Using chains to insert silence at start and end of file

[sik0fewl]

Hi,

I'd like to use chains to truncate silence at the end of each track and then generate a specified amount of silence at the start and end of each tracks.

I figured this would be easy to do, since it's easy to do by hand. However, chains only seem to allow "Effects" to be used and not other Audacity commands, so I can run "Truncate silence" but "Generate -> Silence" can't be. If I overlooked something here, please let me know.

So after that I figure my best bet is using Nyquist in the chain. However, when I try to do this, Audacity crashes. I'm not sure if I'm doing something wrong (most likely) or there's just a bug in Audacity.

Here's what I tried to do to generate silence at the start and end of the track. This is on a mono track.

Code: Select all

(seq (const 0 5) (cue s) (const 0 5))

That call crashes Audacity with this debug info:

Code: Select all

[ gc: total 16640, 1656 free; samples 1KB, 0KB free ]
error: bad argument type - NIL
Segmentation fault

My system:

Code: Select all

$ uname -a
Linux makalu 2.6.28-16-generic #55-Ubuntu SMP Tue Oct 20 19:48:24 UTC 2009 i686 GNU/Linux
$ audacity -version
Audacity v1.3.7 (Unicode)

[Edit: I'm running Jaunty]

Any help that can be offered would be greatly appreciated!

--
Thanks,
Ryan

[steve]

Here's what I tried to do to generate silence at the start and end of the track.

Unfortunately it's not quite that simple - putting silence before "s" in that way won't work because of stuff to do with *warp* that I don't understand

It should not crash Audacity though, but I see you have Audacity 1.3.7 which has some bugs in Nyquist (now fixed in Audacity 1.3.9).

I'm not sure that this is the best way to do it, but this will add 5 seconds of silence to the beginning and end of the selection:

Code: Select all

(feedback-delay (seq (cue s)(cue (stretch-abs 1 (s-rest 10)))) 5 0)

[steve]

A couple of ways to do it with mono or stereo tracks:

Code: Select all

(stretch-abs 1.0
(if (arrayp s)
   (vector
      (sim (at 0 (cue (s-rest 5)))(at 5 (seq (cue (aref s 0))(cue (s-rest 5)))))
      (sim (at 0 (cue (s-rest 5)))(at 5 (seq (cue (aref s 1))(cue (s-rest 5))))))
   (sim (at 0 (cue (s-rest 5)))(at 5 (seq (cue s)(cue (s-rest 5)))))))

Code: Select all

(defun space (s-in time)
(stretch-abs 1.0
   (sim
      (at 0  (s-rest time))
      (at time (seq (cue s-in)(cue  (s-rest time)))))))

(multichan-expand #'space s 5)

[sik0fewl]

Thanks a lot, Steve... I'm pretty sure I would've never figured that out .

Unfortunately I just noticed that the Nyquist prompt isn't even available in "Edit chains".

What would you recommend doing in order to run these commands (Truncate silence and Nyquist script) in batch? Audacity is starting to seem pretty inflexible for this purpose.

--
Thanks again,
Ryan

[steve]

Unfortunately I just noticed that the Nyquist prompt isn't even available in "Edit chains".

Yes, shame isn't it

Batch processing is still pretty rudimentary in Audacity.

There are a few related "Feature Requests" listed here: http://wiki.audacityteam.org/index.php?title=Automation

[sik0fewl]

Batch processing is still pretty rudimentary in Audacity.

Yes, I'm starting to discover that . I'll guess I'll have to play around and see what I can do.

I'm starting to think the easiest way might be to write a new built-in effect .

Thanks again for your help.

[steve]

There was a long discussion about this a while ago - unfortunately the main discussion is no longer on the forum, but there's a partial summary here: http://forum.audacityteam.org/viewtopic ... 03&p=52466

[somenick]

Was this ever solved? Did you find another solution not using Audacity?

[steve]

Was this ever solved? Did you find another solution not using Audacity?

Adding this feature to Audacity is quite a major project and will not be happening before Audacity 2.0 comes out.

If you are comfortable with command line programs and batch scripting it is possible to accomplish the task using SoX http://sox.sourceforge.net/
There is a bit of an introduction to scripting with SoX in this topic: http://forum.audacityteam.org/viewtopic ... 18&t=50426

[bgravato]

Was this ever solved? Did you find another solution not using Audacity?

Adding this feature to Audacity is quite a major project and will not be happening before Audacity 2.0 comes out.

If you are comfortable with command line programs and batch scripting it is possible to accomplish the task using SoX http://sox.sourceforge.net/
There is a bit of an introduction to scripting with SoX in this topic: http://forum.audacityteam.org/viewtopic ... 18&t=50426

Though it is implicit, I'd like to make it explicit that obviously sox can't process audacity project files... So if you have all the shows saved as audacity projects only it won't help much unless you export all of them as WAV, FLAC or similar...