• Publications
  • Influence
The State of the Art Ten Years After a State of the Art: Future Research in Music Information Retrieval
TLDR
A case study of all published research using the most-used benchmark dataset in MGR during the past decade shows that none of the evaluations in these many works is valid to produce conclusions with respect to recognizing genre, i.e. that a system is using criteria relevant for recognizing genre. Expand
A Survey of Evaluation in Music Genre Recognition
  • Bob L. Sturm
  • Computer Science
  • Adaptive Multimedia Retrieval
  • 24 October 2012
TLDR
This paper compiles a bibliography of work in MGR, and analyzes three aspects of evaluation: experimental designs, datasets, and figures of merit. Expand
Music transcription modelling and composition using deep learning
TLDR
This work builds and train LSTM networks using approximately 23,000 music transcriptions expressed with a high-level vocabulary (ABC notation), and uses them to generate new transcriptions to create music transcription models useful in particular contexts of music composition. Expand
Comparison of orthogonal matching pursuit implementations
We study the numerical and computational performance of three implementations of orthogonal matching pursuit: one using the QR matrix decomposition, one using the Cholesky matrix decomposition, andExpand
Deep Learning and Music Adversaries
TLDR
This work builds adversaries for deep learning systems applied to image object recognition by exploiting the parameters of the system to find the minimal perturbation of the input image such that the system misclassifies it with high confidence. Expand
The GTZAN dataset: Its contents, its faults, their effects on evaluation, and its future use
TLDR
This article disprove the claims that all MGR systems are affected in the same ways by these faults, and that the performances of M GR systems in GTZAN are still meaningfully comparable since they all face the same faults. Expand
An analysis of the GTZAN music genre dataset
TLDR
An analysis of its composition is provided, and a machine-readable index of artist and song titles is created that catalogs numerous problems with its integrity, such as replications, mislabelings, and distortions. Expand
MATConcat: An Application for Exploring Concatenative Sound Synthesis Using MATLAB
TLDR
An application in MATLAB implementing concatenative sound synthesis (CSS) using feature matching, which is meant to be a free and open application for demonstrat- ing and experimenting with the process. Expand
Classification accuracy is not enough
  • Bob L. Sturm
  • Computer Science
  • Journal of Intelligent Information Systems
  • 1 December 2013
We argue that an evaluation of system behavior at the level of the music is required to usefully address the fundamental problems of music genre recognition (MGR), and indeed other tasks of musicExpand
Ensemble Models for Spoofing Detection in Automatic Speaker Verification
TLDR
This work investigates why some models on the PA dataset strongly outperform others and finds that spoofed recordings in the dataset tend to have longer silences at the end than genuine ones. Expand
...
1
2
3
4
5
...