Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #228372 > unrolled thread

working with mp3 files

Back to article view | Back to linux.debian.user

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  working with mp3 files mick crane <mick.crane@gmail.com> - 2020-10-29 12:50 +0100
    Re: working with mp3 files Nicholas Geovanis <nickgeovanis@gmail.com> - 2020-10-29 13:00 +0100
    Re: working with mp3 files Celejar <celejar@gmail.com> - 2020-10-29 14:40 +0100
      Re: working with mp3 files Siard <shiems@mailbox.org> - 2020-10-29 19:00 +0100
        Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-10-29 20:00 +0100
          Re: working with mp3 files davidson <davidson@freevolt.org> - 2020-10-30 01:10 +0100
          Re: working with mp3 files Celejar <celejar@gmail.com> - 2020-10-30 01:20 +0100
    Re: working with mp3 files David Christensen <dpchrist@holgerdanske.com> - 2020-10-29 16:20 +0100
      Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-10-29 17:00 +0100
        Re: working with mp3 files ghe2001 <ghe2001@protonmail.com> - 2020-10-29 17:20 +0100
          Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-10-29 17:50 +0100
            Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-10-30 00:40 +0100
              Re: working with mp3 files <tomas@tuxteam.de> - 2020-10-31 09:50 +0100
          Re: working with mp3 files Andrei POPESCU <andreimpopescu@gmail.com> - 2020-11-02 09:40 +0100
            Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-11-02 11:20 +0100
        Re: working with mp3 files Nicolas George <george@nsup.org> - 2020-10-29 17:50 +0100
        Re: working with mp3 files Thomas Pircher <thp+debian@p5r.uk> - 2020-10-29 17:50 +0100
      Re: working with mp3 files ghe2001 <ghe2001@protonmail.com> - 2020-10-29 17:00 +0100
    Re: working with mp3 files Thomas Amm <debili@open-email.net> - 2020-10-31 11:40 +0100

#228372 — working with mp3 files

hello,
I am totally clueless about audio files.
Have for example librivox recordings of "1984" split into a dozen files.
Would like to combine them together into one file with
ffmpeg join_together "files" some_options) out_file
presumably each file has meta data that I only need once ?
Anybody point to a book or resource for a numpty to discover how it 
works?

mick


-- 
Key ID    4BFEBB31

[toc] | [next] | [standalone]

#228373

[Multipart message — attachments visible in raw view] — view raw

I recommend the python pydub package for exactly that. Tear them apart and
put them back together easily.
I wasn't a Eurythmics fan but their 1984 disc was outstanding :-) Sex Crime


On Thu, Oct 29, 2020, 6:43 AM mick crane <mick.crane@gmail.com> wrote:

> hello,
> I am totally clueless about audio files.
> Have for example librivox recordings of "1984" split into a dozen files.
> Would like to combine them together into one file with
> ffmpeg join_together "files" some_options) out_file
> presumably each file has meta data that I only need once ?
> Anybody point to a book or resource for a numpty to discover how it
> works?
>
> mick
>
>
> --
> Key ID    4BFEBB31
>
>

[toc] | [prev] | [next] | [standalone]

#228374

On Thu, 29 Oct 2020 11:42:35 +0000
mick crane <mick.crane@gmail.com> wrote:

> hello,
> I am totally clueless about audio files.
> Have for example librivox recordings of "1984" split into a dozen files.
> Would like to combine them together into one file with
> ffmpeg join_together "files" some_options) out_file
> presumably each file has meta data that I only need once ?
> Anybody point to a book or resource for a numpty to discover how it 
> works?

https://trac.ffmpeg.org/wiki/Concatenate

I concatenate mp4 video files (produced by my smartphone video
recording app, which starts a new file every 4GB) by putting them
(making sure they are named sequentially, e.g. "1.mp4", "2.mp4", etc.)
into an otherwise empty directory and running the following command
(taken from the above page):

ffmpeg -f concat -safe 0 -i <(for f in ./*.mp4; do echo "file '$PWD/$f'"; done) -c copy output.mp4

Celejar

[toc] | [prev] | [next] | [standalone]

#228386

Celejar:
> mick crane:
> > hello,
> > I am totally clueless about audio files.
> > Have for example librivox recordings of "1984" split into a dozen files.
> > Would like to combine them together into one file with
> > ffmpeg join_together "files" some_options) out_file
> > presumably each file has meta data that I only need once ?
> > Anybody point to a book or resource for a numpty to discover how it 
> > works?
> 
> https://trac.ffmpeg.org/wiki/Concatenate
> 
> I concatenate mp4 video files (produced by my smartphone video
> recording app, which starts a new file every 4GB) by putting them
> (making sure they are named sequentially, e.g. "1.mp4", "2.mp4", etc.)
> into an otherwise empty directory and running the following command
> (taken from the above page):
> 
> ffmpeg -f concat -safe 0 -i <(for f in ./*.mp4; do echo "file '$PWD/$f'"; done) -c copy output.mp4

The method that I know is slightly different. I once took it from
https://medium.com/abraia/basic-video-editing-for-social-media-with-ffmpeg-commands-1e873801659
but that site has apparently been simplified now and concatenation has been
removed.
I found that it works fine for mp3's as well.

Create join.txt file that contains the exact paths of the files that you
want to join. All files should be same format (same codec). The path name
of all files should be mentioned one by one like below.
  file /home/xx/myvideos/part1.mp4
  file /home/xx/myvideos/part2.mp4
  file /home/xx/myvideos/part3.mp4
  file /home/xx/myvideos/part4.mp4
Now, join all files using command:
$ ffmpeg -f concat -i join.txt -c copy output.mp4
If you get an error something like below;
  [concat @ 0x555fed174cc0] Unsafe file name '/path/to/mp4'
  join.txt: Operation not permitted
add "-safe 0":
$ ffmpeg -f concat -safe 0 -i join.txt -c copy output.mp4

[toc] | [prev] | [next] | [standalone]

#228390

[Multipart message — attachments visible in raw view] — view raw

Siard (12020-10-29):
> > ffmpeg -f concat -safe 0 -i <(for f in ./*.mp4; do echo "file '$PWD/$f'"; done) -c copy output.mp4
> 
> The method that I know is slightly different. I once took it from
> https://medium.com/abraia/basic-video-editing-for-social-media-with-ffmpeg-commands-1e873801659
> but that site has apparently been simplified now and concatenation has been
> removed.
> I found that it works fine for mp3's as well.
> 
> Create join.txt file that contains the exact paths of the files that you
> want to join. All files should be same format (same codec). The path name
> of all files should be mentioned one by one like below.
>   file /home/xx/myvideos/part1.mp4
>   file /home/xx/myvideos/part2.mp4
>   file /home/xx/myvideos/part3.mp4
>   file /home/xx/myvideos/part4.mp4
> Now, join all files using command:
> $ ffmpeg -f concat -i join.txt -c copy output.mp4
> If you get an error something like below;
>   [concat @ 0x555fed174cc0] Unsafe file name '/path/to/mp4'
>   join.txt: Operation not permitted
> add "-safe 0":
> $ ffmpeg -f concat -safe 0 -i join.txt -c copy output.mp4

That's exactly the same method, you are just using a file instead of a
shell pipe substitution.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228395

On Thu, 29 Oct 2020 Nicolas George wrote:
> Siard (12020-10-29):
>>> ffmpeg -f concat -safe 0 -i <(for f in ./*.mp4; do echo "file '$PWD/$f'"; done) -c copy output.mp4
>>
>> The method that I know is slightly different. I once took it from
>> https://medium.com/abraia/basic-video-editing-for-social-media-with-ffmpeg-commands-1e873801659
>> but that site has apparently been simplified now and concatenation has been
>> removed.
>> I found that it works fine for mp3's as well.
>>
>> Create join.txt file that contains the exact paths of the files that you
>> want to join. All files should be same format (same codec). The path name
>> of all files should be mentioned one by one like below.
>>   file /home/xx/myvideos/part1.mp4
>>   file /home/xx/myvideos/part2.mp4
>>   file /home/xx/myvideos/part3.mp4
>>   file /home/xx/myvideos/part4.mp4
>> Now, join all files using command:
>> $ ffmpeg -f concat -i join.txt -c copy output.mp4
>> If you get an error something like below;
>>   [concat @ 0x555fed174cc0] Unsafe file name '/path/to/mp4'
>>   join.txt: Operation not permitted
>> add "-safe 0":
>> $ ffmpeg -f concat -safe 0 -i join.txt -c copy output.mp4
>
> That's exactly the same method, you are just using a file instead of a
> shell pipe substitution.

Also known as "Process Substitution". See the subheading, with that
name, in bash(1) under EXPANSION section.

Briefly review the effects of doing s/buster/bullseye/g on some file:

$ diff /etc/apt/sources.list <(sed 's/buster/bullseye/g' /etc/apt/sources.list)

-- 
Ce qui est important est rarement urgent
et ce qui est urgent est rarement important
-- Dwight David Eisenhower

[toc] | [prev] | [next] | [standalone]

#228396

On Thu, 29 Oct 2020 19:58:48 +0100
Nicolas George <george@nsup.org> wrote:

> Siard (12020-10-29):
> > > ffmpeg -f concat -safe 0 -i <(for f in ./*.mp4; do echo "file '$PWD/$f'"; done) -c copy output.mp4
> > 
> > The method that I know is slightly different. I once took it from
> > https://medium.com/abraia/basic-video-editing-for-social-media-with-ffmpeg-commands-1e873801659
> > but that site has apparently been simplified now and concatenation has been
> > removed.
> > I found that it works fine for mp3's as well.
> > 
> > Create join.txt file that contains the exact paths of the files that you
> > want to join. All files should be same format (same codec). The path name
> > of all files should be mentioned one by one like below.
> >   file /home/xx/myvideos/part1.mp4
> >   file /home/xx/myvideos/part2.mp4
> >   file /home/xx/myvideos/part3.mp4
> >   file /home/xx/myvideos/part4.mp4
> > Now, join all files using command:
> > $ ffmpeg -f concat -i join.txt -c copy output.mp4
> > If you get an error something like below;
> >   [concat @ 0x555fed174cc0] Unsafe file name '/path/to/mp4'
> >   join.txt: Operation not permitted
> > add "-safe 0":
> > $ ffmpeg -f concat -safe 0 -i join.txt -c copy output.mp4
> 
> That's exactly the same method, you are just using a file instead of a
> shell pipe substitution.

And both variations are mentioned in the page I linked to:

https://trac.ffmpeg.org/wiki/Concatenate

Celejar

[toc] | [prev] | [next] | [standalone]

#228376

On 2020-10-29 04:42, mick crane wrote:
> hello,
> I am totally clueless about audio files.
> Have for example librivox recordings of "1984" split into a dozen files.
> Would like to combine them together into one file with
> ffmpeg join_together "files" some_options) out_file
> presumably each file has meta data that I only need once ?
> Anybody point to a book or resource for a numpty to discover how it works?

https://sourceforge.net/projects/audacity/

https://packages.debian.org/search?keywords=audacity&searchon=names&suite=all&section=all


David

[toc] | [prev] | [next] | [standalone]

#228377

[Multipart message — attachments visible in raw view] — view raw

ghe2001 (12020-10-29):
> I had a very similar problem a few years ago.  SoX will convert those
> mp3s to files you can edit (I convert to flac and edit with audacity),
> stick the edited files all together, and turn that big file back into
> am mp3.  Works real good (CLI, not GUI).

Doing this with a lossy format like MP3 will worsen the quality, it is
not a good idea.

MP3 is an elementary stream, with just a little noise at the beginning
or the end for metadata: concatenation of the streams works.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228379

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256



‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, October 29, 2020 9:54 AM, Nicolas George <george@nsup.org> wrote:

> ghe2001 (12020-10-29):
>
> > I had a very similar problem a few years ago. SoX will convert those
> > mp3s to files you can edit (I convert to flac and edit with audacity),
> > stick the edited files all together, and turn that big file back into
> > am mp3. Works real good (CLI, not GUI).
>
> Doing this with a lossy format like MP3 will worsen the quality, it is
> not a good idea.

Yeah, but he's starting with mp3, so I don't know of a better way to do it.

> MP3 is an elementary stream, with just a little noise at the beginning
> or the end for metadata: concatenation of the streams works.

How do you get rid of the metadata noise?  I've tried 'cating mp3s, and it didn't work very well at all.

Running through SoX makes fine sounding files.  (I've done all this with spoken word sources, though.  Like the ones OP has.)

--
Glenn English
-----BEGIN PGP SIGNATURE-----
Version: ProtonMail

wsBzBAEBCAAGBQJfmumlACEJEJ/XhjGCrIwyFiEELKJzD0JScCVjQA2Xn9eG
MYKsjDKlIQf+KCQtEnyEHU5m9NUwJRgM1ScsIYUVFeCOC3F+P8+NfFLUuqnF
9Mow3/ST8ox/rLOMWnmBTjEmt3ZZckuuWIOLbejXaOb+SAhWmzcO1zsHkqv8
KSD54rC+K/fdJ3ZF61bzmCYlKBuH9XbEWzoT88dWvjGFhC0E2WxHiRHr6j5Z
wNrBpiDs5ntk4Qvgq03q5HGXRIh5ddp9uoTYEw576YnwQ8sKnP3BRaPn36tt
jSUAKbiNGxxDYCdFCuoIkAWvPPy1vt7/LPIzU/fADQb8VnDAK06iLyA9N4hN
BEiy6Xjb54sDa9fa4UhZ8Ch2iH62uGKl2vxScXUrYWmeW5pYXeg4oA==
=OCNo
-----END PGP SIGNATURE-----

[toc] | [prev] | [next] | [standalone]

#228381

[Multipart message — attachments visible in raw view] — view raw

ghe2001 (12020-10-29):
> Yeah, but he's starting with mp3, so I don't know of a better way to do it.

The better way to do it is to avoid decoding then re-encoding.

> How do you get rid of the metadata noise?  I've tried 'cating mp3s, and it didn't work very well at all.

You can just cut the file with any binary tool.

Or you can use the concat demuxer in ffmpeg.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228393

[Multipart message — attachments visible in raw view] — view raw

mick crane (12020-10-29):
> Is it correct that the concat works with the file but the demux streams it
> first through audio device and records?

No, absolutely not. What on Earth could make you think that?!?

The concat demuxer works through each file, outputting packets in them
in order, while taking care of shifting the timestamps of the later
files according to the duration of the earlier. Nothing more.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228399

[Multipart message — attachments visible in raw view] — view raw

On Sat, Oct 31, 2020 at 12:32:24AM +0000, mick crane wrote:
> On 2020-10-29 23:30, Nicolas George wrote:
> >mick crane (12020-10-29):
> >>Is it correct that the concat works with the file but the demux
> >>streams it
> >>first through audio device and records?
> >
> >No, absolutely not. What on Earth could make you think that?!?
> >
> >The concat demuxer works through each file, outputting packets in them
> >in order, while taking care of shifting the timestamps of the later
> >files according to the duration of the earlier. Nothing more.
> 
> I may be a little intoxicated so there may be irregularities in my
> posting but is it possible you could help me out understanding how a
> static file gets to be a stream ( of packets ) ?

By squinting the right way :-)

Actually, in UNIX, every file is a stream of bytes. This view has
imposed itself so pervasively that only the old folks among us
remember those operating systems where files were, by default,
record oriented (I don't remember /that/ part fondly, mind you).

Back to topic, audio/video files have one common trait: their
content is time-related, and (at least when recording/playing
back), your machine has to access them (mostly) in time order.

Add to that that they tend to encapsulate several "things" (e.g.
left channel, right channel for audio; video and audio part for
films; perhaps several (human) languages; one track for each
orchestra instrument, yadda, yadda). At the beginning, for simple
cases, ad hoc solutions emerged: you alternate right and left
channel samples for a typical CD. With time, more complex requirements
emerged (you want "channels" with different bandwidth needs
"intertwined" in a way that their time bases stay in relation).

For that, you chop each channel in chunks of a convenient size.
You then shuffle your channels together (and call them streams,
just because). This shuffling together is called "multiplexing",
since why invent a new terminology, when engineers have got such
a handy one.

At the other end, you demultiplex.

That whole machinery is called the "container format" [1].
Whithin this, each stream has an own "codec", which typically
is (among other things) a lossy or a lossless (de-) compressor.

This works nicely whether the thing is on file or is coming in
via a socket. Yay abstraction layers!

Cheers
[1] https://en.wikipedia.org/wiki/Container_format
 - t

[toc] | [prev] | [next] | [standalone]

#228427

[Multipart message — attachments visible in raw view] — view raw

On Jo, 29 oct 20, 16:11:18, ghe2001 wrote:
> 
> 
> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
> On Thursday, October 29, 2020 9:54 AM, Nicolas George <george@nsup.org> wrote:
> 
> > ghe2001 (12020-10-29):
> >
> > > I had a very similar problem a few years ago. SoX will convert those
> > > mp3s to files you can edit (I convert to flac and edit with audacity),
> > > stick the edited files all together, and turn that big file back into
> > > am mp3. Works real good (CLI, not GUI).
> >
> > Doing this with a lossy format like MP3 will worsen the quality, it is
> > not a good idea.
> 
> Yeah, but he's starting with mp3, so I don't know of a better way to do it.

Import the mp3 directly in audacity ;)

Audacity might still be doing some conversion internally on 
import/export from/to mp3 (didn't check and my ears didn't notice any 
difference).

Audacity is meant for advanced editing, it's likely overkill if one just 
needs to join several files.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser

[toc] | [prev] | [next] | [standalone]

#228430

[Multipart message — attachments visible in raw view] — view raw

Andrei POPESCU (12020-11-02):
> > > Doing this with a lossy format like MP3 will worsen the quality, it is
> > > not a good idea.

> Import the mp3 directly in audacity ;)
> 
> Audacity might still be doing some conversion internally on 
> import/export from/to mp3

It will.

> 			    (didn't check and my ears didn't notice any 
> difference).

Then you are wasting bits.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228380

[Multipart message — attachments visible in raw view] — view raw

Thomas Pircher (12020-10-29):
> How about mp3cut from the poc-streamer package?

I do not know it.

I just remembered another possibility: mkvtoolnix can concatenate while
remuxing into Matroska.

Using a better container than the elementary stream is a good idea, if
no compatibility issue is involved.

Regards,

-- 
  Nicolas George

[toc] | [prev] | [next] | [standalone]

#228382

Nicolas George wrote:
> MP3 is an elementary stream, with just a little noise at the beginning
> or the end for metadata: concatenation of the streams works.

How about mp3cut from the poc-streamer package?

| mp3cut can split and concatenate MP3 files according to time slices
| given on the command line.

Thomas

[toc] | [prev] | [next] | [standalone]

#228378

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256



‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, October 29, 2020 9:14 AM, David Christensen <dpchrist@holgerdanske.com> wrote:

> On 2020-10-29 04:42, mick crane wrote:
>
> > hello,
> > I am totally clueless about audio files.
> > Have for example librivox recordings of "1984" split into a dozen files.
> > Would like to combine them together into one file with
> > ffmpeg join_together "files" some_options) out_file
> > presumably each file has meta data that I only need once ?
> > Anybody point to a book or resource for a numpty to discover how it works?
>
> https://sourceforge.net/projects/audacity/
>
> https://packages.debian.org/search?keywords=audacity&searchon=names&suite=all&section=all
>
> David

I had a very similar problem a few years ago.  SoX will convert those mp3s to files you can edit (I convert to flac and edit with audacity), stick the edited files all together, and turn that big file back into am mp3.  Works real good (CLI, not GUI).  Going through flac keeps the metadata.

Also, audacity is in the Debian repositories.  So is SoX.

--
Glenn English


-----BEGIN PGP SIGNATURE-----
Version: ProtonMail

wsBzBAEBCAAGBQJfmuTbACEJEJ/XhjGCrIwyFiEELKJzD0JScCVjQA2Xn9eG
MYKsjDIT+ggAjAOLBtocrz1kdYr4QBoi96N5dxR892vumXmTr+atrh82xnoZ
flYGo0J16Boh0dgAO3DoKov7fywXwVI9UvXmzfzUYtCvcbn6THj1Q3jLIfYD
QzC6YEPFSQBQwfP6u7YOD4Ho4KgnNvcYLBkO0+wzNZngqifBVThcoD50loYN
gU+OahKyNDc1R+SR1fk/3h+hl1kQm3qhclcJkVk1RCerMc4q0Ht2pPosgsjL
+9eKdJ2VB7F3mDasurt/EZ1N8NCwIXE5K+l/6nvh0X8wpSzfnnyTSqOp2BYR
QfuEFM5BLja8q6ZM66Fi3giaPc8tZhDgTTbnTymmFbwapKEHz8G7AA==
=yy0j
-----END PGP SIGNATURE-----

[toc] | [prev] | [next] | [standalone]

#228400

On Thu, 2020-10-29 at 11:42 +0000, mick crane wrote:
> hello,
> I am totally clueless about audio files.
> Have for example librivox recordings of "1984" split into a dozen
> files.
> Would like to combine them together into one file with
> ffmpeg join_together "files" some_options) out_file
> presumably each file has meta data that I only need once ?
> Anybody point to a book or resource for a numpty to discover how it 
> works?
> 
> mick
> 
> 
Hi, 

mpgjoin actually has a quite readable manpage. In case of mp3 all you
need is 

 mpgjoin -o JOINT_FILE_NAME -j FILE1 FILE2 (...) FILEN

I wouldn't worry about the metadata, you can create new ones with mp3tag
or graphically with EasyTag or similar.
Alernatively -if you don't want to hear "This is a Librivox
recording..." over and over, maybe- you migh just open the MP3s with
audacity, align them in extra tracks, cut what you don't want and export
the whole thing as tagged MP3. 

Greetings, 
tom

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web