eac3to - audio conversion tool [Archive] - Page 238

Gew

2nd November 2012, 20:07

Hi guys!

Say (for instance) I have a an old DVD film that has mono audio, yet they have duped the audio to 2 channel Dolby (AC3) for it's DVD release.

And oh, I know, sometimes they have used artificial tools to "post-produce" true stereo so that i.e. sounds coming from the right goes on right channel, etc.

Anyways, say I just want my computer backup to be of mono audio. I used to use LAME for audio, and then I just pulled the "-m m" syntax to downmix stereo to mono. This worked a charm.

Now I've started using AVC+AAC. Weird though, I use the eac3to utility, and can't seem to find a way to downmix stereo. If AC3 has e.g. 5.1 sound I can use "eac3to in.ac3 out.m4a -down2" to easily downmix the output to stereo.

Hence, I was really hoping for a "-down1" option, yet this won't work. Any easy approach that I have missed out on, that makes this downmix possible with one simple eac3to line. Or will I simply need to combine several tools?

Cheers!

tebasuna51

2nd November 2012, 21:24

There are a eac3to parameter, -mono, to extract the Center channel from a 5.1 or the Left channel from 2.0, but don't exist a downmix 2.0 -> mono.

You can use BeHappy to decode AC3, downmix to mono, maybe normalize, and convert to AAC.

Snowknight26

3rd November 2012, 19:57

madshi,

This TrueHD sample (http://stfcc.org/misc/eac3to_corrupt_mlp.thd) does not decode properly with eac3to but ffmpeg can decode it fine. Might be time to release a new version? :p

eac3to:
eac3to.exe eac3to_corrupt_mlp.thd eac3to_corrupt_mlp.flac
TrueHD, 7.1 channels, 48kHz
Decoding with libav/ffmpeg...
Remapping channels...
Encoding FLAC with libFlac...
[mlp @ 008f5d90] Restart header sync incorrect (got 0x0598)
The libav decoder reported error -1 while decoding.
Creating file "eac3to_corrupt_mlp.flac"...
Aborted at file position 262144.

ffmpeg:
ffmpeg.exe -i eac3to_corrupt_mlp.thd eac3to_corrupt_mlp.flac
ffmpeg version N-46146-g11d695d Copyright (c) 2000-2012 the FFmpeg developers
built on Oct 29 2012 18:10:27 with gcc 4.7.2 (GCC)
configuration: --enable-gpl --enable-version3 --disable-pthreads --enable-runtime-cpudetect --enable-avisynth --enable-bzlib --enable-frei0r --enable-libass --enable-libopencore-amrnb
--enable-libopencore-amrwb --enable-libfreetype --enable-libgsm --enable-libmp3lame --enable-libnut --enable-libopenjpeg --enable-libopus --enable-librtmp --enable-libschroedinger
--enable-libspeex --enable-libtheora --enable-libutvideo --enable-libvo-aacenc --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libxavs
--enable-libxvid --enable-zlib
libavutil 52. 1.100 / 52. 1.100
libavcodec 54. 69.100 / 54. 69.100
libavformat 54. 35.100 / 54. 35.100
libavdevice 54. 3.100 / 54. 3.100
libavfilter 3. 20.109 / 3. 20.109
libswscale 2. 1.101 / 2. 1.101
libswresample 0. 16.100 / 0. 16.100
libpostproc 52. 1.100 / 52. 1.100
[truehd @ 0000000003822f80] max_analyze_duration 5000000 reached at 5000000
[truehd @ 0000000003822f80] Estimating duration from bitrate, this may be inaccurate
Input #0, truehd, from 'eac3to_corrupt_mlp.thd':
Duration: N/A, start: 0.000000, bitrate: N/A
Stream #0:0: Audio: truehd, 48000 Hz, 7.1, s32
Output #0, flac, to 'eac3to_corrupt_mlp.flac':
Metadata:
encoder : Lavf54.35.100
Stream #0:0: Audio: flac, 48000 Hz, 7.1, s16, 128 kb/s
Stream mapping:
Stream #0:0 -> #0:0 (truehd -> flac)
Press [q] to stop, [?] for help
size= 8095kB time=00:00:44.64 bitrate=1485.5kbits/s
video:0kB audio:8087kB subtitle:0 global headers:0kB muxing overhead 0.099678%

bugmen0t

4th November 2012, 23:01

Hi

I found 3 blurays that i cant demux with eac3to 3.24

Brave 3D, Brave 2D and Finding Nemo 3D.

All those are Disney blurays, i allways get this error :

a04 libav Restart header sync incorrect (got 0x0598)
a04 The libav decoder reported error -1 while decoding.

I was able to demux those with tsmuxer.

Could you please have a look at it ?

Thanks

LOG :

F:\BDREMUXES\Brave (2012) 3D>eac3to.exe E:\BDMV\PLAYLIST\00802.mpls -demux
M2TS, 2 video tracks, 6 audio tracks, 4 subtitle tracks, 1:34:12, 108.889p
1: Chapters, 37 chapters
2: h264/AVC (left eye), 1080p24 /1.001 (16:9)
3: h264/AVC (right eye), 1080p24 /1.001 (16:9)
4: TrueHD/AC3, English, 7.1 channels, 48kHz
(embedded: AC3, 5.1 channels, 640kbps, 48kHz)
5: AC3 Surround, English, 2.0 channels, 320kbps, 48kHz
6: AC3 Surround, English, 2.0 channels, 320kbps, 48kHz
7: AC3, French, 5.1 channels, 512kbps, 48kHz
8: AC3, French, 5.1 channels, 640kbps, 48kHz
9: AC3, Spanish, 5.1 channels, 640kbps, 48kHz
10: Subtitle (PGS), English
11: Subtitle (PGS), English
12: Subtitle (PGS), French
13: Subtitle (PGS), Spanish
Creating file "00914 - Chapters.txt"...
a04 AC3 encoding doesn't support back channels. Will mix them into the surround.

v02 Extracting video track number 2...
a04 Extracting audio track number 4...
a05 Extracting audio track number 5...
a09 Extracting audio track number 9...
s11 Extracting subtitle track number 11...
a06 Extracting audio track number 6...
v03 Extracting video track number 3...
s10 Extracting subtitle track number 10...
s13 Extracting subtitle track number 13...
s12 Extracting subtitle track number 12...
a04 Extracting audio track number 4...
a08 Extracting audio track number 8...
a07 Extracting audio track number 7...
v03 Creating file "00914 - 3 - h264 (right eye), 1080p24.h264"...
a04 Extracting TrueHD stream...
a04 Extracting TrueHD stream...
a04 Decoding with libav/ffmpeg...
a04 Remapping channels...
a04 Mixing surround channels...
a04 Encoding AC3 <640kbps> with libAften...
a04 libav Restart header sync incorrect (got 0x0598)
a04 The libav decoder reported error -1 while decoding.

jj666

4th November 2012, 23:52

Those blu-ray's you mention aren't even released yet :-P

Have you tried latest FFMPEG standalone, like the chap posting above you did?

Cheers,

-jj-

Overdrive80

5th November 2012, 04:08

Joniii

5th November 2012, 06:34

When I encode 2.0 AC3 file to DTS, it output 5.0 DTS, anyone know why? I use Surcode 1.0.23.

Sparktank

5th November 2012, 08:28

When I encode 2.0 AC3 file to DTS, it output 5.0 DTS, anyone know why? I use Surcode 1.0.23.

Why not just manage it yourself in SurCode than using eac3to?

You can set it up to create a 2.0 in SurCode.

rapscallion

8th November 2012, 21:36

Atak_Snajpera

8th November 2012, 22:34

libaften is an ac3 encoder not truehd decoder.

tebasuna51

8th November 2012, 22:42

I notice in the eac3to dir there is only libaften.dll, dtd 2008.
No ffmpeg files at all.

avcodec.dll and avutil-50.dll are the special madshi dll's for libavcodec support. Only madshi can update that.

rapscallion

8th November 2012, 23:18

libaften is an ac3 encoder not truehd decoder.
Ok, then I'm a bit confused by this in Madshi's first post in this thread:
The libav/ffmpeg TrueHD decoder works perfectly fine including full 7.1 decoding.

What libav, ffmpeg or other file(s), do I need to decode TruHD 7.1 to wavs?

BTW, I use the gui eac3to and More.

@tebasuna51, avcodec.dll and avutil-50.dll are in my dir and are dated 2010. Are they relevant to TruHD 7.1 decoding?

tebasuna51

9th November 2012, 12:02

... avcodec.dll and avutil-50.dll are in my dir and are dated 2010. Are they relevant to TruHD 7.1 decoding?

Yes, inside there are the TruHD libav/ffmpeg decoder.

rapscallion

9th November 2012, 16:23

Thank you...it can be a bit confusing because of the names.

Any idea why they're causing this error (from previous post) on a retail THD 7.1 track? :

Writing WAVs...
[libav] Substream 0 parity check failed <WARNING>
[libav] Substream 0 checksum failed <WARNING>
[libav] Substream 0 length mismatch. <WARNING>
The libav decoder reported error -1 while decoding. <ERROR>
Aborted at file position 262144. <ERROR>

Snowknight26

9th November 2012, 16:56

No one knows if the discs are bad or if they are using a new format that the old versions of ffmpeg that eac3to uses don't support, but a solution was given so why not go with that?

rapscallion

9th November 2012, 17:02

No one knows if the discs are bad or if they are using a new format that the old versions of ffmpeg that eac3to uses don't support, but a solution was given so why not go with that?

If you're talking about ffmpeg, I wouldn't have a clue on how to apply that with the Eac3to and More Gui.

I'll try to demux with Tsmuxer

Snowknight26

9th November 2012, 17:33

Extract the TrueHD track as you would normally then use ffmpeg to convert it to whatever (intermediate) format you need. I showed how to do the latter part in my first post.

rapscallion

9th November 2012, 17:40

Extract the TrueHD track as you would normally then use ffmpeg to convert it to whatever (intermediate) format you need. I showed how to do the latter part in my first post.
Yes, I saw that, however, that's way beyond my understanding/ability.

Snowknight26

9th November 2012, 17:53

http://ffmpeg.zeranoe.com/builds/

Grab the latest static release. Extract somewhere.

Go to the bin folder and start a command prompt window there.

ffmpeg -i file.thd file.flac

Voila.

rapscallion

9th November 2012, 18:03

Thanks for you help and response.

Yes, just did that from here : http://www.videohelp.com/tools/ffmpeg

Which linked to the site you posted.

Then downloaded the WinFF front end. However, no output option for wavs.
Doing the cmd line your way has output of flac. Will I be able to convert that to 8 wavs?

I'll give it a try.

Edit: must be something funky with this track as I received this :

"Invalid data found when processing input"

sneaker_ger

9th November 2012, 18:04

Try:
eac3to input.flac output.wavs

Snowknight26

9th November 2012, 18:58

"Invalid data found when processing input"

Rename the input to .thd.

rapscallion

9th November 2012, 19:23

Rename the input to .thd.
Done, but still getting errors. This track is definitely skunked.

[truehd @ 024c0900] Lossless check failed - expected 4f, calculated 53.
[truehd @ 024c0900] mlpparse: Parity check failed.
[truehd @ 024c0900] Lossless check failed - expected 33, calculated 83.
[truehd @ 024c0900] mlpparse: Parity check failed.
[truehd @ 024c0900] Lossless check failed - expected 9d, calculated 72.
[truehd @ 024c0900] mlpparse: Parity check failed.
[truehd @ 024c0900] Lossless check failed - expected b2, calculated 41.
[truehd @ 024c0900] Invalid nonrestart_substr.
Error while decoding stream #0:0: Invalid data found when processing input
[truehd @ 024c0900] Stream parameters not seen; skipping frame.
Last message repeated 32 times
[truehd @ 024c0900] mlpparse: Parity check failed.
[truehd @ 024c0900] Lossless check failed - expected da, calculated 77.
[truehd @ 024c0900] mlpparse: Parity check failed.
[truehd @ 024c0900] Lossless check failed - expected f0, calculated 3f.
[truehd @ 024c0900] mlpparse: Parity check failed.

Snowknight26

9th November 2012, 20:47

Most likely because the source uses seamless branching. Only 6 bytes of probably over 3GB is corrupt. You won't even hear it.

rapscallion

9th November 2012, 21:56

Most likely because the source uses seamless branching. Only 6 bytes of probably over 3GB is corrupt. You won't even hear it.
Thanks, that is probably the issue, because the source does have seamless branching.
Though I've never had an issue with seamless branching audio tracks before.

When I demuxed the track, Tsmuxer stated "overlaps" at 2 different time points and "fixed" it.

Problem is that I need to convert to wavs, not use the THD track in it's current format.

Furiousflea

11th November 2012, 02:32

No one knows if the discs are bad or if they are using a new format that the old versions of ffmpeg that eac3to uses don't support, but a solution was given so why not go with that?

Solution doesn't work for "Brave".

Countless lossless check failed/parity check/numerous other errors when using FFMPEG command line.

There's about 5 gaps in audio ranging from 2ms to 30ms so...

Kinda stuck until something gets updated.

Furiousflea

11th November 2012, 02:32

Thanks, that is probably the issue, because the source does have seamless branching.
Though I've never had an issue with seamless branching audio tracks before.

When I demuxed the track, Tsmuxer stated "overlaps" at 2 different time points and "fixed" it.

Problem is that I need to convert to wavs, not use the THD track in it's current format.

That's not the issue, there's something different about these tracks.

Snowknight26

11th November 2012, 05:10

Solution doesn't work for "Brave".

Countless lossless check failed/parity check/numerous other errors when using FFMPEG command line.

It does work. There are 6 errors total, all which are only one byte.

r0lZ

11th November 2012, 12:34

@madshi
As you can see here (http://forum.doom9.org/showthread.php?p=1600104#post1600104), I'm trying to implement TrueHD support in my BD3D2MK3D GUI. Currently, the libav decoder crashes with this message: "Too many audio samples in frame" and when that problem occurs, eac3to hangs.
Is it a solution for that problem? Can you add a check in the next eac3to version to return an error instead of hanging?
(Sorry if that bug has already been reported. I'm too lazy to read the whole thread!)
Thanks!

Furiousflea

11th November 2012, 13:56

It does work. There are 6 errors total, all which are only one byte.

Maybe with your track, which film is it exactly? I'm talking about "Brave" the 2D version I'm trying to remux the main movie but with FLAC audio instead of the TrueHD 7.1 track.

eac3to crashes when using my nero audio encoder (I realize it can't do 7.1 but thought I'd try just to see).

libav give the message you talk about so no-go there.

But when using FFMPEG after demuxing the .THD track, I get about 4 different types error continuously throughout the whole stream...

1. Too many audio sample in frame
2. audio sync mismatch

Well basically lots of red error messages for first 14 mins or so.

The real problem though is that even if this output is close to the original, there's still the audio gaps that can't be removed.

Joniii

11th November 2012, 14:43

I demuxed Pixars new Finding Nemo BD and got some warnings.

First this when demuxing, I wonder if the audio is 14 frames off or did eac3to fix the video overlap?

eac3to v3.24
command line: eac3to F:\FINDING_NEMO 1) 2: d:\nemo.h264 3: d:\nemo.thd
M2TS, 1 video track, 8 audio tracks, 13 subtitle tracks, 1:40:54, 24p /1.001
1: Chapters, 32 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, English, 7.1 channels, 48kHz
(embedded: AC3 EX, 5.1 channels, 640kbps, 48kHz)
...
[a03] Extracting audio track number 3...
[a03] Extracting TrueHD stream...
[v02] Extracting video track number 2...
[v02] Creating file "d:\nemo.h264"...
[a03] Creating file "d:\nemo.thd"...
[v02] Video overlaps for 14 frames at playtime 0:00:49. <WARNING>
Video track 2 contains 145150 frames.
eac3to processing took 5 minutes, 31 seconds.
Done.

Then when trying to extract ac3 core from the TrueHD I just demuxed, I get this.

eac3to d:\nemo.thd h:\nemo.ac3 -core
TrueHD, 7.1 channels, 48kHz
AC3 encoding doesn't support back channels. Will mix them into the surround.
Decoding with libav/ffmpeg...
Remapping channels...
Mixing surround channels...
Encoding AC3 <640kbps> with libAften...
[mlp @ 009e2d10] Substream 0 parity check failed
[mlp @ 009e2d10] Substream 0 checksum failed
[mlp @ 009e2d10] Substream 0 length mismatch.
The libav decoder reported error -1 while decoding.

Extracting ac3 directly from BD:

eac3to F:\FINDING_NEMO 1) 3: h:\nemo.ac3 -core
M2TS, 1 video track, 8 audio tracks, 13 subtitle tracks, 1:40:54, 24p /1.001
1: Chapters, 32 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, English, 7.1 channels, 48kHz
(embedded: AC3 EX, 5.1 channels, 640kbps, 48kHz)
...
a03 Extracting audio track number 3...
a03 Extracting AC3 stream...
a03 Creating file "h:\nemo.ac3"...
a03 Audio overlaps for 7ms at playtime 0:04:46.
a03 Audio overlaps for 28ms at playtime 0:05:38.
a03 Audio overlaps for 20ms at playtime 1:33:24.
a03 Starting 2nd pass...
a03 Realizing (E-)AC3 gaps...
a03 Creating file "h:\nemo.ac3"...
Video track 2 contains 145150 frames.
eac3to processing took 4 minutes, 52 seconds.
Done.

Joniii

11th November 2012, 14:51

Also with the new Total Recall I get this:

H:\eac3to>eac3to d:\total.thd e:\total.ac3 -core
TrueHD, 5.1 channels, 48kHz
Decoding with libav/ffmpeg...
Encoding AC3 <640kbps> with libAften...
Creating file "e:\total.ac3"...
[mlp @ 00712d70] Lossless check failed - expected 0, calculated 93
-[mlp @ 00712d70] Lossless check failed - expected 0, calculated ac
-[mlp @ 00712d70] End of stream indicated
[mlp @ 00712d70] Lossless check failed - expected 0, calculated b3
[mlp @ 00712d70] Lossless check failed - expected 0, calculated 1c
-[mlp @ 00712d70] End of stream indicated
[mlp @ 00712d70] Lossless check failed - expected 0, calculated 26
-[mlp @ 00712d70] Lossless check failed - expected 0, calculated 91
.......

Snowknight26

11th November 2012, 20:47

The real problem though is that even if this output is close to the original, there's still the audio gaps that can't be removed.

So use eac3to to demux the TrueHD track then use ffmpeg to convert it to whatever format you want.

Joniii

13th November 2012, 07:44

Joniii

13th November 2012, 07:50

Anyone know why EAC3TO adds +5ms delay when doing FLAC -> AC3?

5.1 AC3 to FLAC with EAC3TO, I then edit the file with Audacity and save as FLAC. Then again use EAC3TO to encode it back to AC3. After it has encoded it, it is off by +5ms when comparing to the original AC3 or FLAC saved by Audacity.

sneaker_ger

13th November 2012, 07:55

Encoder delay caused by MDCT (https://en.wikipedia.org/wiki/Modified_discrete_cosine_transform), not the fault of eac3to. Many lossy formats suffer from it.

Joniii

13th November 2012, 08:09

Encoder delay caused by MDCT (https://en.wikipedia.org/wiki/Modified_discrete_cosine_transform), not the fault of eac3to. Many lossy formats suffer from it.

Thanks :)

tebasuna51

13th November 2012, 09:44

Anyone know why EAC3TO adds +5ms delay when doing FLAC -> AC3?

You can avoid this delay with external Aften encoder using the -pad parameter:

eac3to input stdout.wav | Aften -b 640 -pad 0 -readtoeof 1 - undelayed.ac3

Without problems if sound begin with silence, else the first 5.33 ms can be distorted.

Joniii

13th November 2012, 16:29

You can avoid this delay with external Aften encoder using the -pad parameter:

eac3to input stdout.wav | Aften -b 640 -pad 0 -readtoeof 1 - %_.ac3

Without problems if sound begin with silence, else the first 5.33 ms can be distorted.

Thanks :)

sneaker_ger

13th November 2012, 18:13

Yes, for movies this is usually fine. No one cares about a few milliseconds of a 2 hour movie that are probably just silence anyways.

Joniii

14th November 2012, 07:44

Yes, for movies this is usually fine. No one cares about a few milliseconds of a 2 hour movie that are probably just silence anyways.

Yeah, noone will notice it and it doesn't bother me either but it's a plus if you can get rid of that +5.33ms in the progress. Also I wanted to understand why and where it is added.

G_M_C

14th November 2012, 10:17

Maybe the problems with truehd are a manifestation of the dreaded cinavia watermarking?

Furiousflea

14th November 2012, 17:02

So use eac3to to demux the TrueHD track then use ffmpeg to convert it to whatever format you want.

That doesn't remove gaps since no decoding\encoding is taking place.

This is still a problem. Brave is currently unwatchable with demux with eac3to -> FFMPEG encoding to FLAC because of the gaps. Audio is a good 200-300ms out.

Maybe eac3to can resolve gaps by simply demuxing doing a 2nd pass with some formats - but not THD - which is what we're dealing with here.

Snowknight26

14th November 2012, 18:13

That doesn't remove gaps since no decoding\encoding is taking place.

Sure it can, just use the .gaps file in conjunction with the FLAC file that ffmpeg creates.

henryperu77

14th November 2012, 19:59

Several new BD's with TrueHD have problems, for example it's impossible to directly encode TrueHD to DTS or AC3 with EAC3TO, it gives some red warnings and hangs. Also demuxing THD works but demuxing THD with AC3 fails.

[mlp @ 00b42d10] Substream 0 parity check failed
[mlp @ 00b42d10] Substream 0 checksum failed
[mlp @ 00b42d10] Substream 0 length mismatch.
a03 The libav decoder reported error -1 while decoding.

Hi i have a .thd audio file that i encoded with Dolby Media Encoder, i tried to make a thd+ac3 in order to mux on a .ts file but i get that same error

TrueHD, 5.1 channels, 48kHz
Decoding with libav/ffmpeg...
Encoding AC3 <640kbps> with libAften...
libav Substream 0 parity check failed
libav Substream 0 checksum failed
libav Substream 0 length mismatch.
The libav decoder reported error -1 while decoding.

Any help?

madshi

14th November 2012, 20:25

@tebasuna51, would you be willing to make a list of the most important things to fix in / add to eac3to? Not too big things, please, just bugfixes and minor changes/improvements. Thanks!

And for those of you who have problems with specific TrueHD tracks, please upload small samples. E.g. 2MB per sample should be enough. But before you upload, please make sure that the problem also occurs with the sample and not just with the full file, otherwise the less won't help. Thx.

henryperu77

14th November 2012, 20:39

How i can make a 2mb sample of a truehd file?? what program could i use?

Atak_Snajpera

14th November 2012, 22:49

dgsplit?

henryperu77

15th November 2012, 00:15

Ok i used dgsplit and this is the sample

http://www.mediafire.com/download.php?oopeorxdvqnk77v

thanks in advance

Snowknight26

15th November 2012, 00:40

@tebasuna51, would you be willing to make a list of the most important things to fix in / add to eac3to? Not too big things, please, just bugfixes and minor changes/improvements. Thanks!

You just made several people very happy. :)

And for those of you who have problems with specific TrueHD tracks, please upload small samples. E.g. 2MB per sample should be enough. But before you upload, please make sure that the problem also occurs with the sample and not just with the full file, otherwise the less won't help. Thx.

http://stfcc.org/misc/eac3to_corrupt_mlp.thd