Sparse Multi-Pitch and Panning Estimation of Stereophonic Signals

Abstract:
In this paper, we propose a novel multi-pitch estimator for stereophonic mixtures, allowing for pitch estimation on multi-channel audio even if the amplitude and delay panning parameters are unknown. The presented method does not require prior knowledge of the number of sources present in the mixture, nor on the number of harmonics in each source. The estimator is formulated using a sparse signal framework, and an efficient implementation using the ADMM is introduced. Numerical simulations indicate the preferable performance of the proposed method as compared to several commonly used multi-channel single pitch estimators, and a commonly used multi-pitch estimator.