结合端点检测的dtw乐谱跟随算法分析-analysis of dtw score following algorithm combined with endpoint detection.docxVIP

ABSTRACTAudio analysis plays a significant role in multimedia information processing. Music is an art composed by organized and regular musical sounds, which expresses human feelings through singing songs or playing musical instruments. The content of music is full of ideals and reflects the real world. People can’t live without music. With the high-speed development of computer science, people have higher demands for computers, hoping computers realize automatic music transcription during musical performance. Score following is an important part of automatic music transcription, which helps computer receive music on live and eventually locates the right position of note in score. To solve the problems that Dynamic Time Warping (DTW) algorithm of score following has low time accuracy and can’t detect error notes, we designed an algorithm based on modified DTW.This paper combines the knowledge of music theory, signal processing and pattern recognitions to propose a system based on DTW algorithm and combined with onset detection algorithm, which extracts the chroma feature based on Constant-Q Transform (CQT), and then compares the two feature vectors of audio signals. Finally, using the DTW algorithm realizes alignment between audio and score, which determines playing notes of each moment. The main work in this paper includes the following aspects:Firstly, framing in the system of score following based on time turns into monophonic or polyphonic, in order that the algorithm is not affected by rhythm and has better robustness. It lays a foundation for detecting error notes.Secondly, we studied the chroma feature in the field of score following and introduced the specific algorithm. This paper uses the chroma feature based on CQT, which is more in line with the music theory characteristics of the scale frequency with exponential distribution.Finally, we compute the Euclidean Distance of two groups of feature vectors, and then figure the optional alignment path using DT