Katherine Mangu-Ward from the February 2009 issue
Meet Eduardo Miranda’s singing robots. Miranda is a researcher at the Interdisciplinary Centre for Computer Music Research at the University of Plymouth. He built these two warbling toasters, gave them both a vocal range and the ability to adapt, then put them in a room together for two weeks. They went in singing radically different “tunes”—random sequences of six notes at a time—and came out crooning an oddly haunting duet consisting of about 20 sounds.
The robots have simplified vocal cords, and their voices are strangely human. They also have microphones for ears and cameras for eyes. Their song was a spontaneous, emergent order. One robot would sing its randomly generated sequence, then the other would follow suit with its own tune. If the first robot judged the sounds to be similar, it would nod and store the notes for future use. If the sounds were too different, it would shake its head and the tunes would be forgotten.
A shared musical vocabulary evolved. “The robots develop their own musical culture,” Miranda told New Scientist. “There are no preprogrammed musical rules.”
Katherine Mangu-Ward (kmw@reason.com) is an associate editor of reason. To hear the robots sing, click here.
Reason needs your support. Please donate today!
Try Reason's award-winning print edition today! Your first issue is FREE if you are not completely satisfied.
(310) 367-6109
3415 S. Sepulveda Blvd.
Suite 400
Los Angeles, CA 90034
(310) 391-2245
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment or disable your ability to comment for any reason at any time.
nfl jerseys|11.5.10 @ 4:27AM|#
eryfgh
Jon|4.24.11 @ 5:00PM|#
http://www.freshporn.org
Scarpe Nike|8.8.11 @ 2:21AM|#
is good
monclers jacket outlet|11.5.11 @ 3:02AM|#
good
caibutou|2.13.12 @ 9:36PM|#
good