Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > General > Audio encoding

Closed Thread
 
Thread Tools Search this Thread Display Modes
Old 1st November 2011, 06:06   #11381  |  Link
Atlantis
Registered User
 
Join Date: Feb 2002
Posts: 758
Here is a 1 sec audio file 48khz 5.1 32bit float.

eac3to can't read it.
Attached Files
File Type: rar audio 32bit float.rar (1,023 Bytes, 221 views)
Atlantis is offline  
Old 1st November 2011, 11:18   #11382  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by Atlantis View Post
Here is a 1 sec audio file 48khz 5.1 32bit float.
Your wav file have a wrong header:
Code:
File ........: D:\Internet\audio 32bit float.wav
Size ........:  1152068 bytes

---------------------------------------------- Header Info
ChunkID .....: RIFF
RiffLength ..:  1152060
Container ...:  WAVE
SubchunkID ..: fmt  (Length: 40)
AudioFormat .:  65534 (WAVE_FORMAT_EXTENSIBLE)
NumChannels .:  6
SampleRate ..:  48000
ByteRate ....:  1152000
BlockAlign ..:  24
BitsPerSample:  32
ValidBitsPS .:  0
MaskChannels :  63 (FL FR FC LF BL BR)
SubType .....:  3 (Float)
SubchunkID ..: data (Length: 1152000)
Offset data .:  68
Duration ....:  1 sec., (0h. 0m. 1s.)
------------------------------------------------- End Info
The field ValidBitsPerSample must be 32 instead 0.

You can hexedit the wav file and put at offset 38 the correct value 32 (hex 20) or fix the file with WavFix:

wavfix "audio 32bit float.wav" -m 0

Add the parameter -ignorelength if your wav file is greater than 4 GB.
After that eac3to can read the file.
__________________
BeHappy, AviSynth audio transcoder.

Last edited by tebasuna51; 1st November 2011 at 11:29.
tebasuna51 is offline  
Old 1st November 2011, 12:43   #11383  |  Link
Atlantis
Registered User
 
Join Date: Feb 2002
Posts: 758
What program did you use to see that information? I want to use it.

Also why the header is wrong? I used audition to get that. Also, if I output as 32bit float it doesn't work but if I output as 32bit integer, it works.

(using adobe audition to output. Works means eac3to can read the file)
Atlantis is offline  
Old 1st November 2011, 14:03   #11384  |  Link
dream88
Registered User
 
Join Date: Aug 2011
Posts: 3
Hi can someone please confirm for me what exactly the bug is with different versions of Arcsoft DTS Decoder, ?

Is it 1.1.0.5 or 1.1.0.0 for 6.1/6.0 audio and 1.1.0.8 for anything else ?
Some people claim that 1.1.0.0 is the best for 6.1/6.0 but others claim 1.1.0.5 is best because 1.1.0.0 lacks many bug fixes ?
I found the following pieces of information around doom9 forums but i'm still confused to which version to use to avoid any bug ?

- 1.1.0.0 can decode DTS(-HD) 6.1/6.0 but can't decode non-standard 7.1
- All versions above v1.1.0.0 do not accurately decode 6.1 DTS tracks with eac3to.
- 1.1.0.8 can't decode DTS(-HD) 6.1/6.0 but can decode non-standard 7.1
- Both decode DTS(-HD) 1.0 correctly, unlike 1.1.0.7.
lossy DTS
1) 1.1.0.0 and 1.1.0.1 always decode lossy DTS as 24 bit, and their decoding results differ from each other.
2) 1.1.0.5 and up decode in proper bitdepth.
3) Versions 1.1.0.5 and up decode lossy DTS identically.
4) 1.1.0.1 and 1.1.0.5 decode 24 bit lossy DTS identically.
5) Decoding of 16 bit lossy DTS was changed from 1.1.0.0 to 1.1.0.1 and from 1.1.0.1 to 1.1.0.5.
6) 1.1.0.7 and 1.1.0.8 decode 6.0 without back center channel.

The conclusion : for lossy DTS is 1.1.0.5. It decodes all configurations, in proper bitdepth, and decoding algorithm didn't change since this version (except 6.0 bug in .7 and .8).

Last edited by dream88; 1st November 2011 at 14:16.
dream88 is offline  
Old 1st November 2011, 20:56   #11385  |  Link
Anakunda
Registered User
 
Join Date: Jan 2010
Posts: 330
HI,

eac3to question:

does the -down16 switch have impact on quality or samplerate/bit depth on DTS -> AAC conversion? (providing the DTS is 24b/48kHz/6ch)
Anakunda is offline  
Old 1st November 2011, 20:58   #11386  |  Link
ramicio
Banned
 
Join Date: Mar 2004
Location: PA, US
Posts: 683
Quote:
Originally Posted by Asmodian View Post
Thanks, it makes a lot more sense now.

It turns out I am just getting 24 bit from the eac3to, there are no more bits to get. (-full gives 24 bit)

I am piping to qaac, I think 24 bit is all that exist for this mp2 source but qaac could take 32bit float if it did exist.

So there is no point.
Oh yeah, you will only get 64 bits when you do any sort of processing.
ramicio is offline  
Old 2nd November 2011, 00:02   #11387  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by dream88 View Post
Hi can someone please confirm for me what exactly the bug is with different versions of Arcsoft DTS Decoder, ?
- I don't found any bug using 1.1.0.0, also decode non-standard 7.1 without problems.

- lossy DTS don't have bitdepth, then you can't recover the original bitdepth source. The data in header only inform about the source bitdepth but this data is useless for 2 reasons:

1) Some encoders put always 24 bits in this field no mather the source was 16 or 24 bits.

2) The error recovering the lossy encode is greater than the precission difference betwen 16 and 24 bits, then it makes no sense decode to 16 or 24 bassed in source bitdeph, decode always to 24 bits and you obtain always the best approach.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 2nd November 2011, 18:31   #11388  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by Atlantis View Post
What program did you use to see that information? I want to use it.
http://forum.doom9.org/showthread.ph...30#post1522330

Quote:
Also why the header is wrong?
I don't know. I don't support commercial soft, ask to the vendor.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 2nd November 2011, 19:13   #11389  |  Link
Atlantis
Registered User
 
Join Date: Feb 2002
Posts: 758
tebasuna51, what you said is not correct.

I did another test. You said that it was because of wrong header.

Well as I said, 32 float was not recognized and 32 integer was recognized by eac3to. So I passed a 32 integer to your tool LeeAudBi and the header is the same incorrect way as you say but it works.

audio 32bit integer.wav
ChunkID .....: RIFF
RiffLength ..: 1152060
Container ...: WAVE
SubchunkID ..: fmt (Length: 40)
AudioFormat .: 65534 (WAVE_FORMAT_EXTENSIBLE)
NumChannels .: 6
SampleRate ..: 48000
ByteRate ....: 1152000
BlockAlign ..: 24
BitsPerSample: 32
ValidBitsPS .: 0
MaskChannels : 63 (FL FR FC LF BL BR)
SubType .....: 1 (Integer)
SubchunkID ..: data (Length: 1152000)
Offset data .: 68
Duration ....: 1 sec., (0h. 0m. 1s.)

I attached the file here. Test it with eac3to and it is recognized so that header problem you mentioned is not the problem. There must be another problem within eac3to.
Attached Files
File Type: rar audio 32bit integer.rar (1.0 KB, 82 views)
Atlantis is offline  
Old 3rd November 2011, 02:31   #11390  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Maybe eac3to don't check this field with int samples, because it make the test to know the real bitdepth of the source, with float sample must know the alignement.

But, how do you explain than changing only this field with a hexeditor the float file is read without problems?

You can read about this field here: http://msdn.microsoft.com/en-us/wind...dware/gg463006
Quote:
Details about wValidBitsPerSample
The field wValidBitsPerSample is used to explicitly indicate how many bits of precision are present in the signal. Most of the time this value will be equal to wBitsPerSample...
Audacity fill this field correctly, I don't know for what Audition don't make the same.
__________________
BeHappy, AviSynth audio transcoder.

Last edited by tebasuna51; 4th November 2011 at 11:15. Reason: explain int/float check
tebasuna51 is offline  
Old 4th November 2011, 15:52   #11391  |  Link
Traps
Registered User
 
Join Date: Feb 2011
Posts: 26
Source is 24bit TrueHD track, i want to convert it to 16bit flac or DTS but i'm getting this error
Code:
command line: eac3to.exe"  "E:\THD.thd" "E:\flac16.flac -down16"
------------------------------------------------------------------------------
TrueHD, 5.1 channels, 48kHz
This audio conversion is not supported.  <ERROR>
Anyone knows what am i doing wrong?
Traps is offline  
Old 4th November 2011, 15:54   #11392  |  Link
ramicio
Banned
 
Join Date: Mar 2004
Location: PA, US
Posts: 683
Try getting rid of the quotes. You don't need them when there are no spaces in file names or paths. The -down16 switch is part of the file name in your command.
ramicio is offline  
Old 4th November 2011, 15:55   #11393  |  Link
Traps
Registered User
 
Join Date: Feb 2011
Posts: 26
Yeah, i figured it out like 2sec after i posted here, thanks anyway!
Traps is offline  
Old 6th November 2011, 17:27   #11394  |  Link
pandv2
Registered User
 
Join Date: Sep 2011
Posts: 37
Quote:
Originally Posted by pandv2 View Post
If the silence duration to insert, is bigger than the insertion position, eac3to trows a error. So:

-edit=0:00:00.789,800 is a error
-edit=0:00:00.789,788 is not
If the source code is not available, can I hope this, and another bugs, get corrected in the future? Or Madish had left this project?.

Now I am using BeSplit to cut sections and eac3to to insert o delete (with delay) in the beginning. This works but BeSplit is not maintaned, and there are also some bugs.

And i post, also, a feature request: Extract a ac3 segment to a new file.
pandv2 is offline  
Old 7th November 2011, 19:44   #11395  |  Link
Boulder
Pig on the wing
 
Boulder's Avatar
 
Join Date: Mar 2002
Location: Finland
Posts: 5,718
How does normalization work when doing a multichannel to multichannel transcode? Is it even useful to apply or will it only mess the balance between the channels?
__________________
And if the band you're in starts playing different tunes
I'll see you on the dark side of the Moon...
Boulder is offline  
Old 7th November 2011, 22:29   #11396  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
-normalize preserve the balance between channels, all channels are amplified by the same value.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 8th November 2011, 19:19   #11397  |  Link
twazerty
AVCHDCoder developer
 
Join Date: Feb 2009
Location: Oss, Netherlands
Posts: 266
HDV 4 ch audio not detected (Mpeg 2 Layer 3)

In the professional world there is a problem regarding the HDV format with 4 audio channels. For example the Sony S270 which I have access to. This is a professional camera which is able to record a 4ch audio stream. When you want to import the footage of the camera to your computer you are unable to access the 4ch audio file inside the M2T file (MPEG-TS). And big names like Vegas, Avid, etc aren't able to see/extract the 4ch audio stream. Only the second stereo stream will be available. On the camera itself you are able to access all 4 audio channels. This is a problem that all camera's have that record in HDV with 4 channels and there isn't a solution from the manufactures. The only way is via SDI but it is very time consuming and HDD hungry (3GB/s).

Applications like MediaInfo and all video players can only access the second audio stream (or a backwards compatible stream but I am not sure of this). This is the mediainfo log:
Code:
General
ID                               : FF
Complete name                    : J:\HDVTest 4ch\00_0002_2010-11-07_214302.M2T
Format                           : MPEG-TS
File size                        : 138 MiB
Duration                         : 44s 160ms
Start time                       : UTC 2010-11-07 21:43:02
End time                         : UTC 2010-11-07 21:43:07
Overall bit rate                 : 26.2 Mbps
Maximum Overall bit rate         : 33.0 Mbps
Encoded date                     : UTC 2010-11-07 21:43:02

Video
ID                               : 2064 (0x810)
Menu ID                          : 100 (0x64)
Format                           : MPEG Video
Format version                   : Version 2
Format profile                   : Main@High 1440
Format settings, BVOP            : Yes
Format settings, Matrix          : Default
Duration                         : 44s 320ms
Bit rate mode                    : Constant
Bit rate                         : 24.0 Mbps
Nominal bit rate                 : 25.0 Mbps
Width                            : 1 440 pixels
Height                           : 1 080 pixels
Display aspect ratio             : 16:9
Frame rate                       : 25.000 fps
Standard                         : Component
Resolution                       : 8 bits
Colorimetry                      : 4:2:0
Scan type                        : Interlaced
Scan order                       : Top Field First
Bits/(Pixel*Frame)               : 0.617
Stream size                      : 127 MiB (92%)

Audio
ID                               : 2068 (0x814)
Menu ID                          : 100 (0x64)
Format                           : MPEG Audio
Format version                   : Version 1
Format profile                   : Layer 2
Duration                         : 44s 208ms
Bit rate mode                    : Constant
Bit rate                         : 384 Kbps
Channel(s)                       : 2 channels
Sampling rate                    : 48.0 KHz
Video delay                      : -232ms
Stream size                      : 2.02 MiB (1%)
But in fact this file contains a 4 channel audio stream. This is the output of eac3to v3.2.4:
Code:
eac3to v3.24
command line: "c:\Program Files (x86)\AVCHDCoder\Tools\eac3to\eac3to.exe"  "J:\HDVTest 4ch\00_0002_2010-11-07_214302.M2T" -demux
------------------------------------------------------------------------------
TS, 1 video track, 2 audio tracks, 0:00:44, 50i
1: MPEG2, 1440x1080 50i (16:9)
2: E-AC3, unknown parameters
3: MP2, 2.0 channels, 384kbps, 48kHz, -264ms
Bitstream parsing for track 2 failed.  <WARNING>
Demuxing this track may still produce correct results - or not.  <WARNING>
[a02] Extracting audio track number 2...
[v01] Extracting video track number 1...
[a03] Extracting audio track number 3...
[v01] Creating file "00_0002_2010-11-07_214302 - 1 - MPEG2, 1440x1080 50i.m2v"...
[a03] Applying MPx delay...
[a03] Creating file "00_0002_2010-11-07_214302 - 3 - MP2, 2.0 channels, 384kbps, 48kHz.mp2"...
Video track 1 contains 1107 frames.
eac3to processing took 1 second.
Done.
In this case a second audio stream will pop up. But I don't think it is an E-AC3 audiostream when we take a look at the HDV specifications that can be found on wikipedia: HDV Specifications. The most important part is this:
Code:
MPEG-1 Part 3 AL 2 Stereo (2-channel) at 384 kbit/s (192 kbit/s per channel);
optional MPEG-2 Part 3 AL 2 4-channel at 96 kbit/s per channel.
As you can see 4 channel audio is part of the HDV format. I am not sure if there are actually 2 audiostreams or only 1. There is no way I can check this because the camera itself only outputs the 4 ch and no 2 ch. But the computer will only play a 2 ch. But when you compare the MediaInfo audio information with the specs I expect that there are 2 audio stream just like eac3to detects. It would be very very nice if HDV support can be integrated into eac3to because many professional cameramen are looking for an solution.

Currently eac3to extracts the m2v and 2ch audio stream fine. Only the 4 channel stream isn't detected nor extracted. I uploaded 2 sample to my server for analysis:
http://tools.twanwintjes.nl/uploads/...-07_153401.M2T (49 seconds / 152MB) - Right Click --> Save As
http://tools.twanwintjes.nl/uploads/...-07_214302.M2T (44 seconds / 137MB) - Right Click --> Save As

In addition to the main problem the camera splits a recording into 4GB chunks. So sync issues are a problem according to the professionals.

Note: I do not own this professional camera. I professional camera guy I know came to me with this problem and he owns a camera. If you are willing to support HDV and you need more material like other resolutions/framerates/interlaced/progressive with only 2ch or combined with a 4ch audio track. We can deliver test footage. We are sure many people are looking for a solution. Companies like Sony do not offer a software solution, also this problem is still available on new camera's.

Additional info

Last edited by twazerty; 13th November 2011 at 23:02.
twazerty is offline  
Old 9th November 2011, 11:54   #11398  |  Link
Joniii
Registered User
 
Join Date: Aug 2007
Posts: 301
Why does eac3to and mkvextractgui2 output different file sizes?

h264 track demuxed from mkv:

eac3to 18 341 959 089
mkvextractgui2 18 341 843 800
__________________

Windows 10 Pro 2004 with Media Center, 2 x DVB-C with cable cards
MPC-HC (madVR, LAV Filters, XySubFilter)
Sony Bravia 85X950H/85XH95
Yamaha RX-V685
Yamaha Piano Black NS-777/555/C444/YST-SW315

Last edited by Joniii; 9th November 2011 at 11:56.
Joniii is offline  
Old 13th November 2011, 05:18   #11399  |  Link
Chouonsoku
Registered User
 
Join Date: Sep 2007
Posts: 52
I'm trying to convert a Dolby TrueHD 1.0 track to FLAC, and I keep getting the following error.

Quote:
eac3to v3.24
command line: eac3to B:\BDMV\STREAM\00001.m2ts 4:japanese.flac
------------------------------------------------------------------------------
M2TS, 1 video track, 3 audio tracks, 2 subtitle tracks, 0:24:13, 24p /1.001
1: h264/AVC, 1080p24 /1.001 (16:9)
2: TrueHD/AC3, English, 5.1 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 5.1 channels, 448kbps, 48kHz, dialnorm: -25dB)
3: TrueHD/AC3, English, 2.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 2.0 channels, 192kbps, 48kHz, dialnorm: -25dB)
4: TrueHD/AC3, Japanese, 1.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 1.0 channels, 96kbps, 48kHz, dialnorm: -25dB)
5: Subtitle (PGS), English
6: Subtitle (PGS), English
[a04] Extracting audio track number 4...
[a04] Extracting TrueHD stream...
[a04] Removing TrueHD dialog normalization...
[a04] Decoding with libav/ffmpeg...
[a04] [libav] Substream min channel cannot be greater than max channel. <WARNING>
[a04] The libav decoder reported error -1 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>
I did some searching and found that this is supposedly caused by an outdated version of ffmpeg included with eac3to. So I grabbed the latest version and ran the command line:

ffmpeg -i japanese.thd -ac 1 -acodec pcm_s24le -f wav japanese.wav

And got the following error:

Quote:
Input stream #0:0 frame changed from rate:48000 fmt:s32 ch:1 to rate:48000 fmt:s32 ch:2
Assertion ctx->channels == out->ch_count failed at /home/kyle/software/ffmpeg/source/ffmpeg-git/libswresample/audioconvert.c:66

This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information.
So now I'm not sure what to do. I don't remember having this problem with any other TrueHD 1.0 tracks in the past.
Chouonsoku is offline  
Old 13th November 2011, 10:24   #11400  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
The message is clear: this track is a mix of 1.0 and 2.0
Try with:
ffmpeg -i japanese.thd -ac 2 -acodec pcm_s24le -f wav japanese.wav
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Closed Thread

Tags
eac3to

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 09:11.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.