Cut or trim blank space out of wave file

[TGB]

Bob,

There was an option in Saw Pro that was very helpful and I was wondering if you could maybe hook it up in SS?

It was and option of trimming out blank spaces of a wave file. It was quit helpful and I use to use it all the time.

SS is the undisputed BEST and I've never regretted the purchase.

Kudos to your engineering Genius!

[Bob L]

The option is built-in on each track in the Track Properties menu... Right-Click on any track label in the MT.

It's called Remove Silence and there are many options... read the details of the feature in the helpfile section... Using The MultiTrack/Operation Normal Mode/MultiTrack Editing And Control - Track Number / Label Display.

The Levelizer plug also has a version of this feature.

This feature is designed in SS to be used with a wav file on the MT... the option cuts the wav file into regions with silence in between.

If you are talking about cutting an actual soundfile down in size stripping out silent spaces and joining the ends together back to back... you would do this on the MT in SS... start with a wav file on the track... simply mark an area you want to trim (B and E keys or drag in the timeline ruler)... then press Shift-Delete to delete the marked section and splice the ends together... go down the timeline doing all the cuts you want.... then use the buildmix option on the Process/Mixdown menu to write a new wav file with all the cuts... this will not destroy the original file and will do the whole process in one pass... rather than re-writing the file with each cut in the SoundFile View.

Bob L

[TGB]

What I'm talking about has nothing to do with MT operation at all. What I was talking about is soley in the wave file by itself. Cutting the wave file down in size.
In Saw 32 it was very easy. You didn't have to build to new soundfile or anything, just mark the b & e of the wave file and go to I think it was cut soundfile. This is great for 2 track mixes that have to much dead roll on the head or to long at the end.

[Bob L]

Well.. that is what I was talking about then in the second part of my reply.

The operation in SAW32 was pretty much the same... it had to rebuild the file for each cut... so it was doing the same thing in the background... and the point I was trying to make is that it had to do that for each cut you made... so if you trimmed a few separate spots out of the middle of the file, it had to rebuild each time over and over again.

The MT method I suggested allows you to not only make dozens or hundreds of cuts at once with only one rebuild, but you can also eq, or otherwise process the file and adjust the levels etc... all in the same one rebuild... much more powerful and generally much more efficient.

Set up one big track on the MT and it now looks like th SF view.... do the same work there.

Bob L

[lownoise]

What I'm talking about has nothing to do with MT operation at all. What I was talking about is soley in the wave file by itself. Cutting the wave file down in size.
In Saw 32 it was very easy. You didn't have to build to new soundfile or anything, just mark the b & e of the wave file and go to I think it was cut soundfile. This is great for 2 track mixes that have to much dead roll on the head or to long at the end.
I also find occasions where it is desirable to trim the head or tail of the WAV file, and I still make use of SAW32 to do it. Another option I like in SAW32 is the ability to Clear Regions (L / R or BOTH) and Mix or Paste L to R or R to L.

Of course, these are things that can be accomplished easily with SS MT/Automation, but the point is that sometimes a final correction or glitch removal is required in a master and it's nice not to have to remix at that point. (Just do the surgery on a copy!)

As to the head/tail issue, there is a nice utility WavTrim that will add or remove silence, normalize, and some other things. WavTrim.zip (http://www.lownoiserecords.com/downloads/_freeware/WavTrim.zip) Downloadable at http://www.lownoiserecords.com/downloads/

73, Douglas

[Bob L]

A remix is not reprocessing... simply place the wav file on a clean MT in a clean edl and trim... the resulting buildmix is then an identical copy, unless you add levels and processing... exactly the same as trimming in the SAW32 SoundFile view which is, as I said, also rebuilding the entire wav file from the beginning anyway.

The only time it does not have to rebuild is if you are trimming only the very end of the file... then it can simply change the EOF marker in the file... but if you trim anything from the front or middle... it rebuilds the file completely.

This operation is definitely much more powerful on a MT track, if you let go of the old Soundfile idea. :)

Bob L

[lownoise]

A remix is not reprocessing... simply place the wav file on a clean MT in a clean edl and trim... the resulting buildmix is then an identical copy, unless you add levels and processing... exactly the same as trimming in the SAW32 SoundFile view which is, as I said, also rebuilding the entire wav file from the beginning anyway.

The only time it does not have to rebuild is if you are trimming only the very end of the file... then it can simply change the EOF marker in the file... but if you trim anything from the front or middle... it rebuilds the file completely.

This operation is definitely much more powerful on a MT track, if you let go of the old Soundfile idea. :)

Bob L

Right, Bob. The times that we have resorted to the old method have been when for whatever reason the buildmix process itself creates the glitch, for example a nasty transient at the head of the output file.

Douglas

[Bob L]

The buildmix process should never create a glitch on its own... something else is going on with your session... probably a nasty vst plugin that is retaining data from the last playback.

SAWStudio will not create a glitch liike that randomly. The buildmix is one of the most powerful features of the program and one that you should consider feeling very comfortable with... including learning how to manipulate all its features and output file creation options. :)

Bob L

[lownoise]

The buildmix process should never create a glitch on its own... something else is going on with your session... probably a nasty vst plugin that is retaining data from the last playback.

SAWStudio will not create a glitch liike that randomly. The buildmix is one of the most powerful features of the program and one that you should consider feeling very comfortable with... including learning how to manipulate all its features and output file creation options. :)

Bob L

I'm sure that's what it is, exactly. But as a practical matter, especially when time is tight, it's still nice to have the aforementioned Soundfile Edit options in the arsenal!

Douglas

[Bob L]

Well, for those special occasions... its nice and easy to pop into SAW32 or SAWPro or SoundForge or whatever is your favorite. :)

Again I say, for the problem you describe, the SAW32 method is building the mix again exactly as SAWStudio would, but in the background... a nice difference in SS is that you can cut the front and use a simple soiftedge to ramp up from zero nicely so there is no noticable signal jump in headphones at the beginning of the file... in SAW32, you are stuck with an instant jump in signal after you cut behind the glitch.

Bob L

[MikeDee]

Hi, Bob -

For my clarification, suppose I do the following:

1. Build a mix in my project .edl
2. Save and close my .edl
3. Create a region of the entire (newly mixed) SoundFile and add to the MT, Track 1
(Or: Open that SF and add to MT)
4. Build a mix of that one region (no level changes or processing at all)
5. Close the .edl (not necessary to save it)
6. Repeat steps 3 through 5 a total of ten times

Result: We have the original BuildMixed file (from step 1), plus ten additional BuildMixed files, each a build of the immediately prior mix.

Question: Will all eleven files sound identical? Better yet, will all eleven files BE identical?

Theoretically, I understand that they should sound, should be identical...the "numbers" should be identical all eleven times. Will they in practice? I'd like to think so, perhaps even be so bold as to expect so. :)

We know what would happen in the world of analog tape...hiss upon hiss...compounding with each new "mix" of the prior "mix." I mean, just a simple bounce of a few tracks can yield audible tape hiss.

But in the digital world, there is, of course, no tape hiss. Yet I am wondering if anything, anything at all, no matter how slight, happens to each new file during each subsequent BuildMix...a difference of "just one number." If so, there would most likely be an audible difference between the first and eleventh file. Based on what you've mentioned throughout this thread, however -- and, again, theoretically (and hopefully in practice) -- I would tend to think not. The end result would (and should) be that all eleven files will sound and be identical.

Drumagog roll, please, Bob....... :D

[studio-c]

I'm counting on it being identical. And if there is a random number here or there, I don't think it's going to ruin your reputation as an engineer. That's WAAAAAY down there on the list of how someone could botch up a mix. There are a lot of things you could be spending your time agonizing over. I don't think this is one that needs to take up your brainspace.

And on the other points. Embrace the idea of non-destructive editing. Worst case, it's really nice to have a backup:
mysong_uncut.wav
mysong_trimmed.wav

I usually have several versions of mixes, including
mysong_original.wav
mysong_normalized.wav
mysong_CLWH.wav (compressed/limited/wavehammered for certain streaming uses. It's not pretty...)
mysong_64kbps_streaming.mp3
mysong_320kbps_streaming.mp3
etc.

I haven't worried about destructive editing for years.
For that, you can use Soundforge, etc. But if you do, you can't go back and use the same file in SS because your event times will all be screwed up. SS is best used for its nondestructive and pristine multitrack and mixbuilding functions.

Scott

[Bob L]

We had this discussion before somewhere in other threads... yes... all files will in fact be perfect.... you can test for yourself... put up the eleventh copy against the original in the MT... they should play at exact double level on the out meter... reverse the phase on one of the tracks and you should get 100% cancellation... zero output.

My summing engine should not introduce a random mistaken number value anywhere.

Bob L

[MikeDee]

On the ball with you there, Scott. I don't believe I have ever done destructive editing...ever. (Well, DAW-wise, that is...I mean, I could think of other ways to destroy a mix. ;) ) I went straight from the razor blade & block right to SAWPlus32 ==> SAWStudio.

My post is hypothetical and, as you said, not one over which I will agonize. I save "incremented" versions as well, such as "premix" and "normalized," etc. My question deals with one BuildMixed file ("whatever" version) BuildMixed again...as is...then taking that new BuildMix and BuildMixing it again...also as is...continuing to do this a total of ten times.

The end result should be that all eleven files should sound and be identical, since no processing or changing of levels is taking place during any of the BuildMix operations. I simply want to determine whether any of these unadulterated "mixes" are different in any way, no matter how slight.

My guess is that if Bob were to "check the numbers" for each file, they would be identical. So...I would think that all files would sound identical as well. This is the nature of my hypothetical post.

In fact, to take this a step further...the files DEFINTELY should be identical. If not, who could ever trust a simple file backup (copy/paste), right? Guess I answered my own question, perhaps. :D

Anyway, in the real "bread and butter" world, I'm definitely on the same page with you...non-destructive editing (SS), version control, and backups all the way. :)

Best personal regards,

[MikeDee]

Excellent, Bob. Just the answer for which I was hoping (and expecting). :)

Best personal regards,

[studio-c]

So you're a splicing block kinda guy too? Kewl. Makes me appreciate SS all the more :)
Glad to get the verification from Bob. I thought if anyone was going to keep the math pure, it would be him.
Wasn't sure if the thread topic was holding you up or if it was just a curiosity. It's easy to get sidetracked by 'what if' and 'why can't we', when there's a way to do just about everything in SS. And when not, there are other apps for everything else. Glad to hear you weren't losing sleep over trimming blank space :)

Cheers,
Scott

[Bob L]

You don't need to wait for my answer... try it yourself... the cancellation can only happen if all values are exact... any phasing problems or floating point math roundoff errors will add up to leakage when you reverse the phase between the different builds... SAWStudio's mix engine will not cause these problems... there was another discussion that got into a debate that all DAWs will do this... and I said then also... I sure hope so... but... you may be surprised. :)

This claim should be matter of fact... this is not something special about SS... although it may be something you want to confirm with your DAW.

Bob L

[studio-c]

I've done it.
I just love hearing you say it :)
Uncle Bob makes us all feel safe... lol...

But seriously, the fact that you have paid such painstaking attention to detail allows me to do what I do and not feel like I have to prove it. Things like headroom, I know that you work in the real world and know how unexpected loud bits just HAPPEN, and you've allowed for it. There's no slop in your design and it's very much appreciated for us folks that wanna just jump in and drive the thing rather than know how to adjust our own timing belts.

Again, thanks so much.

Scott

[MMP]

I have used modulation plugins that don't start at the same phase on each build. Two mixes in this rare case would not be identical. So, like most every rule, there are exceptions...but, this is not a result of anything in Bob's engine.

Regards,

MM

[Bob L]

But that would still not apply to the situation that was proposed here... to simply take an existing mix and rebuild it over and over as in trimming and cutting in the MT... rather than in the SoundFile View.

The interesting thing is that doing this type of trim work in the MT allows you to do multiple cuts and adjustments with only one rebuild and the ability to listen and adjust and fix and change on the fly... but doing that in the SoundFile view rebuilds again for each cut... without the ability to go back and adjust the last cuts... I just don't understand why people cling so hard to methods that are clearly not as flexible and powerful... It certaintly IS hard to teach an old dog new tricks. :)

Bob L

[MMP]

But that would still not apply to the situation that was proposed here... to simply take an existing mix and rebuild it over and over as in trimming and cutting in the MT... rather than in the SoundFile View.
Bob L


Right!

Just wanted to make the point that not all build mixes are identical if using certain plugins, before the non-zeroing reports started coming in.:rolleyes:

Regards,

MM

[MikeDee]

I've done it.
I just love hearing you say it :)
Uncle Bob makes us all feel safe... lol...

But seriously, the fact that you have paid such painstaking attention to detail allows me to do what I do and not feel like I have to prove it. Things like headroom, I know that you work in the real world and know how unexpected loud bits just HAPPEN, and you've allowed for it. There's no slop in your design and it's very much appreciated for us folks that wanna just jump in and drive the thing rather than know how to adjust our own timing belts.

Again, thanks so much.

Scott

Ditto on all counts! :cool: :)

I know a producer who swears by Nuendo. I'd like to see him perform this test and see whether he winds up swearing at Nuendo (while, of course, demoing SS). :D

[Dave Labrecque]

But that would still not apply to the situation that was proposed here... to simply take an existing mix and rebuild it over and over as in trimming and cutting in the MT... rather than in the SoundFile View.

The interesting thing is that doing this type of trim work in the MT allows you to do multiple cuts and adjustments with only one rebuild and the ability to listen and adjust and fix and change on the fly... but doing that in the SoundFile view rebuilds again for each cut... without the ability to go back and adjust the last cuts... I just don't understand why people cling so hard to methods that are clearly not as flexible and powerful... It certaintly IS hard to teach an old dog new tricks. :)

Bob L

I think some of us still like the simplicity of chopping off an end of a file without having to go through the extra steps that using the MT entails. When it's something that simple, I still do a quick chop in Sound Forge. No having to place a region on the timeline, etc. Open it and chop it then save it. It's really only a couple extra steps in SAW, but if I have several files to chop...

Plus, if I can keep my current SAW session open and do the quick chop elsewhere, I like that. That having been said, a second instance of SAW is only a double-click away. :)

But as soon as it requires anything a bit more involved, it's SAW's MT all the way. Tweaking an end fade for example without having to process it off-line over and over is killer.

[studio-c]

I use Soundforge for that. Agreed there are some times when I just want to quickly snip something out of a file. And the destructive solution is the quickest.

Maybe the good thing about exiting SS and going into another app is that I"m not going to grab a file that's in my multitrack and corrupt it by creating an event offset. I like just closing the whole SS down and snipping it elsewhere. Same goes for mp3 conversions and other activities. I like having the multitrack software totally out of the equation.

Just a way I've gotten used to working. No reason or justification.

Scott

[Bob L]

No having to place a region on the timeline, etc. Open it and chop it then save it.

Using the Add SoundFile To MT on the file menu is exactly like opening the file in the SF view. No regions to mark and create... its the same.

Chopping it can be done in numerous ways... even easier than marking the SF view and previewing... to chop the end... use the U key... or mark a chunk and use Shift-Delete... etc. and the advantage is you can hear and adjust the results very quickly with the proper use of the Shift and Ctrl Alt boundary adjust options, especially when creating a splice in the middle of the file... and you might find adding a small softedge at the end chop smooths out the results as it fades smoothly to zero instead of chops off.

Saving it is as simple as using the BuildMix instead of the Save command... for files that have been altered in the middle, the resulting time is about the same since the SF edit has to rewrite the file from the beginning anyway... but, if you actually use the Undo option in the old SAW, then the build mix is fater because it is not having to write 2 copies of the whole file.

The only advantage I see in speed and simplicity is when trimming the very end... and as I said earlier, the use of the softedge makes that process much cleaner and I would not want to do it any other way now that I have heard the difference.

Other advantages of the MT proces is the facft that the original has not been destroyed, so you can change your mind later after multiple cuts especially... you can also trim the cuts back and forth which you can't easily do in a SF destructive edut... you mention doing one type of edit this way and another type of edit that way... getting in the flow of using the MT method in SS allows you to do ALL edit types the same way... simplifies the overall process and makes things less complicated.

So for the one process of cutting the tail end off of a 2 hour file, the SF destructive version may truly be more efficient... but in general, I personally find the MT method much more to my liking and leaves me feeling much more in charge of the final decision making.

I simply felt that all the destructive SF editing code and the long process of backing up long wav files as undos for these processes just does not really make sense anymore. Besides, there are dozens of program choices with those options, including the old SAW products which many of the SS users already own. :)

Bob L

[Dave Labrecque]

True enough. Powerful stuff.

The only thing I'd counter with is that the build mix process adds a couple steps (potential extra dialogue box, path/options settings and/or overwrite prompt), whereas a quick ctrl-S in Sound Forge often is all that I need. :)

[Les Woollam]

... but, if you actually use the Undo option in the old SAW, then the build mix is fater because it is not having to write 2 copies of the whole file.

Was there an Undo option for Cut Soundfile in SAW PRO? I don't think there was.

For sure, trimming the end off a long soundfile is quicker than building a new mix, but, as Bob says, cutting chunks out of the middle of a soundfile is going to take a similar amount of time to process as using the build mix function.

The only thing that has me popping back to SAW PRO to edit the soundfile is the fact the the file name is left unchanged, and I don't have to go through the process of naming and re-naming files.

I would like to see this function (maybe along with 'Process Marked Area Back To Soundfile') included in SS, with the ability to disable it in 'Options' (as per 'Disable Retake-All Function').