One of the earliest motivations for developing humanoid robots centered on creating robots that may coexist with humans in environments created for human beings. For several years, at Waseda University, the development of the anthropomorphic flutist player has been focused on improving the musical interaction between the human and the robot to clarify the human flute playing and to propose novel assisted music teaching tools. In this paper, a new architecture for autonomously transfer skills from robot to human using the flutist robot is introduced. Furthermore, the new version of the flutist robot, the WF-4R (Waseda Flutist No.4 Refined) is presented; where the arm system was added to assure the positioning accuracy of the flute and the development of a melody recognition system to enable the robot to interact with students at the same logical level of perception. An experimental setup has been performed in order to verify the effectiveness of both mechanical and perceptual systems. As a result, using the arms system, we have assured the repetitiveness of the flute positioning. Furthermore, the implemented music recognition system was able of recognizing the melody of flutist players (an overall recognition rate of 90%); demonstrating that HMM (usually used for speech recognition) is also effective for flute melody identification.