How Does One Export Audio In The Same Quality As Source?

Hello. :slight_smile:

I was resynching an audio track that I want to add to a video. After finishing the resynching I tried to export the audio only using the same Sample Rate and Bitrate listed in the source but when the file finished exporting it was significantly smaller in size compared the original. I am not sure how to preserve the same quality and size as the original with the settings offered.

These are the specs of the source track:

ID : 3
Format : DTS
Format/Info : Digital Theater Systems
Mode : 16
Format settings, Endianness : Big
Codec ID : A_DTS
Bit rate mode : Constant
Bit rate : 1 509 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 1.60 GiB (20%)
Default : No
Forced : No

The highest I was able to export the file to was around 600 mb which as you can see is significantly less than the original file. What are the settings I should set in Shotcut to export the audio with the same quality as the original?

Thanks.

Tried one of the lossless export presets?

Which one though? Also, does Shotcut only export audio as stereo? I noticed that when I tried to export them they did not come back as 6 channels but only as stereo.

You choose.
Lossless = Lossless doesn’t it?

Sorry, I don’t know how SC handles multi channel audio.
Perhaps you need to use a commercial pro-grade NLE?

[quote=“Steve_Ledger, post:4, topic:961, full:true”]
You choose.
Lossless = Lossless doesn’t it?[/quote]

If there is more than one option than that means that they all don’t do the same thing.

Besides, the Lossless suggestion is meaningless here since I’ve already said I matched the Bit Rate and Sample Rate of the original track.

Does anyone know if Shotcut only exports audio as stereo?

I believe that DTS can fit in an AAC container (if Dolby digital can, the DTS competitor, can then why not DTS???)
just go to export, stock
select AAC

under Codec tab check the diaable video
under Audio Tab, go to Codec drop down arrow, and select DCA
maybe I’m wrong but its worth a shot

for example VLC media player implements DTS with libDCA (used to be called libDTS)
where libDCA is the "opensource “version” of DTS)

can we expect other “open source” softwares like Shotcut to implement proprietrary DTS? would there be
royalties to pay?
Or should we expect opensource softwares to support each other and go to
othe open souce software like lib DCA

I haven’t tried this
so you’ll have to be the scientific-spirited experimente rfor us all
I’m keeping my fingers crossed for you that DCA will do the trick for you

Sorry. I tried some experimenting after giving my (bad) advice to you.

I tried to bring in a 5 track Ogg audio file from Audacity but Shotcut just showed it as one audio track.

Then I tried a mono wav file import to Shotcut. Using shotcut I cut it up into 5 tracks.
I experimented everyway I could think of to get DCA export… no luck.

So I assume Shotcut does not yet implement DCA (that is opensource DTS)

And I asume that Shotcut also does not implement multitrack audio.

Can’t think of any other way to try. Anyone else have any ideas?

You should probably work audio in multrack audio software (like audacity) as 5 mono tracks and .
separate the 5 audio tracks into 5 files and bring each of the 5 files into Shotcut, then sync each separately
to the right video loctions in shotcut…
This is how I work with all audio files… las when I do voice over narration… I record audio, edit it, then, and only then bring the voice overs and background music to Shotcut to cut up and move around to where I want it on the video.

Should add that when I brought in the 5 track Ogg file and
when looked at Properties in Shotcut it did show it as 5 channel
but it just puts it all to display on one audio track,
instead of 5 tracks

so don’t think Shotcut is reducing the audios down to stereo as you questioned
it is still there as 5 channels
but just displayed as 1 track.

Also, for every audio only format I tried… I tried all I could see on the export menu…
the resulting files were all 2 channel only.

So, at least in exporting, Shotcut mixes down to 2 channel.

(Don’t know why then the Properties in Shotcut would show 5 channels if it mixes down on export.)

So you’ll have to forget about doing multitrack in Shotcut and as I suggested, bring multi audio files in and sync them

  • you can’t preview it except as mixed down stereo.
    Now way to get synced-to-video multitrack audio-only out of Shotcut.

Thanks a lot for your efforts, rayandlise. I really appreciate it! :slight_smile:

I didn’t know anything about having to pay DTS in order to encode audio that way. So there is no way to export audio multi-channel tracks like 5.1 and beyond unless money is paid to Dolby Digital and DTS?

Does Audacity have the same limitation?

DRM

Better ask that question on Audacity forums.

TO my knowledge both Shotcut and Audacity use FFmpeg to encode/decode many formats. And usually with audacity export you have to go to “external program” export and then usel ffmpeg directly (the audacity version of ffmpeg). Personally I have never been able to get DTS files (DCA ) out of Audacity… though maybe others have. And some say ffmpeg does not implement the full DTS specs.

But have got dolby digital like files out (open source approximations) of audacity.
so 5 channel dolby 5.1 is possible there.

With Shotcut I usually work multi channel dolby this way. Make a rough 5 track in audacity. Export it as 5 separate files rather than one 5 channel file. (audadcity can do it either way) Make sure you label the files so you know wihich file is for which speaker.

In Shotcut I make 5 audio tracks and import each Dolby 5.1 file (that is each separte file-channel) one by one and lay them on the audio tracks in Shotcut.

Then I sink the audio to video in Shotcut. Slice and move the audios around etc.

When done I export the finished audio tracks back out from Shotcut as separate files again.

That is, not only I turn off the video codec but I mute the 4 audio tracks I don’t want (big M button on left side of track in Shotcut) and then export to say AAC.
It’s a long process. Five times you, mute, 4 audio tracks and leae one to export.
So then I have 5 audio tracks files exported again.representing the dolby 5.1 channels.

I could reunite them in audacity. But since audacity uses FFmpeg I just go to FFmpeg directly.
Use ffmpeg to join the 5 audio files into a 5 track audio file.
Then use ffmpeg to join the audio back to the video only (video exported from Shotcut)

In this way thanks to ffmpeg you can end up with a video that has dolby 5.1 audio for 5 speaker systems.

It’s kind of long but its cheaper than usingf the commercial apps doing 5.1 audio with video.

Also some people tell me that dolby is better than DTS (DCA ) anyway. But a lot depends on the software
you use (not just on bit depth and frequency) and the hardware.

So you have to ignore some of the hype about high bit and frequency. (It’s like with cameras I’d rather have a 8 mpixel from a major manufacturer than a 20 mpixel phone camera any day, high numbers are just ad hype for phones)

Of course if FFmpeg does not do DTC fully then I’ll never really know anyway.if DTC is better than Dolby for audio… will have nothing to compare with as long as can’t get DTS files from ffmpeg.

Ray

If you want to preserve your channels, you can add channels=6 in the Other tab of the export dialog.

You can replace the 6 with the number of channels you actually have. I had to look into the source code to see how it was implemented. The default value, when not specified, is 2.

2 Likes

Thanks a lot for that, joel.ringuette. :slight_smile:

I tried it and while the file did it indeed come out with 6 channels intact and a bigger file size, it still came out far smaller than the original file. However, you did manage to solve a big mystery in that Shotcut can indeed preserve multi-channel audio tracks! Again, thanks for that!

The original file was encoded in DTS and as can be seen in my OP was 1.6 GB is size. I tried it with the 6 channel command and it came out to a bigger file of 946 MB. So I imagine that this is an issue now with the codec.

Can anyone recommend the appropriate audio codec on the Shotcut list to use for exporting if one wants to keep the quality and size of the original audio file?

Maybe it’s a 7.1 sound.
Or a DTS-HD High Resolution Audio and Shotcut doesn’t reach the max bitrate setting…

Did you analised the output files datas & compare it with the first post?

PS : But seriuosly, isn’t a 1Gb sound enought? Do you really feel the loss?

joel, thanks a heap. I wondered about that since openshot, also built on MLT, can export multi channel audio

Experimenting with your suggestion I brought in 6 audio files, 1 for each channel, to Shotcut, set them on separate audio tracks, split the audio, cut, paste etc.
to edit audio and sync to the video,
then I exported it as an aac file (as per your suggestion with channels=6) and sure enough
it did save as 6 channel audio. (rather than mixing down to stereo)
I tested it in audacity and I could work on all 6 channels tracks
But…
if I bring in a 6 channel audio file to Shotcut (without first breaking it to 6 separate files) it puts the whole file onto just one audio track in Shotcut… rather than 6 audio tracks (as in audacity)

It’s great to have 6 channel export possible (with the channels=6 setting) but do you know of any comparable way to import 6 channel audio onto 6 audio tracks in Shotcut?

Thanks for any suggestions.