Singing voice separation with pre-learned dictionary and reconstructed voice spectrogram
Singing voice separation with pre-learned dictionary and reconstructed voice spectrogram is a scholarly work, published in 2018 in ''Neural Computing and Applications''. The main subjects of the publication include superposition principle, independent component analysis, separation, rank, artificial intelligence, matrix, annotation, singing, Speech enhancement, spectrogram, speech recognition, audio signal processing, Source separation, pattern recognition, and computer science. Based on these considerations, in this paper, authors present two categories, time–frequency-based source separation algorithms.