Non-negative matrix factorization for polyphonic music transcription

By Paris Smaragdis, Judith C. Brown. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2003

1,035 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/34343b4ae08f6069d78b02c26d64c17f7c92a906

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Signal Processing

Citations in the corpus

Onsets and Frames: Dual-Objective Piano Transcription (2017, methodology): “early popular method used in the task of polyphonic music transcription [19].”
Automatic Music Transcription: An Overview (2019, methodology): “Recent approaches include traditional signal processing methods [11], [12], probabilistic modeling [8], Bayesian approaches [13], NMF [14]–[17], and neural networks (NNs) [18], [19].”
Nonsmooth nonnegative matrix factorization (nsNMF) (2006, methodology): “NMF has been successfully used in diverse fields of science suchasbiomedicalapplications[4],[5],[6],[7],faceandobject recognition [8], [9], color science [10], [11], and polyphonic music transcription [ 12 ], among others.”
The Sound of Pixels (2018, methodology): “Classical approaches include signal processing methods such as Nonnegative Matrix Factorization (NMF) [42,8,40].”
An End-to-End Neural Network for Polyphonic Piano Music Transcription (2015, methodology): “Unsupervised NMF approaches [6], [7] aim to learn a dictionary of pitch spectra from the training examples.”
Automatic music transcription: challenges and future directions (2013, methodology): “first introduced as a tool for music transcription in [119].”
Algorithms for Nonnegative Matrix Factorization with the β-Divergence (2010, background): “…a few, let us mention the problems of learning parts of faces and semantic features of text (Lee and Seung, 1999), polyphonic music transcription (Smaragdis and Brown, 2003), object characterization by reﬂectance spectra analysis (Berry et al., 2007), portfolio diversiﬁcation (Drakakis et al.,…”
Exemplar-Based Sparse Representations for Noise Robust Automatic Speech Recognition (2011, methodology): “In most speech applications of sparse representations, the dictionary atoms either consist of fundamental basis functions such as Fourier coefﬁcients or wavelets, or are learned [11]–[14], [21].”
Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria (2007, methodology): “The algorithms have been used in several unsupervised learning tasks and also in the analysis of music signals, where the nonnegativity constraints alone have turned out to be sufficient for the separation of sound sources [8].”
Nonnegative Matrix Factorization: An Analytical and Interpretive Tool in Computational Biology (2008, background): “…and facial pattern recognition [8–19], natural language processing such as in text mining and document clustering (see [20–22] and references therein), sparse coding [23–27], information retrieval [28,29], speech recognition [30–33], video summarization [34], and Internet research [35,36].”
Nonnegative Matrix Factorization with the Itakura-Saito Divergence: With Application to Music Analysis (2009, background): “To cite a few, we mention the problems of learning parts of faces and semantic features of text (Lee & Seung, 1999), polyphonic music transcription (Smaragdis & Brown, 2003), object characterization by reflectance spectra analysis (Berry, Browne, Langville, Pauca, & Plemmons, 2007), portfolio…”
Generalized Nonnegative Matrix Approximations with Bregman Divergences (2005, methodology): “An somewhat offbeat application to Polyphon ic music transcription is presented by Smaragdis and Brown [33].”
Convolutive Speech Bases and Their Application to Supervised Speech Separation (2007, methodology): “Applications of NMF on audio data are presented in [7].”
Nonnegative signal factorization with learnt instrument models for sound source separation in close-microphone recordings (2013, methodology): “NMF methods have been shown to be specially useful for musical analysis tasks [11], including not only SSS but also others, such as automatic music transcription [12] or acoustic space characterization [13].”
Spectral vs. spectro-temporal features for acoustic event detection (2011, background): “, columns of a spectrogram), and have each NMF bases consist of a single short time frame [3].”