Musical transcription of sampled audio data is the process of taking a sequence of digital data corresponding to the sound waveform and extracting from it the symbolic information i.e. MIDI scores. In this work, two different approaches for transcription of Indian Classical Music are discussed, one with prior knowledge of Raga and another without it. For this purpose, pitch detection is done using Fourier of Fourier Transform with Parabolic Interpolation to spectral peaks. This method for pitch detection is very accurate and suitable for shrutis in Indian Classical Music. For polyphonic music, we performed source separation by classifying frames as monophonic/polyphonic one. These separated monophonic frames are handled by a monophonic transcription system. Finally results are shown with conversion of Raga audio sample into MIDI format. © 2012 ICST Institute for Computer Science, Social Informatics and Telecommunications Engineering.
CITATION STYLE
Akant, K., Pande, R., & Limaye, S. (2012). Automatic music transcription of Indian classical music into MIDI data. In Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (Vol. 108 LNICST, pp. 347–350). https://doi.org/10.1007/978-3-642-35615-5_55
Mendeley helps you to discover research relevant for your work.