eac3to - audio conversion tool

[Mark_A_W]

Originally Posted by TinTime

Rename the .pcm files to .raw and sox should recognise them. As you said you'll have to specify bitrate, bit depth, etc.

Thanks.

Took me all that time to sort out the syntax...

sox -r 96000 -b 32 -c 1 -f lfe.raw lfe.wav (I think that's right..)


Job done

[nwg]

How do I work out the delay in the audio is eac3to doesn't give me anything? Is there any software that will? I tried Tsmuxer but that saids 0ms.

[alc0re]

If eac3to doesn't mention any unfixed audio delay, then there isnt any required.

[MuteyM]

Originally Posted by FoxRacR17

I am having a problem and I cannot figure it out. I'm trying to encode a dts master audio file to wma pro 5.1 but the audio keeps coming out all garbled when I play it on my 360. I have arcsoft total media theater 2.1.6.125 installed, and nero 7 installed. It is coming from a .m2ts files...

You could try my HDConcertRipper app, it's a GUI that uses eac3to to convert Blu-Ray (or HD DVD) discs to WMA Pro 5.1 (or FLAC). It won't work if all you have is the M2TS file though, it needs a full Blu-Ray BDMV directory structure.

[idbirch2]

Originally Posted by sidekick2

I took my burn of Bourne Identity that would not work and it plays *FINE* in my Dad's bdp-s350! This disk would not play at all in my PS3 (as described in my previous posts). I would have never guessed the PS3 wouldn't be able to decode DD+.

But it does, it worked fine for me with all 3 Bourne Films, the PS clearly said Dolby Digial Plus. Maybe it's because I authored to AVCHD (4GB splits to a FAT32 HDD) and not burned to disc.

[DrNein]

Is the conversion valid from THD (no core) to THD+AC3? The output audio constantly "drops out" during playback. The original THD is otherwise not recognized by tsMuxeR though and I understand incompatible with BD too.

[lchiu7]

Originally Posted by sidekick2

..

All we need now, is for Madshi to come up with a DD+ converter! Create a core from the hd-dvd DD+, then stick the extra bits on top of the core with some duct tape and spit. Ohh, and write a new firmware for the PS3 so I can play DD+. (minor detail I almost left out)

And what loss in AQ is there if the DD+ is converted to say DD at 640Kb? I guess only if the DD+ stream is 7.1 but then you could convert to DD-EX?

I am facing a similar situation. Finally got an AVR that has HDMI input ( in fact it decodes DD+ over bitstream) but nonoe of the playback devices I have, the A100, Toshiba A3 or PS3 can output DD+ as bitstream. The A3 outputs it as LPCM, the A100 not at all (perhaps as DD+ converted to PCM) and similarly for the PS3.

So I just convert the DD+ to DD and put up with the AQ reduction, if any

[peterjcat]

Originally Posted by lchiu7

And what loss in AQ is there if the DD+ is converted to say DD at 640Kb? I guess only if the DD+ stream is 7.1 but then you could convert to DD-EX?

So I just convert the DD+ to DD and put up with the AQ reduction, if any

DD+ on HD DVD is not the same as DD+ on Blu-ray -- there's no core to extract, so converting DD+ to anything requires a complete lossy recode so you'll technically lose some quality even if you're going from 640k DD+ to 640k DD. Since the majority of DD+ tracks are 1536k you'll lose even more quality going to DD. It doesn't make any difference whether the DD+ is 7.1 or 5.1 (if there are any 7.1 DD+ movies).

However, the libavcodec conversion to 640k DD is pretty good so the quality loss shouldn't be that noticeable.

[peterjcat]

Originally Posted by DrNein

Is the conversion valid from THD (no core) to THD+AC3? The output audio constantly "drops out" during playback. The original THD is otherwise not recognized by tsMuxeR though and I understand incompatible with BD too.

What are you using for playback? And are you using the most recent (.18 or later) version of tsMuxeR?

eac3to's THD to THD+AC3 has almost always worked fine for me (playing back via Popcorn Hour A-110), using old versions of tsMuxeR fixed with ts4Np, or the new version of tsMuxeR.

[lchiu7]

Originally Posted by peterjcat

DD+ on HD DVD is not the same as DD+ on Blu-ray -- there's no core to extract, so converting DD+ to anything requires a complete lossy recode so you'll technically lose some quality even if you're going from 640k DD+ to 640k DD. Since the majority of DD+ tracks are 1536k you'll lose even more quality going to DD. It doesn't make any difference whether the DD+ is 7.1 or 5.1 (if there are any 7.1 DD+ movies).

However, the libavcodec conversion to 640k DD is pretty good so the quality loss shouldn't be that noticeable.

What about DD+ to LPCM? Should be no loss at all then but a slight(!) increase in filesize (but you would have saved some on the video presumably?)

[edit]

OK - I can see some challenges with that. Just did the first evo file of a HD-DVD title and while the eac3 track was 978Mb the PCM was 4.6Gb! And I guess that would be almost double for the whole movie! As an aside I can see why folks are clamoring for multichannel FLAC support for the Popcorn Hour

[DrNein]

Originally Posted by peterjcat

What are you using for playback? And are you using the most recent (.18 or later) version of tsMuxeR?

eac3to's THD to THD+AC3 has almost always worked fine for me (playing back via Popcorn Hour A-110), using old versions of tsMuxeR fixed with ts4Np, or the new version of tsMuxeR.

Using latest tsMuxeR. Playback with MPC-HC and CyberLink Audio Decoder (though dodgy sound is similar with other decoders). The duration of the converted file is also wrong (too long and the counter runs fast). The THD cannot be played back in MPC-HC (although fine in source with PDVD) and MediaInfo reports no specs. But when converted to DTS is it okay though so eacto must be extracting it correctly and just not converting to THD+AC3 properly.

[Snowknight26]

eac3to seems to report the number of fields for interlaced content, not the number of frames.

eac3to v3.12
command line: eac3to.exe "Z:\temp\threekings.ts" "Z:\Encoding Tools\temp\threekings.source.mkv"
------------------------------------------------------------------------------
TS, 1 video track, 1 audio track, 1:54:49, 50i
1: h264/AVC, 1080i50 (16:9)
2: AC3 Surround, 2.0 channels, 384kbps, 48khz, dialnorm: -27dB, -2081ms
[v01] The video track contains the (probably incorrect) "full range" flag. <WARNING>
[v01] Extracting video track number 1...
[a02] Extracting audio track number 2...
[v01] Removing "full range" flag from video bitstream.
[a02] Removing AC3 dialog normalization...
[a02] Applying (E-)AC3 delay...
[v01] Muxing video to Matroska...
[a02] A remaining delay of -1ms could not be fixed.
[a02] Creating file "Z:\Encoding Tools\temp\threekings.source - 2 - AC3 Surround, 2.0 channels, 384kbps, 48khz.ac3"...
[a02] [1:48:15] The source file seems to be damaged (transport error indicated). <WARNING>
[v01] [1:48:14] The source file seems to be damaged (transport error indicated). <WARNING>
[a02] [1:48:15] The source file seems to be damaged (discontinuity). <WARNING>
[v01] [1:48:14] The source file seems to be damaged (discontinuity). <WARNING>
[a02] [1:48:15] The source file seems to be damaged (discontinuity). <WARNING>
[v01] [1:48:14] The source file seems to be damaged (discontinuity). <WARNING>
[a02] [1:48:15] The source file seems to be damaged (discontinuity). <WARNING>
[v01] [1:48:14] The source file seems to be damaged (discontinuity). <WARNING>
[a02] This track is not clean. <WARNING>
[a02] The last (E-)AC3 frame is incomplete and thus gets skipped. <WARNING>
[a02] Audio has a gap of 96ms at playtime 1:48:13. <WARNING>
[a02] Starting 2nd pass...
[a02] Realizing (E-)AC3 gaps...
[a02] Creating file "Z:\Encoding Tools\temp\threekings.source - 2 - AC3 Surround, 2.0 channels, 384kbps, 48khz.ac3"...
Added fps value to MKV header.
Video track 1 contains 344450 frames.
eac3to processing took 7 minutes, 38 seconds.
Done.

[madshi]

Originally Posted by GZZ

Did you get my file Madshi ?

Yes, but didn't have time to look into it yet...

Originally Posted by TinTime

I think I've found a very, very minor bug. If I run...

eac3to.exe 1.eac3 1.dts -core

...then it causes eac3to to crash.

Thanks, will check that out.

Originally Posted by jmonier

There appears to be a problem converting RAW/PCM 3.0 (Vicky Cristina Barcelona) to WAV. The center channel ends up being tagged as LFE. eac3to does correctly identify it (although PowerDVD identifies it as LPCM 3.0) but then doesn't seem to handle it as such.

Can't reproduce that problem. The WAV file I end up with is reported by eac3to as:

"WAV, 3.0 channels, 0:00:20, 24 bits, 3456kbps, 48khz"

Originally Posted by jmonier

Using -down2 also seems to have a similar problem and apparently throws away what it thinks is the LFE channel.

Downmixing does not properly support anything other than 5.1 or higher. Downmixing funny channel configurations like 3.0 or 2.1 is currently not supported. That's on my to do list, but it doesn't have a high priority...

Originally Posted by Thunderbolt8

what does the message mean which eac3to sometimes gives with some captures "this video track contains the (probably incorrect) 'full range' flag" and what does eac3to do then by default (or how can I see what eac3to does by default and if this is correct?)

I've already explained this a number of times, please do a search. Thanks...

Originally Posted by jamos

Ok some final thoughts on this issue. Tsmuxer is not the issue as it muxes fine using dts 7.1 and plays back fine.

That is not a proof at all. tsMuxeR might work just fine with DTS and still screw up LPCM muxing. Perfectly possible.

Originally Posted by jamos

The issue must exist in eac3to or Pcm3tsmu as I am getting this with multiple dts-ma 7.1 sources. You have to have a 7.1 reciever to test this as it looks correct in the file but on playback the center channel plays through the right front.

It could also be your receiver which doesn't handle 7.1 LPCM correctly. Or it could be whatever source device you're using. I don't really know where the bug is introduced in your situation. But from what I can see the bug is not caused by eac3to. So there's nothing I can do for you.

Originally Posted by zeropc

i never used the demux feature in eac3to, so i wonder if this would be the right cmd (example)...

eac3to "F:\MOVIE\BDMV\STREAM\00001.M2TS" "G:\FILES\" -demux

Originally Posted by yonta

eac3to demuxes to the current folder.
If you want your source file demuxed to a specific folder,
go to the folder and run command there.

Yes, yonta is right. Alternatively you can demux to a specific folder by using "eac3to source G:\FILES\movie.*".

Originally Posted by NanoBot

Yesterday I did my first tests in converting a DTS MA audio track to a multichannel flac, which works fine at all. Nevertheless, when the converting process was finished, I ended up with a flac file, which has a bitdepth of 24bit, while eac3to told me, that the bitdepth of the original track was only 16bit.

That is not supposed to happen. Are you sure you're using the latest eac3to version? If so, please post the eac3to log.

Originally Posted by SomeJoe

why is it that Dolby Digital Plus (.eac3) tracks from HD-DVD cannot be muxed to Blu-Ray and played on a Blu-Ray player? Is there that much of a fundamental difference between the .eac3 used on HD-DVD and what should/could be used on BD?

I have never tried and thus cannot answer any questions about what plays on a Blu-Ray standalone player or not.

What I can say is that Dolby Digital Plus is indeed fundamentally different on HD DVD compared to Blu-Ray. Well, at least Blu-Ray main audio. For Blu-Ray secondary audio (e.g. Director's comments) DD+ is identical to HD DVD.

Originally Posted by SomeJoe

Well, yes, that's what I've read. But it doesn't seem to make a lot of sense. Why would DD+ be so fundamentally different on HD-DVD compared to Blu-Ray?

Because DD+ is optional on Blu-Ray. So they used a core+extension logic for Blu-Ray main audio, where the core is simple Dolby Digital (not Plus). For HD DVD DD+ is mandatory, so they didn't need to provide a compatible core.

Originally Posted by idbirch2

ToNMT converts HD-DVD Dolby Digial Plus streams into BluRay compliant DDP streams.

No. It does not.

It doesn't change one bit in the DD+ stream. It just slightly changes the way the DD+ stream is stored in the m2ts container.

Originally Posted by sidekick2

All we need now, is for Madshi to come up with a DD+ converter!

That's not possible without reencoding. And there is no (affordable) DD+ encoder available. So this is not going to happen.

Originally Posted by anode

as far i have seen, in the output of eac3to's demuxed sup-files from blu-ray they contain the necessary PTS info but no DTS (decoder time stamp) information. I had some subtitles which do not play on my Samsung BD-P2500 (longer, more complex ones) (remuxed with tsMuxeR to BR-AVCHD). Why are the DTS fields always zero?
When demuxing the same track with tsMuxeR, DTS info is present.

I didn't think the DTS info would have any value. Well, seems I was wrong. I'll fix that...

Originally Posted by HanSolo00

Please add support for -24.975fps source framerate (the product of RePAL'd material.) It would be very useful to be able to "-24.975 -ChangeTo24.000" in order to revert audio back to original film speed, to match RePAL+AssumeFPS(24) video.

I'm sorry to say, but the SSRC resampler used by eac3to only supports very specific changes. I don't think it will do "24.975 -> 24.000".

Edit: Scratch that. I just tried, it works. "-24.975" will be supported in the next build.

Originally Posted by glood1

Problem remuxing a bluray

2: h264/AVC, 1080p24 (16:9)
[v02] The video framerate is correct, but rather unusual. <WARNING>

the framerate is not stable (speedup the video track is bad)
a few moment there is shifting frames (lost frames)
it s working with another movie whith 23.XXX

I don't think this is a bug in eac3to. I think it's your playback chain that somehow doesn't like 24.000 movies. So the best solution would probably be to slow the movie down to 23.976. The eac3to option "-slowdown" will do that for you. Afterwards you'll have a movie which would behave just like your average 23.976 Blu-Ray. You will need to apply "-slowdown" on both audio and video tracks.

Originally Posted by zeropc

say, can you add a "set progressive flag" option into eac3to for slowdown clips/movies?

that would make some pal to ntsc conversions proper.

here is why...

source is 1080i50 (film) and doesnt show any interlacing. slowing it down to 23.976fps works fine but the output file is not 1080/24p, it's 1080i48/1001.
this gives playback problems on the ps3 for example. it's all jerky, like frame-dropping.

A while ago I tried to do that, but I failed. The problem is that especially with h264 progressively encoded movies are very VERY different to interlaced encoded movies. I don't think patching 1080i48 to 1080p24 is possible without doing a full reencoding. At least not with h264. It might be possible with MPEG2 or VC-1. But you're talking about h264, right?

Originally Posted by FoxRacR17

I'm trying to encode a dts master audio file to wma pro 5.1 but the audio keeps coming out all garbled when I play it on my 360.

Try converting to something else, e.g. AC3, and check whether that plays fine. If so, the problem is most likely caused by whatever software you're using after eac3to is done.

Originally Posted by FoxRacR17

I'm running on Vista x64 sp1, and i'm using eac3to to change the dts-ma file into a single .wav (the resulting file is ALWAYS 4.50 gigs exactly, shouldnt it be bigger?)

Depends on the bitdepth and runtime. But 4.5GB is a perfectly reasonable size for a movie audio track.

Originally Posted by tebasuna51

But now, when a 2nd pass is needed, I get a WARNING:
The 2nd pass can't be started due to "stdout".

That's intended behavior. What would you like eac3to to do instead?

Originally Posted by tebasuna51

And with the 'normalize' parameter (or '2pass', working ok before 2.09) [...] the "stdout.pass1.wav" is created but nothing is sended by STDOUT to encoder.

That's a bug. Will fix that in the next build.

Originally Posted by tvjunky

Would it be much work to automatically create a second file containing only the forced captions while extracting this stream?

I guess it would be possible, but I don't know how difficult it would be. I'll put it on my to do list.

Originally Posted by TinTime

eac3to seems to occasionally use the wrong title number when it's specified in the command.

For example I get the following results getting all titles with Twelve Monkeys on HDDVD

Ouch, thanks.

Originally Posted by Thunderbolt8

seems like the "a remaining delay of Xms could not be fixed" message is missing when using -slowdown

When you "slowdown", eac3to has to do a full reencoding of the audio track. In this situation it applies delay on the decoded data. Which means that delay can always be perfectly fixed.

Originally Posted by Mark_A_W

Madshi, a little feature request:

I'm trying to convert a mono 32 bit IEEE Float, 96khz .pcm file generated from DRC (Digital Room Correction software), to a mono 32bit IEEE Float 96khz .wav file.

Eac3to works, but only up to 24 bit (if I put "-32", it gives the specify info error).

Hmmmm... I guess I could add support for that. But "-32" alone will not be enough. Because there is 32bit integer and 32bit float. So you will also have to specify "-float" or something like that...

Originally Posted by DrNein

Is the conversion valid from THD (no core) to THD+AC3? The output audio constantly "drops out" during playback. The original THD is otherwise not recognized by tsMuxeR though and I understand incompatible with BD too.

I'm not aware of any bug in thd+ac3 creation. It should work just fine, and you're the first one who reports a problem. My first guess would be that tsMuxeR does something wrong.

Originally Posted by DrNein

Using latest tsMuxeR. Playback with MPC-HC and CyberLink Audio Decoder (though dodgy sound is similar with other decoders). The duration of the converted file is also wrong (too long and the counter runs fast). The THD cannot be played back in MPC-HC (although fine in source with PDVD) and MediaInfo reports no specs. But when converted to DTS is it okay though so eacto must be extracting it correctly and just not converting to THD+AC3 properly.

Why do people always think that eac3to is at fault if anything goes wrong?

Originally Posted by Snowknight26

eac3to seems to report the number of fields for interlaced content, not the number of frames.

That's true. Hmmmmm... I guess I could count frames and fields separately.

[yesgrey]

Originally Posted by madshi

Why do people always think that eac3to is at fault if anything goes wrong?

Because eac3to's developer is the only one answering the questions and fixing the bugs when they exist?
Maybe if you stop eac3to development people will think that it's already perfect and the problems must be on any other part...

[tebasuna51]

Originally Posted by madshi

That's intended behavior. What would you like eac3to to do instead?

It's OK. Was only to document the next problem:

That's a bug. Will fix that in the next build.

Thanks.

[jmonier]

Originally Posted by madshi

Can't reproduce that problem. The WAV file I end up with is reported by eac3to as:

"WAV, 3.0 channels, 0:00:20, 24 bits, 3456kbps, 48khz"

Did you try checking it with something other than eac3to? I would think that it's possible that if eac3to converts it incorrectly it would also report it incorrectly.

In my case, AC3Filter reported it as 2.1 and played it that way as well. And when I fed the WAV directly to my receiver via SPDIF the center channel was not there.

[Kurtnoise]

@madshi: could you add Dirac detection from TS files please ?

> specs
> samples

thanks...

[DrNein]

Originally Posted by madshi

I'm not aware of any bug in thd+ac3 creation. It should work just fine, and you're the first one who reports a problem. My first guess would be that tsMuxeR does something wrong.

Why do people always think that eac3to is at fault if anything goes wrong?

I knew that was coming
But the problem appears to occur in the conversion before tsMuxeR is involved as the output plays wrong on its own. Then again, the extracted unconverted track cannot be played or identified outside of PDVD with the tools I have. In PDVD the track is identified as MLP rather than TrueHD as by eac3to. I understand they are related but do not recall seeing that before. It could be the track is an oddball but eac3to does convert it to DTS nicely and the bitrate was not higher anyway.

[Snowknight26]

eac3to can't properly decode this AC3 file to WAV.

[peterjcat]

Originally Posted by lchiu7

What about DD+ to LPCM? Should be no loss at all then but a slight(!) increase in filesize (but you would have saved some on the video presumably?)

[edit]

OK - I can see some challenges with that. Just did the first evo file of a HD-DVD title and while the eac3 track was 978Mb the PCM was 4.6Gb! And I guess that would be almost double for the whole movie! As an aside I can see why folks are clamoring for multichannel FLAC support for the Popcorn Hour

You got it. Most DD+ tracks on HD DVD are 1.536mbps, to unpack that to 24-bit 5.1-channel LPCM take 6.9mbps, so it's 4.5x the size to maintain the same quality. Even if you've got plenty of disc space, it's the principle of the thing!