somebody posted this last summer, and I can take no credit for it:
So decide where you want one track to end and the other to begin. If your feed back is 10 seconds long and you want to have each track containing roughly 5 seconds of feedback, then slit her right down the middle - MAKING SURE you consider the audio at the splice point is at a 0 sample value.
To check this, zoom in as close as you can to the audio till you get down to sample values, find your split point at a point where the audio is 0. Once you've found this point, select one half of the audio (lets say the final have) and cut this out (pasting it to your computer's clip board).
Rename the first half of feedback and then start a new waveform..
Paste in the second half of feedback and give this a similar but different name (well, use your judgement - clarity and organization are important values in file handling)
Open your first song and place the first half of feed back in, making sure that it is the VERY LAST waveform to be playing in the song (even if by a few milliseconds).
Save and close the session
Open your second song and place the second half of feed back in. This time, make sure it is the first piece of audio to play....
Use your judgement to make everything sound smooth and euphonic.
or cacophonic, it's your call..
When you want to test how the segue sounds; mix down both songs and then paste them together.
and that's all there is to it!
