Abstracto

Study of Algorithms for Separation of Singing Voice from Music

Madhuri A. Patil, Harshada P. Burute, Kirtimalini B. Chaudhari, Pradeep B. Mane

Audio signal is an acoustic signal which has frequency range roughly in 20 to 20,000 Hz. Human auditory system has a wonderful ability of effectively focusing on sound in the surrounding. Most audio signals are from the mixing of several sound sources. Separation of singing voice from music has wide range of application such as lyrics recognition, alignment, singer identification, and music information retrieval. Music accompaniment that is often non-stationary & harmonic. Basically, audio signal is time frequency segments of singing voice. An audio signal classification system should be able to categorize different audio format like speech, background noise, and musical genres, singer identification, karaoke etc. In this paper, discuss about separation technique and classifier which are used for singing voice separation from music. Non-negative matrix factorization (NMF) is used for separation from music, Gaussian mixture model (GMM) & Support vector machine (SVM) classifier for the classification.