Drum Transcription in the presence of pitched instruments using Prior Subspace Analysis

This paper demonstrates of Prior Subspace Analysis (PSA) as a method for transcribing drums in the presence of pitched instruments. PSA uses prior subspaces that represent the sources to be transcribed to overcome some of the problems associated with other subspace methods such as Independent Subspace Analysis (ISA) or sub-band ISA. The use of prior knowledge results in improved robustness for transcription purposes and enables the method to work more readily in the presence of pitched instruments than other subspace methods. The effectiveness and robustness of PSA as a tool for drum transcription in the presence of pitched instruments is demonstrated in a simple drum transcription algorithm. __________________________________________________________________________________________ I INDEPENDENT SUBSPACE ANALYSIS Independent Subspace Analysis (ISA) is a technique for separating sound sources from single channel mixtures [1]. Based on the concept of reducing redundancy in time-frequency representations it represents sound sources as low dimensional subspaces in the time-frequency plane. The signal is transformed to a magnitude spectrogram by means of a Short-Time Fourier Transform (STFT). ISA assumes that the mixture signal spectrogram Y can be decomposed into l statistically independent spectrograms Yj: