Digital Nightfall on 19/2/2008 at 08:47
Without access to the source files, there's no way to separate the voiceover from the background music. That CoK was able to work so much out of it in spite of that limitation is remarkable.
Beleg Cúthalion on 19/2/2008 at 09:38
Plus, the music in the cutscenes is often simply made of the ambient sounds or other material you can find (like the music from the intro, credits etc.) elsewhere. There are only a few pieces that have a remarkable music like the Hammer choirs before Cragscleft, the Viktoria theme, the "City Warden Shuffle" or whatever you call the little loop before Ramirez and this little string theme you can hear before RotC.
Child Of Karras on 19/2/2008 at 13:34
Well, basically it's not difficult to extract the music from the videos. But there're two important conditions the video has to offer: first, Garrets voice has to be at 0°, meaning completely mono. Furthermore the music then has to be stereo.
Then it's only a question of combining phase erasement tools, equalising, reverb and noise reduction in the right order.
The Magpie on 19/2/2008 at 16:05
Quote Posted by Child Of Karras
basically it's not difficult
Ladies and gentlemen, this was spoken by a true artisan. Hat's off.
;)
--
L.
Child Of Karras on 20/2/2008 at 00:04
*lol* Okay, I have to confess that it's a bit hard for me to get into a audio noob point of view. I do this for a couple of years now.
To tell it in the mathematical way: we need to get the difference of both audio channels. If we consider Garret's voice as "1", then on both channels (left and right) garrett is "1", so he's in the middle. Now we just subtract one channel from the other and Garrett is gone (1-1=0). And only music it what's left. :p
Child Of Karras on 20/2/2008 at 10:26
Well, the same way would be just adding both channels. Audition has got a function called "Center Channel Extractor" which works FFT-based (frequency analysis) and provides better results. I also used this for the videos but in the opposite way, so I reduced the center channel, because it keeps the Stereo image. Taking the difference of both channels leaves only one channel left.
Chade on 20/2/2008 at 12:41
EDIT: Sorry, I am talking about the subtraction of the two speaker sounds here ...
But surely there are an infinite number of possible stereo sounds which could have generated the remaining wave? Is only one of these possibilities reasonable?
Or am I just plain wrong? :eww:
ignosco on 21/2/2008 at 06:01
:D Thank you, thank you, thank you! The music is so awesome! But what I was saying before is that I thought there would of been intire songs or something like that made for the game. I can't get over how awesome this is!:)
Dooh Nibor on 23/2/2008 at 14:25
:eek: Great job! Thank you, thank you very much! :thumb:
