… non-negative matrix deconvolution (NMD) which models spectrogram matrix as a linear
combination of dictionary … While the speaker recognition accuracy is lower for the new approach, …

… is proposed to model a spectrogram. This model is similar to … nonnegative matrix deconvolution
in a speaker recognition task. In order to apply BNMD dictionaries in speaker recognition, …

… In our application X = | X|.2 is a speech power spectrogram, … , where speakers are characterized
through dictionaries while … based on supervised non-negative matrix factorization,” in In …

… dimensional deconvolution (NMF2D) with adaptive sparsity … tensor factor two dimensional
deconvolution (K-wNTF2D) has … : Spectrogram of the original and estimated sources by using …

… tic pattern recognition tasks such as speaker identification [8] … MEL-magnitude spectrograms
[25]. The MEL-magnitude … to select the correct number of dictionary elements in the learned …

… -based non-negative matrix … spectrogram are estimated through the source dictionary
and are then applied to the target dictionary to generate the corresponding target spectrogram. …

… In this sense, with the aim of validating its use as a new limited-data automatic speaker
verification technique, this paper proposes to use the BLPOC function as a simple matching …

… Because the dictionary are built using real audio sounds of drum strikes, we have … spectrogram
of the dictionary as IX.,.,pl = W.,. ,P . We can reconstruct a synthetic drum signal using the …

… on noise robustness of ASR using a nonnegative matrix factorization (NMF) -… non-negative
matrix deconvolution NMF … This kind of DTFT-based Mel scale magnitude spectrograms have …

… (ASR), hearing aids and speaker identification systems. … spectrogram features for the
NMF-based tasks. The idea of … the frequency domain and non-negative matrix deconvolution …