You'll need to get rid of repeats. A nice way is with a set.
import re
mystring = """
DEALER: 'S up, Bubbless?
BUBBLES: Hey.
DEALER: Well, there you go.
JUNKIE: Well, what you got?
DEALER: I got some starters. """
p = re.compile('([A-Z]*):')
s = set(p.findall(mystring))
print s
This results in a unique set of names
set(['JUNKIE', 'DEALER', 'BUBBLES'])