Study on Multi-Tone Piano Transcription Algorithm Based on BPNN

BPNN ensemble can significantly improve the generalization ability of knowledge system by training multiple BPNNs and synthesizing their conclusions. It not only helps scientists to study machinery knowledge and neural computing in depth, but also helps ordinary engineers and technicians to use BPNN technology to solve real-world problems. Ensemble knowledge has become one of the hot spots in the territory of machinery knowledge in recent years, and selective integration method has become an important direction of ensemble knowledge because of its advantages in adaptability, generalization and combination. In this paper, the transcription of multi-tone piano is studied based on BPNN. The corresponding study methods are used in the research. Through the establishment of data graph and algorithm formula, the corresponding study is carried out. From the research, it can be found that the piano transcription efficiency based on BPNN is very high, up to about 90.43%. In the future, people may pay more Focus to the piano study of BPNN.