eac3to - audio conversion tool

[Chouonsoku]

I'm having an issue with a DTS-HD MA 6.1 audio track from Ghost in the Shell 2: Innocence's Absolute Edition. The disc has a TrueHD 6.1 and a DTS-HD MA 6.1 track, and I can compress the TrueHD to FLAC, extract the DD5.1-EX core and also extract the DTS-ES 6.1 core, but when I compress the DTS-MA HD track to FLAC, it ends up being around 2.5 GB (vs. the DTS-MA HD's 1.8 GB) and the output file is also stretched somewhat. The original runtime of the film is 1hr 39m, however the FLAC of the DTS track reports it's 1hr 53m long, and during playback the audio is there, but it's distorted and stretched. No warnings or errors are reported by eac3to, and again all other tracks playback fine.

eac3to v3.24
command line: eac3to 1) 1:chapters.txt 2:innocence.h264 3: japanese.ac3 -core 3:japanese-truehd.flac 4:japanese.dts -core 4:japanese-dts.flac 7:english.sup
------------------------------------------------------------------------------
M2TS, 1 video track, 2 audio tracks, 3 subtitle tracks, 1:39:06, 24p /1.001
1: Chapters, 20 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, Japanese, 6.1 channels, 48kHz
(embedded: AC3 EX, 5.1 channels, 640kbps, 48kHz)
4: DTS Master Audio, Japanese, 6.1 channels, 16 bits, 48kHz
(core: DTS-ES, 6.1 channels, 16 bits, 1509kbps, 48kHz)
5: Subtitle (PGS), Japanese
6: Subtitle (PGS), Japanese
7: Subtitle (PGS), English
Creating file "chapters.txt"...
[a04] Extracting audio track number 4...
[v02] Extracting video track number 2...
[a04] Extracting audio track number 4...
[a03] Extracting audio track number 3...
[a04] Extracting DTS core...
[s07] Extracting subtitle track number 7...
[a04] Remapping channels...
[a03] Extracting TrueHD stream...
[a03] Extracting audio track number 3...
[a04] Patching bitdepth to 24 bits...
[a03] Decoding with libav/ffmpeg...
[a03] Extracting AC3 stream...
[a04] Decoding with ArcSoft DTS Decoder...
[a03] Remapping channels...
[a04] Encoding FLAC with libFlac...
[a03] Encoding FLAC with libFlac...
[a04] Creating file "japanese.dts"...
[v02] Creating file "innocence.h264"...
[a04] Creating file "japanese-dts.flac"...
[a03] Creating file "japanese-truehd.flac"...
[a03] Creating file "japanese.ac3"...
[s07] Creating file "english.sup"...
[a03] The original audio track has a constant bit depth of 16 bits.
[a04] The original audio track has a constant bit depth of 16 bits.
[a03] Superfluous zero bytes detected, will be stripped in 2nd pass.
[a03] Starting 2nd pass...
[a03] Decoding FLAC...
[a03] Reducing depth from 24 to 16 bits...
[a03] Encoding FLAC with libFlac...
[a03] Creating file "japanese-truehd.flac"...
[a03] The processed audio track has a constant bit depth of 16 bits.
Video track 2 contains 142560 frames.
Subtitle track 7 contains 814 captions.
eac3to processing took 34 minutes, 53 seconds.
Done.

[Snowknight26]

Most likely an issue with the ArcSoft DTS decoder. What version is it?

[sneaker_ger]

Thanks for clarification.
Does such pair of fps preserve or change the pitch?

Change.

[TDiTP_]

Stupid question. Why eac3to need MkvToolNix?

[Mark_A_W]

Hi all

Quick question..

What is the command line to upconvert a 6.1 TrueHD track to 7.1 FLAC (with centre rear doubled to L back and R back)?

There's a trilogy coming out soon (finally..) with 6.1 sound ;)

[TDiTP_]

What is the command line to upconvert a 6.1 TrueHD track to 7.1 FLAC (with centre rear doubled to L back and R back)?

-double7

But this command does work incorrect, you need to add -0,1,2,3,5,6,4:
eac3to input output -double7 -0,1,2,3,5,6,4

[Chouonsoku]

Most likely an issue with the ArcSoft DTS decoder. What version is it?

eac3to (v3.24) is up to date
Nero Audio Decoder (Nero 7) works fine
ArcSoft DTS Decoder (1.1.0.8) works fine
Sonic Audio Decoder (4.3.0.169) works fine
Haali Matroska Muxer (2011-03-03) is up to date
Nero AAC Encoder (1.5.4.0) is up to date
Surcode DTS Encoder doesn't seem to be installed
http://www.surcode.com
MkvToolnix (4.6.0.0, release version) is up to date

Is there a newer version?

Edit: Something I just noticed though, eac3to reports the track as being 16-bit however the bits aren't stripped in the second pass like they are for the TrueHD track. I'm testing decoding with Sonic Audio Decoder now to see if that works, although I'd rather not use it.

Edit 2: Sonic decoded it fine, same exact bitrate as the TrueHD (figured). Plays back fine and has the correct length. Still don't know why ArcSoft is having a problem with it.

[TDiTP_]

For decode track 4:

4: DTS Master Audio, Japanese, 6.1 channels, 16 bits, 48kHz
(core: DTS-ES, 6.1 channels, 16 bits, 1509kbps, 48kHz)

you need ArcSoft DTS Decoder 1.1.0.0 (you can find it here (http://www.mediafire.com/?4vsw2dydnz1evme)). Versions 1.1.0.7 and 1.1.0.8 can't decode DTS-ES 6.1 properly, it's known bug.


Sonic decoded it fine, same exact bitrate as the TrueHD (figured).

Sonic can't decode DTS-ES 6.1 properly too. BC-channel is empty always.

[Mark_A_W]

What about 1.1.0.5? Can that do 6.1?

[TDiTP_]

What about 1.1.0.5? Can that do 6.1?

i don't know. If you give me this version, i can test it.

[Thunderbolt8]

I dont remember any more, was there something 1.1.0.7 or .8 could do 1.1.0.0 couldnt do? either were one or two things (dtsma 1.0 or 2.0?), or 1.1.0.0 was just as good or better in each situation.

btw. madshi, would it make sense to integrate remuxing with LAVsplitter instead of haali in the future? are there any improvements (or even bugs) or can there be any improvement at all in this area from a technical point of view?

[Chouonsoku]

Alright, testing with 1.1.0.0 now. Now that you mention it I remember that being an issue but I didn't know which versions it affected.

Edit: It worked. Thanks. :)

[DivxBr]

Hi,

Don't get me wrong, I really love eac3to, and how it handles all audio formats, convert every audio stream to almost every audio format available. It's more than half job done for BD backup.

But there's one 5min+ stream (indexed by an mpls - playlist) that eac3to just can't handle right, and is getting me crazy to understand what I could be making wrong, looks like eac3to is messing up with timestamps and destroying A/V Sync.

That's what I'm doing:


eac3to 00202.mpls

1) 00202.mpls, 0:05:40
[44+45+46+47+48+49+50+51].m2ts
- Chapters, 8 chapters
- VC-1, 1080p24 /1.001 (16:9)
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz


I First extract the video and the audio tracks.


eac3to v3.24
command line: eac3to 00202.mpls 1) 2: "w:\test\ExtractedVideo.mkv" 3: "w:\test\ExtractedAudio.ac3"
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS), English
8: Subtitle (PGS), French
9: Subtitle (PGS), Spanish
10: Subtitle (PGS), Dutch
11: Subtitle (PGS), Spanish
12: Subtitle (PGS), Portuguese
13: Subtitle (PGS), Danish
14: Subtitle (PGS), Finnish
15: Subtitle (PGS), Norwegian
16: Subtitle (PGS), Swedish
[a03] Extracting audio track number 3...
[a03] Removing AC3 dialog normalization...
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
[a03] Creating file "w:\test\ExtractedAudio.ac3"...
[v02] Video overlaps for 7 frames at playtime 0:03:00. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:03:31. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:04:00. <WARNING>
[v02] The MKV file was created without making use of the gap/overlap information. <WARNING>
[v02] Please check whether audio is in sync. If it is in sync everything is fine. <WARNING>
[v02] Otherwise ask eac3to to repeat the muxing. It will then automatically make <WARNING>
[v02] use of the detailed gap/overlap information. <WARNING>
[a03] Audio overlaps for 120ms at playtime 0:00:40. <WARNING>
[a03] Audio overlaps for 95ms at playtime 0:01:37. <WARNING>
[a03] Audio overlaps for 52ms at playtime 0:01:55. <WARNING>
[a03] Audio overlaps for 96ms at playtime 0:02:27. <WARNING>
[a03] Audio overlaps for 119ms at playtime 0:03:00. <WARNING>
[a03] Audio overlaps for 105ms at playtime 0:03:31. <WARNING>
[a03] Audio overlaps for 91ms at playtime 0:04:00. <WARNING>
[a03] Starting 2nd pass...
[a03] Realizing (E-)AC3 gaps...
[a03] Creating file "w:\test\ExtractedAudio.ac3"...
Added fps value (24 /1.001) to MKV header.
Video track 2 contains 8144 frames.
eac3to processing took 15 seconds.
Done.



So it detected Audio AND Video overlaps. It corrected the audio, and gave me the possibility to correct the video overlap also.

So I do it:


eac3to v3.24
command line: eac3to 00202.mpls 1) 2: "w:\test\ExtractedVideo_gaps.mkv"
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS), English
8: Subtitle (PGS), French
9: Subtitle (PGS), Spanish
10: Subtitle (PGS), Dutch
11: Subtitle (PGS), Spanish
12: Subtitle (PGS), Portuguese
13: Subtitle (PGS), Danish
14: Subtitle (PGS), Finnish
15: Subtitle (PGS), Norwegian
16: Subtitle (PGS), Swedish
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
[v02] Video overlaps for 7 frames at playtime 0:03:00. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:03:31. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:04:00. <WARNING>
[v02] The MKV file was created without making use of the gap/overlap information. <WARNING>
[v02] Please check whether audio is in sync. If it is in sync everything is fine. <WARNING>
[v02] Otherwise ask eac3to to repeat the muxing. It will then automatically make <WARNING>
[v02] use of the detailed gap/overlap information. <WARNING>
Added fps value (24 /1.001) to MKV header.
Video track 2 contains 8144 frames.
eac3to processing took 12 seconds.
Done.


eac3to v3.24
command line: eac3to 00202.mpls 1) 2: "w:\test\ExtractedVideo_gaps.mkv"
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, English, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS), English
8: Subtitle (PGS), French
9: Subtitle (PGS), Spanish
10: Subtitle (PGS), Dutch
11: Subtitle (PGS), Spanish
12: Subtitle (PGS), Portuguese
13: Subtitle (PGS), Danish
14: Subtitle (PGS), Finnish
15: Subtitle (PGS), Norwegian
16: Subtitle (PGS), Swedish
[v02] Video gap description file detected, will be used for muxing...
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
Added fps value (24 /1.001) to MKV header.
Video track 2 contains 8144 frames.
eac3to processing took 12 seconds.
Done.


Now, I have 'ExtractedAudio.ac3' (automatically overlap-corrected by eac3to), 'ExtractedVideo.mkv' without overlap-correction and 'ExtractedVideo_gaps.mkv' with overlap-correction.

When I use MKVMERGE to MUX ExtractedVideo.mkv (no overlap corrected) & ExtractedAudio.ac3, the A/V sync at the start of the clip is fine, but at the end of it, the Audio is 250ms earlier.

When I use MKVMERGE to MUX ExtractedVideo_gaps.mkv (overlap corrected) & ExtractedAudio.AC3, the A/V sync at the star of the clip still fine, but at the end of it, the Audio is approx. 650mS late.

Any ideas?


Fernando

[laserfan]

I First extract the video and the audio tracks.


eac3to v3.24
command line: eac3to 00202.mpls 1) 2: "w:\test\ExtractedVideo.mkv" 3: "w:\test\ExtractedAudio.ac3"

So it detected Audio AND Video overlaps. It corrected the audio, and gave me the possibility to correct the video overlap also.

So I do it:


eac3to v3.24
command line: eac3to 00202.mpls 1) 2: "w:\test\ExtractedVideo_gaps.mkv"

Any ideas?
Did you try repeating (for the 2nd pass) exactly the same command line as initially?

[DivxBr]

Did you try repeating (for the 2nd pass) exactly the same command line as initially?

Yes, just done it. Result = Audio with a (approx.) +650mS delay at the end of the clip.

[TDiTP_]

eac3to 00202.mpls

1) 00202.mpls, 0:05:40
[44+45+46+47+48+49+50+51].m2ts
- Chapters, 8 chapters
- VC-1, 1080p24 /1.001 (16:9)
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz
- AC3, English, stereo, 48kHz

This is a short fragment. Maybe you can upload it to any fileshare?

[DivxBr]

This is a short fragment. Maybe you can upload it to any fileshare?

My connection isn't rock stable enough to upload almost 500MB, but I could try. You have any suggestion which upload-site accepts uploads so big?

[TDiTP_]

You have any suggestion which upload-site accepts uploads so big?

AFAIK mediafire.com

[DivxBr]

I had to separate the files in a multipart rar archieve to obey maximum size limit of the site, so there's the links:

http://www.mediafire.com/?j76bubvfnvnrbkz
http://www.mediafire.com/?n9jw82nek0xlauu
http://www.mediafire.com/?to0zlc8zau5bqu8
http://www.mediafire.com/?0hd1d7up9plb416
http://www.mediafire.com/?v0ixbf7se3zuvm8

[TDiTP_]

Thanks for sample :)
It's strange but i can't confirm your problem.

My two logs:

eac3to v3.24
command line: eac3to Segment\BDMV\PLAYLIST\00202.mpls 2: video.mkv 3: audio.ac3
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS)
8: Subtitle (PGS)
9: Subtitle (PGS)
10: Subtitle (PGS)
11: Subtitle (PGS)
12: Subtitle (PGS)
13: Subtitle (PGS)
14: Subtitle (PGS)
15: Subtitle (PGS)
16: Subtitle (PGS)
[a03] Extracting audio track number 3...
[a03] Removing AC3 dialog normalization...
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
[a03] Creating file "audio.ac3"...
[v02] Video overlaps for 7 frames at playtime 0:03:00. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:03:31. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:04:00. <WARNING>
[v02] The MKV file was created without making use of the gap/overlap information. <WARNING>
[v02] Please check whether audio is in sync. If it is in sync everything is fine. <WARNING>
[v02] Otherwise ask eac3to to repeat the muxing. It will then automatically make <WARNING>
[v02] use of the detailed gap/overlap information. <WARNING>
[a03] Audio overlaps for 120ms at playtime 0:00:40. <WARNING>
[a03] Audio overlaps for 95ms at playtime 0:01:37. <WARNING>
[a03] Audio overlaps for 52ms at playtime 0:01:55. <WARNING>
[a03] Audio overlaps for 96ms at playtime 0:02:27. <WARNING>
[a03] Audio overlaps for 119ms at playtime 0:03:00. <WARNING>
[a03] Audio overlaps for 105ms at playtime 0:03:31. <WARNING>
[a03] Audio overlaps for 91ms at playtime 0:04:00. <WARNING>
[a03] Starting 2nd pass...
[a03] Realizing (E-)AC3 gaps...
[a03] Creating file "audio.ac3"...
Added fps value (24 /1.001) to MKV header.
Video track 2 contains 8144 frames.
eac3to processing took 9 seconds.
Done.


eac3to v3.24
command line: eac3to Segment\BDMV\PLAYLIST\00202.mpls 2: video.mkv
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS)
8: Subtitle (PGS)
9: Subtitle (PGS)
10: Subtitle (PGS)
11: Subtitle (PGS)
12: Subtitle (PGS)
13: Subtitle (PGS)
14: Subtitle (PGS)
15: Subtitle (PGS)
16: Subtitle (PGS)
[v02] Video gap description file detected, will be used for muxing...
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
Added fps value (24 /1.001) to MKV header.
Video track 2 contains 8144 frames.
eac3to processing took 7 seconds.
Done.

I used MMG 4.6.0 to mux video and audio in MKV. Everything is
- ok in case of video with corrected overlaps
- not ok in case of video without corrected overlaps.
I used TsMuxeR 1.10.6 to mux video and audio to BDAV, all the same.

PS. I can hear the difference of a ~600ms.

PPS. With this command line

eac3to v3.24
command line: eac3to Segment\BDMV\PLAYLIST\00202.mpls 2: video2.vc1 3: audio2.ac3
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 10 subtitle tracks, 0:05:39, 7.702p
1: Chapters, 8 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
4: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
5: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
6: AC3 Surround, 2.0 channels, 192kbps, 48kHz, dialnorm: -27dB
7: Subtitle (PGS)
8: Subtitle (PGS)
9: Subtitle (PGS)
10: Subtitle (PGS)
11: Subtitle (PGS)
12: Subtitle (PGS)
13: Subtitle (PGS)
14: Subtitle (PGS)
15: Subtitle (PGS)
16: Subtitle (PGS)
[a03] Extracting audio track number 3...
[a03] Removing AC3 dialog normalization...
[v02] Extracting video track number 2...
[v02] Creating file "video2.vc1"...
[a03] Creating file "audio2.ac3"...
[v02] Video overlaps for 7 frames at playtime 0:03:00. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:03:31. <WARNING>
[v02] Video overlaps for 7 frames at playtime 0:04:00. <WARNING>
[a03] Audio overlaps for 120ms at playtime 0:00:40. <WARNING>
[a03] Audio overlaps for 95ms at playtime 0:01:37. <WARNING>
[a03] Audio overlaps for 52ms at playtime 0:01:55. <WARNING>
[a03] Audio overlaps for 96ms at playtime 0:02:27. <WARNING>
[a03] Audio overlaps for 119ms at playtime 0:03:00. <WARNING>
[a03] Audio overlaps for 105ms at playtime 0:03:31. <WARNING>
[a03] Audio overlaps for 91ms at playtime 0:04:00. <WARNING>
[a03] Starting 2nd pass...
[a03] Realizing (E-)AC3 gaps...
[a03] Creating file "audio2.ac3"...
Video track 2 contains 8144 frames.
eac3to processing took 11 seconds.
Done.

and then mux in MMG or TsMuxeR everything is ok with synchronization too.

[DivxBr]

I used MMG 4.6.0 to mux video and audio in MKV. Everything is
- ok in case of video with corrected overlaps
- not ok in case of video without corrected overlaps.
I used TsMuxeR 1.10.6 to mux video and audio to BDAV, all the same.

PS. I can hear the difference of a ~600ms.



Not sure if I understood....

I'm getting audio BEFORE video (250mS) on non-corrected overlaps, and getting audio AFTER video (600mS) on overlap-corrected MUX. Is it what you got?

I'd made another test using DGIndexNV demuxing Video&Audio, and the Result = No AV sync glitch.

I'll double check it later, but looks like the demuxed video is identical (to the not corrected for overlap), and demuxed audio is 8 AC3 frames bigger (+256mS) - which could explain the delay I found.

Making some calculations, if I remove 21 video frames (7 x 3 frames shown overlaped by eac3to), I get +256-(21/23.976) = -619mS (!!!) close enough to delay I get when I let the eac3to correct the video stamps.

I'll test the demux/remux VC-1 stream directly like you showed. Thanks!

I think I'll get some interesting info analysing timestamps file generated by dgindexnv too (at least near the 1st point eac3to finds the 7 overlaped video frames, the timecodes make a big jump, and at this point maybe eac3to is losing track of things).

[Carpo]

Would people suggest using this over TsMuxer, in the mood to do a bit of experimenting :-)

[hubblec4]

hi

i would install TMT5. is eac3to able to use the codecs from this version of TMT?

[PowerGamer]

I extracted PGS subtitles from BluRay into a .sup file using eac3to. I muxed that file into .mkv file using mkvmerge. When I try to extract subtitles from created .mkv file using eac3to I get a new .sup file much shorter than the original .sup file. SupRip fails to open that new .sup file. MPC-HC fails to display subtitles after muxing shorter .sup file into new .mkv file. So it seems that .sup file demuxed by eac3to from .mkv is invalid. I wonder if the author of eac3to can look into this issue?

P.S. Also there is another (very small) issue with eac3to. When listing tracks from .mkv eac3to shows no language for tracks that should be labelled as "English" (see http://www.bunkus.org/videotools/mkvtoolnix/faq.html#default_values).

[Weirdo]

Hi there madshi, is it possible for eac3to to not require Haali Media Splitter, but work with LAV instead?

(edit: Just saw Thunderbolt8 asked the same thing above...)

[Vasch the stampede]

eac3to v3.24
command line: eac3to "...\4.thd" "...\4.flac"
------------------------------------------------------------------------------
TrueHD, 5.1 channels, 48kHz
Decoding with libav/ffmpeg...
Encoding FLAC with libFlac...
Creating file "...\4.flac"...
[libav] Lossless check failed - expected 5d, calculated 63 <WARNING>
[libav] Lossless check failed - expected d, calculated 3 <WARNING>
[libav] Lossless check failed - expected fe, calculated b1 <WARNING>
[libav] Lossless check failed - expected 8b, calculated 5f <WARNING>
The original audio track has a constant bit depth of 16 bits.
Superfluous zero bytes detected, will be stripped in 2nd pass.
Starting 2nd pass...
Decoding FLAC...
Reducing depth from 24 to 16 bits...
Encoding FLAC with libFlac...
Creating file "...1\4.flac"...
The processed audio track has a constant bit depth of 16 bits.
Done. Should I be worried?

[Kurtnoise]

Hi there madshi, is it possible for eac3to to not require Haali Media Splitter, but work with LAV instead?

(edit: Just saw Thunderbolt8 asked the same thing above...)
eac3to uses haali's matroska muxer...LAV filters are just splitter/decoders.

An alternative would be to use the solveig mkv muxer or ffmpeg's one.

[Kurtnoise]

eac3to v3.24
command line: eac3to "...\4.thd" "...\4.flac"
------------------------------------------------------------------------------
TrueHD, 5.1 channels, 48kHz
Decoding with libav/ffmpeg...
Encoding FLAC with libFlac...
Creating file "...\4.flac"...
[libav] Lossless check failed - expected 5d, calculated 63 <WARNING>
[libav] Lossless check failed - expected d, calculated 3 <WARNING>
[libav] Lossless check failed - expected fe, calculated b1 <WARNING>
[libav] Lossless check failed - expected 8b, calculated 5f <WARNING>
The original audio track has a constant bit depth of 16 bits.
Superfluous zero bytes detected, will be stripped in 2nd pass.
Starting 2nd pass...
Decoding FLAC...
Reducing depth from 24 to 16 bits...
Encoding FLAC with libFlac...
Creating file "...1\4.flac"...
The processed audio track has a constant bit depth of 16 bits.
Done. Should I be worried?
I don't think so...

[deathlord]

Hi

It has been discussed, here and in other places, how to play back 5.1 and 7.1 sources in a common setup. There does not seem to be a satisfactory solution. I have opend a new thread where I would like to discuss this topic in some detail:

http://forum.doom9.org/showthread.php?t=160857

[nibus]

Is there a way to preserve artist info (ID3 tag) when converting from FLAC to AAC with eac3to?

Here is my command line:
eac3to.exe "file.flac" "file.m4a" -normalize -quality=0.45

I've included NeroAacTag.exe in the same folder but the output files are not tagged.

[kypec]

I don't think eac3to cares about any meta data in general. I'd suggest to try foobar2000 (http://www.foobar2000.org/) or latest LameXP (http://forum.doom9.org/showthread.php?p=1499797#post1499797) for such simple conversions.

[nibus]

I don't think eac3to cares about any meta data in general. I'd suggest to try foobar2000 (http://www.foobar2000.org/) or latest LameXP (http://forum.doom9.org/showthread.php?p=1499797#post1499797) for such simple conversions.

Ok that's what I thought. Thanks for the suggestions.

[mini-moose]

probably something that's been discussed a lot before but nothing seems to work for me.

Trying to convert an aac file to ac3 with eac3to and I always get "eac3to Getting "Nero Audio Decoder 2" instance failed"

I don't have nero7 installed and I'm not sure I want to install it either or whether it even works on win7 x64. I do have the NeroAAC decoder/encoder pack but am unsure how to make eac3to use it...

hope someone can help.

[TinTime]

eac3to will not decode aac without Nero's DirectShow decoder (it does work on Win7 64).

[mini-moose]

eac3to will not decode aac without Nero's DirectShow decoder (it does work on Win7 64).

so I must install nero7 to convert aac with eac3to or can I just get the directshow dll somehow? I'm not a big fan of Nero though I must say I dont think I tried 7 but most of the ones after 6 seemed like bloated crap

[Boulder]

You can use Nero Lite if you like.

[mini-moose]

You can use Nero Lite if you like.

installed Nero Micro 7. Seems to work now. thanks.

[Richard1485]

3: DTS Master Audio, Italian, 1.0 channels, 16 bits, 48kHz
(core: DTS, 1.0 channels, 16 bits, 384kbps, 48kHz)
[a03] Extracting audio track number 3...
[a03] Decoding with DirectShow (Sonic Audio Decoder)...
[a03] The WAV writer didn't receive the format information. <ERROR>
Aborted at file position 19418609664. <ERROR>

Any ideas what is happening here? This happens when I try to demux the mono to wav (or anything else for that matter).

[Moondust]

Lately I've come across some DTS-HDMA 6.1 tracks which I want to convert to FLAC. The FLAC file however doesn't play right. The length is wrong (2:14:58 instead of 1:57:50) and the sound is distorted in the way that the pitch seems to be way off. Lots of high frequencies and sharpness. Don't know how to describe it properly, but you can hear immediately that it's all wrong. Is this a known bug?

I tried to downmix it to 5.1, but the result is the same - shitty audio.

C:\Program Files (x86)\EAC3TO>eac3to d:\sources\infernalaffairsIII 1) 3:d:\audio
\InfernalAffairsIII.flac -down6
M2TS, 1 video track, 3 audio tracks, 3 subtitle tracks, 1:57:50, 24p
1: Chapters, 20 chapters
2: h264/AVC, 1080p24 (16:9)
3: DTS Master Audio, Chinese, 6.1 channels, 16 bits, 48kHz
(core: DTS-ES, 6.1 channels, 16 bits, 1509kbps, 48kHz)
4: AC3 EX, Chinese, 5.1 channels, 640kbps, 48kHz
5: AC3 EX, Chinese, 5.1 channels, 640kbps, 48kHz
6: Subtitle (PGS), Chinese
7: Subtitle (PGS), Chinese
8: Subtitle (PGS), English
v02 The video framerate is correct, but rather unusual.
a03 Extracting audio track number 3...
a03 Remapping channels...
a03 Decoding with ArcSoft DTS Decoder...
a03 Mixing surround channels...
a03 Encoding FLAC with libFlac...
a03 Creating file "d:\audio\InfernalAffairsIII.flac"...
a03 The original audio track has a constant bit depth of 16 bits.
a03 The processed audio track has a constant bit depth of 16 bits.
Video track 2 contains 169690 frames.
eac3to processing took 12 minutes, 26 seconds.
Done.

Is there a solution?

[nautilus7]

Use Sonic decoder instead of the Arcsoft one. It's known to cause problems with 6.1 DTS(-HD) tracks.

[TDiTP_]

Sonic can't decode DTS(-HD) 6.1 properly (blank BACK_CENTER always). You need ArcSoft v. 1.1.0.0.

[Moondust]

Sonic can't decode DTS(-HD) 6.1 properly (blank BACK_CENTER always). You need ArcSoft v. 1.1.0.0.
I just converted the DTS-HDMA 6.1 track from Infernal Affairs 3 to FLAC 6.1 using Sonic Audio Decoder version 4.3.0.169 and I checked all 6 channels with Audacity. None of them is empty. So it seems that using this version of Sonic Audio Decoder solves the problem I had with Arcsoft decoder version 1.1.0.7. Thanks for the helping hand nautilus7 and TDiTP_.

[TDiTP_]

Ouu, excuse me. Sonic can't decode properly DTS-HD 6.0, not 6.1.

[Thunderbolt8]

Ive lost track again, was there actually one thing which 1.1.0.7 or 1.1.0.8 could do 1.1.0.0 couldnt do, or was there no reason to switch to 1.1.0.7/8 at all, when having 1.1.0.0 working?

[nevcairiel]

1.1.0.7/1.1.0.8 cannot decode 6.0 properly (results in 5.0), and 1.1.0.7 cannot decode Mono.

1.1.0.0 can decode everything, as far as i am aware.

[Thunderbolt8]

alright, no sense to talk about .7 and .8 any further :p

[Greif]

Hey madshi, is there any appetite to adding the ability to mux streams together via either Matroska or MP4 (via mp4box)?

[forumator]

alright, no sense to talk about .7 and .8 any further :p

I wish that were true, I'm a newbie who just wants to rip all my BD audio to .flac, but there's conflicting information in this thread about the limitations of different Arcsoft versions. On page 545 TDiTP_ says ArcSoft 1.1.0.0 can decode only "normal setup" (-logdts: $84b). ArcSoft 1.1.0.7 and 1.1.0.8 decode all configuration But then on the next page we get some disagreement, xkodi says DTS Master Audio, 7.1 (strange setup) channels, 24 bits, 96kHz" and in fact 1.1.0.8 output is the same as 1.1.0.0 Another response from TDiTP_, WAV of Arcsoft 1.1.0.8, Sonic and Nero are identical byte-in-byte (i'm using HxD) to each other and not idential to Arcsoft 1.1.0.0. to which xkodi replies with his own analysis, in very short there is no version of Arcsoft decoder that can handle DTS-MA 7.1 "strange setup" files in bit-perfect way, i agree that there is small improvement regarding "L" and "R" channel with version 1.1.0.8, but still the decoded information is not correct in terms of bit-perfect decode. the only way to get bit-perfect decode from DTS-MA 7.1 "strange setup" files is to use Sonic decoder and convert it to correct 5.1 WAV file as i tried to explain here:

http://forum.doom9.org/showthread.ph...85#post1443085

Of course on the next page TDiTP_ comes back with another refutation in this post: http://forum.doom9.org/showthread.php?p=1472896#post1472896


Needless to say I'm confused about which version of Arcsoft I should be using, because if there's a problem with 1.1.0.0 I'd like to know about it before using it for all my encodes :/

[Snowknight26]

WAV, 2.0 channels, 0:08:35, 32 bits <float>, 2822kbps, 44.1kHz
Reading WAV...
Reducing depth from 64 to 24 bits...

I'm guessing that's merely a cosmetic issue.

[sub24ox7]

I have noticed that same 64bits to 24bits as well and figured it was a typo as well