how to add or remove silence at the beging and at end of each audio file using batch process

[pavlos22]

Hi all,

I am new to this forum. I have been using Audition 3 for the last couple of years and these days I have to coop with an extra difficult situation, which is; I have more than 2000 files that start and finish with a random length of silence, it varies 0 to 7 sec. My question is, how can I set the silent duration at the beginning and end of each file to a constant amount of 2 sec by using batch process?

Kind regards,

Pavlos

[SteveG_AudioMasters_ · Accepted solution]

pavlos22 wrote:

My question is, how can I set the silent duration at the beginning and end of each file to a constant amount of 2 sec by using batch process?

You are going to have to perform two processes to achieve this. The first is that you will have to remove all of the variable amounts of silence from the files, so that they all start from zero and stop at the end. You can do this automatically by using Edit>Delete Silence, but you will have to be careful about how you set it up.

After that, you need a script to add a fixed amount of silence at each end of each file, and run this as a batch process. The information about how to do that is in this thread, in the fourth post, and as far as I know, it's the only way to do it. There is more information in the rest of the thread too, explaining how this actually works.

[pavlos22]

Dear Steve,

Many thanks for you reply. You sussest to use Edit>Delete Silence in order to all start from zero and stop at the end, this procedure will allso trace silence between the audio! Am I missing something here?

[SteveG_AudioMasters_]

You may have to play around with the length settings to get this to work, but by and large how well it works overall is very dependent on the material you have in the files. With music, it generally works quite well, but I could certainly see a few potential problems with speech.

The other thing you may be able to do (and I haven't tried this, so I don't know how practical it would be) would be to have the delete silence operation in a batch file with some selection parameters. So you'd only operate it over the first n seconds of a file (you can select this quite easily) where n is slightly longer than the largest gap you have to detect. In fact, if this works, you could combine all of the operations necessary for both parts of your problem into one script, I think. I might have a look at the possibilities later if I get time...

[pavlos22]

I like your last suggestion, however, I have no clue how to write the sript. Can you help please?

[SteveG_AudioMasters_]

I'll look at the options over the weekend.

[pavlos22]

Thanks Steve!

[SteveG_AudioMasters_]

The progress report on this is that all of the bits work individually, but trying to get them to work correctly in a single sequence is a whole different matter!

One of the snags is that whilst it's relatively easy to select an absolute time at the beginning of a file (we discovered how to do this a while back), the same technique simply doesn't work at all at the end of it. The only solution to this that I'm aware of is to reverse the file as an integral part of the top and tail operation, which then gives you the opportunity to do the same operation at both ends - you just reverse the file again when you've finished.

Well, that's what should happen when you get the parameters right... but Audition's scripting language is, to say the least, rather arcane and completely undocumented (despite promises from the developers), so there's rather a lot of trial and error involved in getting this to work properly, I'm afraid. I've so far tried two completely different sequences, and one of them nearly does - but it's still screwing up the reversing when combined with another function in the sequence.

I shall persevere, because now I want to know what the hell's wrong with it!

[pavlos22]

>it's relatively easy to select an absolute time at the beginning of a file (we discovered how to do this a while back), the same technique simply doesn't work at all at the end of it

Indeed! I had a few trials last couple of days and noticed it.

Definitely we need help on Audition's scripting language from the people in Adobe. Why don't they produce a scripting language documentation?!   

[SteveG_AudioMasters_]

pavlos22 wrote:

Definitely we need help on Audition's scripting language from the people in Adobe. Why don't they produce a scripting language documentation?!   

Dunno - it's always been like that. Even though it's possible to reverse-engineer some of what's there, I always get the impression that there may be very simple ways of issuing commands to do things that you simply don't get by using the 'learn a sequence' method of doing things. For instance, the settings for selecting absolute time at the start of a file weren't discovered like that at all - ryclark only discovered this by experimenting. Same thing applies to my 'mode' table.

For a start, I really would like to know what all of the available commands are - along with lists for each one of the parameters you can vary. Somebody somewhere obviously has this information, so why shouldn't we have it as well? I mean, it's us that are trying to use it, for heaven's sake... what's the big secret?

[pavlos22]

I agree! I wonder if the Audition engineers read this Forum...