10/2006 14:30 v2.83
twolame.exe 294.912 22/03/2007 13:20 v03.10b
and
mppenc.exe 109.568 13/11/2006 02:00 v1.16

I use http://www.rarewares.org/ for:
LAME 3.97
twoLame 0.3.10b
Oggenc2.83 using aoTuVb5
Flac v1.14

Also:
http://kurtnoise.free.fr/index.php?dir=Aften/
http://www.wavpack.com/
http://www.musepack.net/

And this NicAudio.dll (http://nic.dnsalias.com/NicAudio_alpha3.zip)

What is good for wav2mono & wav2stereo extension?
Sometimes we need split multichannel input in mono/stereo wav's to be edited in stereo audio editors. Split in mono wav's is a BeSweet tool than can be accomplished with these utils (compressed are only 5 KB and are optional)

Here are (http://www.mytempdir.com/1265679) BassAudio.extension, NicAudio.extension and bass_cd.dll. This bass library (in AviSynth 2.5\plugins) enable BeHappy to open CD Audio directly.

[序列人]

Starting job bits0001.mpa->bits0001.m4a
Error: System.AccessViolationException: Attempted to read or write protected memory. This is often an indication that other memory is corrupt.
at BeHappy.AviSynthClip.dimzon_avs_init(IntPtr& avs, String func, String arg, AVSDLLVideoInfo& vi, AviSynthColorspace& originalColorspace, AudioSampleType& originalSampleType, String cs)
at BeHappy.AviSynthClip..ctor(String func, String arg, AviSynthColorspace forceColorspace, AviSynthScriptEnvironment env)
at BeHappy.Encoder.encode()



Some one can help me?

[setarip_old]

This is often an indication that other memory is corrupt.Sounds like the program is suggesting that you may have some bad RAM...

[alwa]

There are some files I don't know if are needed:
lame_enc.dll
wvgain.exe
wvselfx.exe
wvunpack.exe
BeHappy works fine without those files.

LAME 3.97 produces files with the wrong bitrate in ABR mode. Someone else had problems with it? At the moment i use ver. 3.96.1 which produces files with correct bitrate.

Aften (http://win32builds.sourceforge.net/aften/index.html)(Other link don't work right now)

@序列人: Can you reproduce the error or does it happen incidentally?

[Chumbo]

Starting job bits0001.mpa->bits0001.m4a
Error: System.AccessViolationException: Attempted to read or write protected memory. This is often an indication that other memory is corrupt.
at BeHappy.AviSynthClip.dimzon_avs_init(IntPtr& avs, String func, String arg, AVSDLLVideoInfo& vi, AviSynthColorspace& originalColorspace, AudioSampleType& originalSampleType, String cs)
at BeHappy.AviSynthClip..ctor(String func, String arg, AviSynthColorspace forceColorspace, AviSynthScriptEnvironment env)
at BeHappy.Encoder.encode()



Some one can help me?
You can try alternative ways of feeding your audio:
- load your mpa into graphedt to see what your filter chain is. Your system may not be able to create a filter chain or even if it does, you may have to server it to behappy differently.

- server your mpa file via an AVS script with directshowsource. Make sure to use the fps parameter. If you still get the same error, add the framecount parameter

- use graphedt to save a grf file of your filter chain without the last output, i.e., default directsound device. So your audio output pin should be open. Create an avs script and use your file.grf as the input to directshowsource. Same here, use the fps parameter and you may have to use the framecount parameter.

The other thing to consider is what is your mpa file? Is it actually an ac3 file? If it is, rename it to ac3 and feed it accordingly. Is it a dts file? You can try renaming it to dts. Check out the thread on evodemux as there's plenty of info there on the demuxed audio tracks from both hd dvds and blu-ray dics.

[EDIT] Here's a sample of the avs line:
directshowsource("myfile.grf",fps=23.976,video=false,audio=true,framecount=xxxxx)
Of course set your fps according to your specific feature's setting. Some say fps and framecount are not necessary with audio files, but I guess you can check for yourself. I find it doesn't hurt to use them.

[Chumbo]

@shon3i,
Thanks to tebasuna and alwa, I updated my folder with the latest files (except enc_aacPlus.exe because I couldn't find the version tebasuna has). I also fed the help text to the encoder's xxxxHelp.txt files for convenience. I also put the avisynth plugin DLLs in their own folder for you.

Here's my updated folder (http://www.mytempdir.com/1266896) link if you want to use it. I hope it helps lessen the work you have to do.

[tebasuna51]

I updated my folder with the latest files (except enc_aacPlus.exe because I couldn't find the version tebasuna has).

I don't remember from where I download the "AACPlus v2 Encoder (using Winamp 5.3 enc_aacplus.dll)" but here (http://www.mytempdir.com/1266951) is.

[shon3i]

Thanks, i alredy have lastest enc_aac files, plus i use 5.33/5.34 which is CT 8.03

[shon3i]

New package is out

Link (http://www.box.net/shared/nkihizx1dh)

i've add whole changelog later

[buzzqw]

thanks shon3i!

BHH

[Deckard2019]

New package is out
Thank you but could you just provide a zip or rar file instead ?

[shon3i]

Thank you but could you just provide a zip or rar file instead ?
Why?

I don't see the point?

[buzzqw]

not an installer...

BHH

[Deckard2019]

not an installer...
Thank you buzzqw ;)

[shon3i]

Then use all files from BeHappy folder and just zip that's all, including avs audio plugins, nothing simpler ;)

BeHappy

aften.exe
aud-x.extension
audxlib.dll
AviSynth Plugin DLLs
AvisynthWrapper.dll
BassAudio.extension
BeHappy.exe
BeHappy.exe.config
ConvertSample.extension
DownMix.extension
DuplicateChannels.extension
enc_aacplus.dll
enc_aacPlus.exe
enc_AudX_CLI.exe
ffmpeg.exe
ffmpeg.extension
flac.exe
flac.extension
lame.exe
libmmd.dll
MP4Box.exe
mppenc.exe
musepack.extension
neroAacEnc.exe
neroAacEnc_SSE2.exe
NicAudio.extension
nscrt.dll
oggenc2.exe
SSRC.extension
twolame.exe
twolame.extension
UpMix.extension
wav2mono.exe
wav2mono.extension
Wav2stereo.exe
Wav2stereo.extension
wavpack.exe
wavpack.extension


Avs Plugins

bass.dll
bassAudio.dll
bass_aac.dll
bass_ac3.dll
bass_ape.dll
bass_cd.dll
bass_flac.dll
bass_mpc.dll
bass_spx.dll
bass_wma.dll
bass_wv.dll
NicAudio.dll
soxfilter.dll


you can use this my package exe and run in administrative mode with "BeHappy_20070324.exe" /a, then setup will extract all files from package to C:\ without real install.

This package is designed for all ppl's who want install and encode easly, for that purpose, i planing to make package who will include lastest AviSynth and Net Framework 2.0.

[buzzqw]

no...no need to include .net2 just offer to download

BHH

[Chumbo]

no...no need to include .net2 just offer to download

BHH
Yeah, I agree on this one. :) You should be able to build your package to require the .net frmwrk 2 and then download it if it needs it. It's too much overhead to download for folks who don't have a fast connection and already have it.

[madshi]

Hi guys,

I'm trying to mux audio tracks from my German PAL DVDs to HD-DVD. For that to work I need of course to change the German audio track from 25fps to 23.976fps. I've tried "TimeStretch", but it's more intelligent than I like. It seems to do pitch correction. But that's *BAD* in my case, cause by far most German audio tracks have been sped up without pitch correction. So basically I need a stupid and simple stretch, much less sophisticated than BeHappy's TimeStretch.

Can anybody help me, please?

Thanks!

[tebasuna51]

Open your track in BeHappy and apply the timestretch needed and other changes, instead Enqueue use Export AviSynth Script.

Edit with Notepad (or similar) the avs changing:
"tempo" by "rate" in line:

TimeStretch(last, tempo=95.904)

The new line:

TimeStretch(last, rate=95.904)

Now open the avs with BeHappy and encode (uncheck the DSP's functions now).

[Chumbo]

Hi guys,

I'm trying to mux audio tracks from my German PAL DVDs to HD-DVD. For that to work I need of course to change the German audio track from 25fps to 23.976fps. I've tried "TimeStretch", but it's more intelligent than I like. It seems to do pitch correction. But that's *BAD* in my case, cause by far most German audio tracks have been sped up without pitch correction. So basically I need a stupid and simple stretch, much less sophisticated than BeHappy's TimeStretch.

Can anybody help me, please?

Thanks!
If you export the avs script, you'll note that all BeHappy does is do the math and put a line in the resulting avs script similar to this:
########################################
# [DSP: TimeStretch - 25 -> 23.976]
########################################
TimeStretch(last, tempo=95.904)
I know BeSweet is outdated, but it does frame conversion if you want to try it and compare.

[EDIT] I knew tebasuna would have the answer. :)

tebasuna, what do you think about updating the UI to include a checkbox for pitch in the stretch dialog? If checked, then tempo is used, otherwise rate is used.

[madshi]

Thank you guys! :) That should do the trick for me. Of course having this option in the GUI would be great, but as long as I can make it work, I'm happy.

[tebasuna51]

tebasuna, what do you think about updating the UI to include a checkbox for pitch in the stretch dialog? If checked, then tempo is used, otherwise rate is used.
There are three options with TimeStretch:
- rate: tempo and pitch changed
- tempo: only tempo changed
- pitch: only pitch changed

this third option can be used to repair the tracks in this case if we don't need change the fps in video.

Maybe for the next version.

[Chumbo]

There are three options with TimeStretch:
- rate: tempo and pitch changed
- tempo: only tempo changed
- pitch: only pitch changed

this third option can be used to repair the tracks in this case if we don't need change the fps in video.

Maybe for the next version.
From the way you broke it down, sounds like the default behavior is what madshi needed, i.e., tempo, because he didn't want pitch change. The suggestion to use rate would actually do both. I just want to make sure I understand. Thanks.

Here's the usage from the help file (just like you broke it down):
TimeStretch allows changing the sound tempo, pitch and playback rate parameters independently from each other, i.e.:
Sound tempo can be increased or decreased while maintaining the original pitch.
Sound pitch can be increased or decreased while maintaining the original tempo.
Change playback rate that affects both tempo and pitch at the same time.
Choose any combination of tempo/pitch/rate.

[tebasuna51]

From the way you broke it down, sounds like the default behavior is what madshi needed, i.e., tempo, because he didn't want pitch change. The suggestion to use rate would actually do both. I just want to make sure I understand.

The normal usage is 'tempo': modify the audio duration (preserving the pitch) to maintain the sync with a video with a new duration because is played at other fps. If the video is modified adding or deleting frames to change the fps, but maintain the duration, the audio don't need change.

But now madshi say: "cause by far most German audio tracks have been sped up without pitch correction", then the pitch is incorrect with the old fps, and tempo is incorrect for the new fps.

[Chumbo]

The normal usage is 'tempo': modify the audio duration (preserving the pitch) to maintain the sync with a video with a new duration because is played at other fps. If the video is modified adding or deleting frames to change the fps, but maintain the duration, the audio don't need change.

But now madshi say: "cause by far most German audio tracks have been sped up without pitch correction", then the pitch is incorrect with the old fps, and tempo is incorrect for the new fps.
Yeah, that makes sense which is what I thought. Thanks.

btw, I went ahead and started the next round of changes. ;) I had time so I added the UI for those options.

2007-03-30 (Chumbo)
+ added rate control to the TimeStretch DSP module - you can select rate, pitch or both
+ updated the version to 0.1.10.* from 0.1.9.*. With all the changes we made we should've update the version sooner

Get latest BeHappy update (http://www.mytempdir.com/1276813). Please test and report any problems.

[=Wolf=]

I has tried from AC3 5.1 448kbps-> Mp3 Sterio 2.0 and here that at me has left....
http://home.farlep.net/~wolf2/17_jpn.mp3

old version is work fine....

p.s sorry for my bad English :(

[madshi]

Hey, that's what I call a quick implementation of a feature suggestion! Thanks, Chumbo!! :)

May I suggest a renaming of the options, though? I think the names are not clear enough. Basically "change frame rate and pitch" can mean two things:

(1) It could mean: Do no pitch correction, so that the pitch is practically changed during conversion.

(2) It could mean: Do pitch correction, so that the pitch after the conversion is the same as before.

Basically I'm not sure when you say "change pitch" do you mean "do pitch correction"? Or do you mean "don't do pitch correction" (the latter of which would result in a pitch change during conversion).

I'd suggest these labels for the 3 options:

(1) Change frame rate without pitch correction.
(2) Change frame rate with pitch correction.
(3) Perform pitch correction without frame rate change.

Alternatively you could also replace the 3 radio buttons with 2 check buttons:

(1) Pitch correction.
(2) Run length correction (or "run time correction" or "frame rate correction").

Actually I think I'd prefer the 2 check buttons over 3 radio buttons. That's the most intuitive option set IMHO. Of course having both check buttons unchecked would simply do no modification to the audio stream at all.

------------

Btw, after doing some tests yesterday, the "rate" conversion is really exactly what I needed. The converted audio tracks are practically perfect. Perfect sync, perfect pitch.

Don't know if this is a known problem, but with the "tempo" method, the sync was not perfect. The final audio file was a few seconds too short. I believe it played slightly too "fast". No such problems with the "rate" option. Also I noticed that the whole conversion process needs significantly less time when using the "rate" option. That was to be expected, of course, cause it does much simpler manipulations to the audio file.

Anyway, for all people trying to convert PAL audio tracks to 23.976, I suggest this approach:

(1) Check whether the PAL track has too high pitch (this is the case for by far most PAL tracks, but not for all).
(2) If the pitch of the PAL track is too high, use the "rate" option.
(3) If the pitch is not too high, use the "tempo" option. However, you might have sync problems this way.

[tebasuna51]

The parameters 'tempo', 'pitch' and 'rate' are well know not only by AviSynth users but also by BeLight-BeSweet users then the name must appear in the options.

I'm opposed to use the words 'frame rate' when speak about audio, the real audio frame rate is never changed with this options.

Then I propose:

(1) Tempo changed preserving pitch.
(2) Pitch changed preserving tempo.
(3) Rate, tempo and pitch changed.

@madshi: "with the "tempo" method, the sync was not perfect. The final audio file was a few seconds too short."
Sometimes the exact change needed is not the exact relation between video framerates, then we need use the 'custom' option.

[tebasuna51]

I has tried from AC3 5.1 448kbps-> Mp3 Sterio 2.0 and here that at me has left....
http://home.farlep.net/~wolf2/17_jpn.mp3

old version is work fine....
Please send the original ac3 5.1 to try reproduce your problem.

Here with an ac3 5.1 448 kb/s, a downmix DSP and encode with Lame work fine.

[=Wolf=]

@tebasuna51
Here. Take it

http://file4.webfile.ru/1362905/17.ac3
48000Hz 448 kb/s tot , 6 chnls (3/2 .1)

[madshi]

The parameters 'tempo', 'pitch' and 'rate' are well know not only by AviSynth users but also by BeLight-BeSweet users then the name must appear in the options.
Agreed, that makes sense.

I'm opposed to use the words 'frame rate' when speak about audio
Yeah, makes sense to me, too.

Then I propose:

(1) Tempo changed preserving pitch.
(2) Pitch changed preserving tempo.
(3) Rate, tempo and pitch changed.
Again for me "pitch changed" is not clear. It could mean "pitch corrected" in the sense "changed=corrected". Or it could mean "pitch changed because it was not corrected". I suggest making use of the terms "pitch correction" and "no pitch correction", as that is a fixed term everybody understands.

Sometimes the exact change needed is not the exact relation between video framerates, then we need use the 'custom' option.
"Rate" gave me perfect sync where "Tempo" give me incorrect sync with the same audio track.

[Chumbo]

The parameters 'tempo', 'pitch' and 'rate' are well know not only by AviSynth users but also by BeLight-BeSweet users then the name must appear in the options.

I'm opposed to use the words 'frame rate' when speak about audio, the real audio frame rate is never changed with this options.

Then I propose:

(1) Tempo changed preserving pitch.
(2) Pitch changed preserving tempo.
(3) Rate, tempo and pitch changed.

@madshi: "with the "tempo" method, the sync was not perfect. The final audio file was a few seconds too short."
Sometimes the exact change needed is not the exact relation between video framerates, then we need use the 'custom' option.
Thanks guys for your feedback. That's exactly what I was looking for. I was wondering if I worded the options correctly. ;) I'll make the changes sometime today. Thanks again.

[tebasuna51]

Again for me "pitch changed" is not clear. It could mean "pitch corrected" in the sense "changed=corrected". Or it could mean "pitch changed because it was not corrected". I suggest making use of the terms "pitch correction" and "no pitch correction", as that is a fixed term everybody understands.
Pitch changed means more acute or more deep we don't know if is correct or incorrect.

This functions can be used for other purpose than audio movie tracks. For instance we can need change the pitch, preserving tempo, to accommodate music for different karaoke singers.

"Rate" gave me perfect sync where "Tempo" give me incorrect sync with the same audio track.
Yes, yes, the 'tempo' algorithm is more complex, and more inexact, like you say, for that we need any custom correction and try with different values (really I don't found seconds of difference, only ms, in movie tracks).

[tebasuna51]

@tebasuna51
Here. Take it

http://file4.webfile.ru/1362905/17.ac3
48000Hz 448 kb/s tot , 6 chnls (3/2 .1)

Sorry I get a "404 - Not Found" error with this link.

[alwa]

@tebasuna51
Here. Take it

http://file4.webfile.ru/1362905/17.ac3
48000Hz 448 kb/s tot , 6 chnls (3/2 .1)
link fix (http://webfile.ru/1362905)
I can't reproduce your issue.
I'm using NicAC3Source/Bass -> Downmix DSP -> LAME and the output is fine.

Again for me "pitch changed" is not clear. It could mean "pitch corrected" in the sense "changed=corrected". Or it could mean "pitch changed because it was not corrected". I suggest making use of the terms "pitch correction" and "no pitch correction", as that is a fixed term everybody understands.
then:
(1) Tempo changed, pitch correction.(hint: changes the length of the audio track while preserving the original pitch)
(2) Pitch changed preserving tempo.(hint: manipulates pitch, but Track length will kept unchanged)
(3) Rate, tempo and no pitch correction.(hint: changes the length of the audio track without preserving the original pitch )
?

Are the details right(i may be completely wrong)? :rolleyes:

[Chumbo]

...then:
(1) Tempo changed, pitch correction.(hint: changes the length of the audio track while preserving the original pitch)
(2) Pitch changed preserving tempo.(hint: manipulates pitch, but Track length will kept unchanged)
(3) Rate, tempo and no pitch correction.(hint: changes the length of the audio track without preserving the original pitch )
?

Are the details right(i may be completely wrong)? :rolleyes:
Thanks alwa. I'll add the hints too.

[EDIT]2007-03-31 (Chumbo)
+ updated TimeStretch rate control radio button labels per suggestions by tebasuna, alwa and madshi.
+ added hints/popup text to rate control radio buttons. I wound up using the AVISynth text for the hint text as it's straightforward.

Latest BeHappy update (http://www.mytempdir.com/1277580)

[=Wolf=]

I have tried to make 5.1 -> 2.0 so here that has left....
http://webfile.ru/1362905 <- ac3

AC3 5.1 -> MP3
Starting job 17_jpn.ac3->17_jpn.mp3
Found Audio Stream
Channels=2, BitsPerSample=32 int, SampleRate=48000Hz
lame.exe -b 128 -h -S --silent - "C:\FTP\Incoming\17_jpn.mp3"
Writing RIFF header to encoder's StdIn
Writing PCM data to encoder's StdIn
Finalizing encoder
Complete


AC3 5.1 -> Ogg
Starting job 17_jpn.ac3->17_jpn.ogg
Found Audio Stream
Channels=2, BitsPerSample=32 float, SampleRate=48000Hz
oggenc2.exe -Q --quality 3 -o "C:\FTP\Incoming\17_jpn.ogg" -
Writing RIFF header to encoder's StdIn
Writing PCM data to encoder's StdIn
Finalizing encoder
Complete

[tebasuna51]

I have tried to make 5.1 -> 2.0 so here that has left....
Seems all perfect.

1) Ac3 5.1 decoded by NicAc3Source produce 32 float. The downmix function works in 32 float, perfect. Now we want encode with Lame. Warning Lame don't support 32 float, the best quality supported is 32 int, then we need a conversion to 32 int. Done. All the process is make with the best quality possible.

2) Ac3 5.1 to Ogg. All the same but oggenc2 accept 32 float. Congratulations! we don't need convert the 32 float and is delivered to the encoder directly.

I test your sample and all is ok here.

[=Wolf=]

@tebasuna51
Then why at me such result turns out?
At work lame...
http://home.farlep.net/~wolf2/17_new.mp3

[tebasuna51]

Here (http://www.mytempdir.com/1278134) is your sample with:
- Open with NicAc3Source("17.ac3", DRC=0)
- Douwmix -> 2.0 with DLP II
- Normalize(45%) to be comparable with your sample
- Automatic conversion to 32 int
- Encoded with Lame 128 Kb/s CBR (like your sample)

Only your mp3 is broken? The ogg play fine?

[tebasuna51]

BeHappy update (http://www.mytempdir.com/1278153) :
- Changes to support new enc_aacPlus.exe from Shon3i. (NeroDigitalEncoder.cs, CodingTechnologiesAAC.cs)

- New enc_aacPlus.exe version to support enc_aacPlus.dll from WinAmp 5.33/5.34 and new MP4mux instead MP4Box. (enc_aacPlus.exe, MP4mux.exe)