The Recognition Of The Character Based On The Piano Music Signal

The study of music recognition has started as early as in the 1970s. Computers promote the development of modern electronic music technology. Music recognition, classification and feature extraction, such as a series of questions, which base on computer and electronic technology, have been increasingly concerned about by many researchers. Music recognition is a rising interdisciplinary research field, which involves physics, signal processing, human-computer interaction, music theory and music psychics.The article starts with the basic process of music recognition and focus on analyzing and picking-up of the piano performance characters. Employing audio signal analysis and processing, we make it possible for computer to recognize piano music characters. The recognition of piano music can assist the piano performance of teaching to some extent.Firstly, based on music theory and the notes physical attributes the thesis advances collection the music characters which may be used to distinguish, and expands the identifiable characters in the past. Among these seven feature (Timbre, First partial, Time valve, Intensity, Tonality, Speed performance length), we focus on pitch, duration and tonality characters detection. Experi- mental results show that the advanced Short Time Fourier Transform(STFT) pitch detection method achieves high recognition ratio. The paper implements the computer detection of the music piece tonality further, integrating the property of each note with the style of the whole piece.The automatic piano performance evaluation system makes it possible for computers to evaluate music performance automatically for first time. Expect the intelligent music recognition will help the music performance learning. The experiment shows that the system fulfills the anticipated requirements, therefore establishes solid foundation for the future work.