Efficient implementation of a system for solo accompaniment separation in polyphonic music

Our goal is to obtain improved perceptual quality for sepa-rated solo instruments and accompaniment in polyphonic mu-sic. The proposed approach uses a pitch detection algorithm in conjunction with a spectral filtering based source separa-tion. The algorithm was designed to work with polyphonic signals regardless of the main instrument, type of accompani-ment or musical style. Our approach features a fundamental frequency estimation stage, a refined harmonic structure for the spectral mask and a post-processing stage to reduce ar-tifacts. The processing chain has been kept light. The use of perceptual measures for quality assessment revealed im-proved quality in the extracted signals with respect to our pre-vious approach. The results obtained with our algorithm were compared with other state-of-the-art algorithms under SISE