Exploring automatic music annotation with "acoustically-objective" tags

109 citations recorded by the source corpus.

How researchers see this paper

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

How algorithmic confounding in recommendation systems increases homogeneity and decreases utility (2017, background): “movie genre), or they can be learned independent of time t with, for example, a topic model [6] for text or from the acoustic structure of a song [57] for music; when learned, the attribute tags can be”
Learning-Based Methods for Comparing Sequences, with Applications to Audio-to-MIDI Alignment and Matching (2016, unknown): “, 2007), CAL10k (Tingle et al., 2010), and uspop2002 (Berenzweig et al.”
FMA: A Dataset for Music Analysis (2016, unknown): “…3,115 2010 yes 1517-Artists [40] 3,180 1,517 2008 yes LMD [42] 3,227 - 2007 no EBallroom [23] 4,180 - 2016 no USPOP [1] 8,752 400 2003 no CAL10k [44] 10,271 4,597 2010 no MagnaTagATune [20] 25,863 3 230 2009 yes 4 Codaich [28] 26,420 1,941 2006 no FMA 106,574 16,341 2017 yes OMRAS2 [24] 152,410…”
The Million Song Dataset (2011, methodology): “This makes it the largest available dataset for tagging evaluation, as compared to Mag-natagatune [9], Swat10K [10] and the Last.”