I tested Rapiro with 13.8V power supply and so far it didn't give me any burning smell and everything seems working. Of course I also checked Block chart of communication and power supply of RAPIRO before I did that.
https://s3.amazonaws.com/ksr/projects/581514/posts/552418/image-299410-full.jpg?1375036965

I have also tried a laptop DC power supply but all the motor just rotate in one direction continuously which is really bad for the motors.

On the other hand I also tried New Alkaline battery. Yes it would not give you any luck.

NOTE: I recommend put your robot in #M0 to make sure your motor won't suddenly run crazy when you connect the connector.

FURTHER NOTE on walking: I realized the walking demo highly toward the left sides of the body to create the motion. It need to touch his left hand to help balance. I wonder is that because of the tension from the cables.

Update on 1st March, 2014: #M0 position motor noise can be heard. I think a record motor action replay function will be needed.

Update on 2nd March, 2014: I found out you cannot send commend from serial monitor after you connect the raspiro board to the rasberrypi. You would still able to program the raspiro for calibration and others. The solution of this is I send
"pi@rapiro ~ $ echo "#M0" | sudo minicom -b 57600 -o -D /dev/ttyAMA0" from pi to send the commend instead. This line of code credit to fvan.
Note: When they both connected I disconnect the power supply and solely use microUSB to program raspiro.
On the other hand, I was also doing calibration try to eliminate the noise when the motor is on. Noise is better, but is still there.
This pic is used during calibration http://hidakanoteimages.s3.amazonaws.com/2014/02/rapiro_servo_all.jpg