Separation of singing voice from music accompaniment for monaural recordings

Separating singing voice from music accompaniment is useful in a wide range of applications, such as lyrics recognition and alignment, singer identification, and music information retrieval. Compared to speech separation, which has been extensively studied for decades, singing voice separation has been little explored. We have developed a pitch‐based system to separate singing voice from music accompaniment for monaural recordings. Our system consists of three stages. The singing voice detection stage partitions and classifies an input into vocal and nonvocal portions. For each vocal portion, the predominant pitch detection stage detects the pitches of the singing voice. Finally, the separation stage uses the detected pitch to group the time‐frequency segments of the singing voice. Quantitative results show that the system performs the separation task successfully. [The work is supported by AFOSR and AFRL.]