Factors in automatic musical genre classification of audio signals

By Tao Li, George Tzanetakis. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2003

2,270 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/602f31242e577d2d05f918a3080fd50095e7faed

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Search Result Clustering in Collaborative Sound Collections (2020, methodology): “equencycontentofsoundsignals.Combiningthesedifferenttypes of features enables the representation of the timbre of musical instruments or more high-level characteristics such as music moods and genres [11, 18, 23, 31, 41]. Recently, techniques using Artificial Neural Networks have been able to provid”
Content-based organization and visualization of music archives (2002, background): “[30]) and a combination might yield superior results.”
SoundSense: scalable sound sensing for people-centric applications on mobile phones (2009, methodology): “For voice and music input, techniques such as speech recognition [32], speaker identification [33], or music genre classification [41] can be applied.”
Automatic mood detection and tracking of music audio signals (2006, methodology): “A number of previous research works [4], [26] utilized MFCC and so-called spectral shape features to represent the timbre of audio and music signals.”
Monitoring temporal change of bird communities with dissimilarity acoustic indices (2014, methodology): “…cepstral coefficents (MFCCs) are features used in automatic speech and speaker recognition (Rabiner & Juang 1993), music information retrieval (Tzanetakis & Cook 2002) and now in animal vocalization identification (Lee, Lee & Huang 2006a; Lee et al. 2006b) but were used here for the first…”
Features for audio and music classification (2003, background): “In comparison to results of other studies of music genre classification (Tzanetakis et al., 2001; Tzanetakis & Cook, 2002), our features looks quite promising.”
A Regression Approach to Music Emotion Recognition (2008, methodology): “Marsyas is a free software framework for rapid development and evaluation of computer audition applications [ 29 ].”
MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Training (2023, methodology): “In this task, some models were trained on other datasets, and the full GTZAN set was used as the test set.”
A Unified Perspective on Multi-Domain and Multi-Task Learning (2014, methodology): “We consider a music-speech discrimination task introduced by Tzanetakis & Cook (2002), which includes 64 music and speech tracks.”
Freesound Datasets: A Platform for the Creation of Open Audio Datasets (2017, background): “While we believe that annotator agreement is required for defining ground truth based on human-sourced annotations [8], it is also true that a number of datasets do not meet this condition [24,23,39].”
Aggregate features and ADABOOST for music classification (2006, methodology): “This is the same dataset used in Tzanetakis et al. (2002), Tzanetakis and Cook (2002), and Li and Tzanetakis (2003).”
Are the Emotions Expressed in Music Genre-specific? An Audio-based Evaluation of Datasets Spanning Classical, Film, Pop and Mixed Genres (2011, methodology): “…role in determining what those features might be, particularly since the high success rate of genre-recognition algorithms show that genres do indeed have distinct musical features (Tzanetakis & Cook, 2002b; Bergstra, Casagrande, Erhan, Eck, & Kögl, 2006; Silla Jr, Kaestner, & Koerich, 2007).”
PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition (2019, background): “5) GTZAN: The GTZAN dataset [57] is a music genre classification dataset containing 1,000 30-second music clips of 10”
Song-Level Features and Support Vector Machines for Music Classification (2005, methodology): “Tzanetakis and Cook (2002) also calculate song-level features.”
Cross-Collection Evaluation for Music Classification Tasks (2016, positive): “6 A collection of 1000 tracks for 10 music genres (100 per genre) [23], including blues, classical, country, disco, hip-hop, jazz, metal, pop, reggae, and rock.”
Open-Source Practices for Music Signal Processing Research: Recommendations for Transparent, Sustainable, and Reproducible Audio Research (2019, background): “, Which one of n genres describes this song?—which was often evaluated on the GTZAN data set (1,000 musical excerpts balanced across ten genres) [1].”