I have a several very long WAV files to split into seperate clips on 1 sec of silence points. Apart from spooling through the waveform is there any other quick way to find these points in either Vegas or SF7?
Auto regions really sounded promising but I'm finding it not much help and a search of the SF7 forum seems I'm not alone with this problem.
I'm working with spoken word so the whole beats and measures thing is meaningless. What I'm simply trying to find is every point where the level drops below a certain point for a set duration. In this specific case say -60dB for more than 2 seconds. Now I cannot believe this is rocket science stuff, radio stations have had alarms that trigger when this happens on their broadcast for obvious reasons and they've had them since before transistors were invented.
Maybe Vegas and / or SF aren't the right tools for this kind of work, if that's the case fair enough, can anyone suggest a toolset that does make light work of these kinds of tasks? Other sorts of tools I could find useful is being able to search for every occurance of a particular waveform with a tolerance band, say to find everytime a particular word or segment of music occured in the track.
Of course none of this would be necessary if those who'd made the recording in the first place provided a track list!
It sounds like this is a perfect use for the auto regions tool. I hanen't used it much, so can't give any recommended settings,.It always seemed to work for songs with 1 saecond gaps for me. Try clicking on the help button under the auto regions ui box and it gives a good description of what each setting does. You should be able to make a good preset that will work for you.
"If this was done as the external editor linked to Vegas, would it leave the new waves 'in place' in the Vegas arr?"
I would say no, although you could certainly try it out to see for sure. I'm thinking no, because the regions get extracted to new wave files and the original file in Vegas gets untouched. The orignal file does get region markers added to it. You can right click on a region point and change it to a marker. When you save the file in SF the marker points will appear in Vegas.
Red,
thanks for trying to help. I've read through the docs, seems to me the issue is that the thing is edge triggered which would be just fine on music. Problem is with spoken word there's LOTS of gaps. I can get the settings to the point where it'll put a marker at the start of every word, no problem.
What I really need I think is a tool that is level triggered, say counts samples below a certain level and when the count exceeds a defined value, place a marker. Also what's not helping is it sounds like this stuff has been recorded on a analogue deck, even the 3 second 'silence' is at -62dB, if it was off a CD where the silence was -infinity then probably the edge triggers could be set to find that without a problem. I just don't understand what these guys workflow must have been, obviously they've edited in a DAW, I think Protools, but even that should have been able to handle true silence, unless they went out of that analogue into their DAT.
Anyways, I've got it to the point where it puts about 40 markers (should be only around 8) and it does find all the breaks, so I'll try exporting that and opening it in Vegas, SF doesn't seem to have a jump to next marker like Vegas does, so in Vegas I can just jump to each marker and delete the ones I don't want.
Farss,
I think you're right on with this one. I tried it out a little more, and it doesn't seem to function properly. I adjusted all the settings on a test file I created and couldn't get any results that I expected. It seems like the "minimum beat" duration adjustment does nothing. Then the minimum level setting seems like an ON or OFF switch also. I created a file with a 60 second 0dB Pink noise and reduced a 1 second section by -20 dB, then another 1 second section I muted to -inf. Seems like if I increase the minimum beat duration greater than the 1 second duration, then the region should not be created by either reduced section, since they're only 1 second in duration....well they are still created, even with the maximum 3 second setting. Another thing I tried to do was raise the "minimum level" setting, thinking the section that wasn't muted would get undetected as a region and the muted section would get detected. Well I got to a setting of 89.2% and 3 regions where detected. I raised that setting to 89.3% and NO regions where detected. Well there certainly was a level difference between these 2 sections, yet I had no way of making an adjustment to discern the difference between them.
"SF doesn't seem to have a jump to next marker like Vegas does, so in Vegas I can just jump to each marker and delete the ones I don't want."
Open up the region list view and you can easily select the regions by clicking on the number. You can also right click on the number and delete the region. You can also use the keyboard shortcuts of "CNTRL+Left/right arrow" to jump to the start of a region/marker, then right click on the marker to delete it.
Thanks (again) Red,
who said the audio guys aren't helpful. I'll give it another go tonight. Glad to see you found the same things as I did. I did find running it through the noise gate first helped a bit, I imagine as that put a drop to -inf in the break it then had a decent edge to work off. Now if you could only use the noise gate to place a marker all would be well.
I hate to talk about someone else's software here but Blaze Audio have a neat little Track Splitter program that seems to work a treat. 30 day full functional trial which is more than enough time to get this job done but I think I'll give them the $9.99 asking price anyway.
Seem Nero' s audio editor also has a similar function.
I've found Blaze stuff to be OK, a bit consummer oriented with twee interfaces but in the past they've taken my suggestions on board and emailed me to say the latest version incorporates the features I'd requested, not bad considering they hadn't got a dime out of me.