• Corpus ID: 10768781

MIREX 2019: VAMP PLUGINS FROM THE CENTRE FOR DIGITAL MUSIC

@inproceedings{Cannam2013MIREX2V,
  title={MIREX 2019: VAMP PLUGINS FROM THE CENTRE FOR DIGITAL MUSIC},
  author={Chris Cannam and Emmanouil Benetos and Matthias Mauch and Matthew E. P. Davies and Simon Dixon and Christian Landone and Katy C. Noland},
  year={2013}
}
In this submission we offer for evaluation several audio feature extraction plugins in Vamp format. Most of these plugins were also submitted to the 2013, 2014, 2015, 2016, 2017, and 2018 editions of MIREX; most of them are also unchanged since 2016, and may offer a useful baseline for comparison across years. The methods implemented in this set of plugins are described in the literature and are referenced throughout this paper. All of the plugins are written in C++ and have been published… 
Extracting Ground-Truth Information from MIDI Files: A MIDIfesto
TLDR
This work enumerates the types of information available in MIDI files and describes the steps necessary for utilizing them and quantifies the reliability of this data by comparing it to human-annotated ground truth.
EXTRACTING GROUND TRUTH INFORMATION FROM MIDI FILES:
TLDR
This work enumerates the types of information available in MIDI files and describes the steps necessary for utilizing them and quantifies the reliability of this data by comparing it to human-annotated ground truth.
An exploration of the current state-of-the-art in automatic music transcription - with proposed improvements using machine learning
TLDR
A novel approach of applying complex-valued neural networks in music transcription is examined, by modifying research in an existing deep complex neural network model, indicating that complex neural networks may develop the research area of music transcription even further.
End-to-end musical key estimation using a convolutional neural network
TLDR
An end-to-end system for musical key estimation, based on a convolutional neural network that out-performs existing key estimation methods proposed in the academic literature and is capable of learning a unified model for diverse musical genres that performs comparably to existing systems specialised for specific genres.
Supervised Metric Learning for Music Structure Feature
TLDR
This work takes a supervised metric learning approach, and proposes a batch sampling scheme to ensure the labels in a training pair are interpreted meaningfully and can improve a traditional MSA algorithm significantly in both intraand cross-dataset scenarios.
Deeper Convolutional Neural Networks and Broad Augmentation Policies Improve Performance in Musical Key Estimation
TLDR
This paper adapts the Inception architecture to the specific needs of harmonic music analysis and uses it to create a model (InceptionKeyNet) for the task of key estimation that can significantly outperform state-of-the-art single-task models when trained on the same datasets.
Music Boundary Detection using Convolutional Neural Networks: A comparative analysis of combined input features
TLDR
This work establishes a general method of pre-processing inputs to the CNN by comparing the inputs calculated from different pooling strategies, distance metrics and audio characteristics, also taking into account the computing time to obtain them.
Supervised Metric Learning For Music Structure Features
TLDR
This work trains a deep neural network to output embeddings that are near each other for two spec-trogram inputs if both have the same section type (accord-ing to an annotation), and otherwise far apart, and proposes a batch sampling scheme to ensure the labels in a training pair are interpreted meaningfully.
DECIBEL: Improving Audio Chord Estimation for Popular Music by Alignment and Integration of Crowd-Sourced Symbolic Representations
TLDR
DECIBEL is proposed, a new ACE system that exploits widely available MIDI and tab representations to improve ACE from audio only and shows that the integration of musical knowledge from heterogeneous symbolic music representations is a suitable strategy for addressing challenging MIR tasks such as ACE.
Deep learning-based automatic downbeat tracking: a brief review
TLDR
A background introduction of downbeat tracking problem, detailed discussions of the following topics: system architecture, feature extraction, deep neural network algorithms, data sets, and evaluation strategy, and a look at the results from the annual benchmark evaluation—Music Information Retrieval Evaluation eXchange.
...
...

References

SHOWING 1-10 OF 20 REFERENCES
MIREX 2010 : CHORD DETECTION USING A DYNAMIC BAYESIAN NETWORK
TLDR
This submission MD1 to the Chord Estimation Task of the 2010 Music Information Retrieval Evaluation eXchange (MIREX 2010) has achieved the highest chord overlap ratio (0.8022) among the participating methods.
MIREX 2006 Audio Beat Tracking Evaluation: BeatRoot
TLDR
The original simple onset detection algorithm, which caused problems for beat tracking music without prominent drums, has been replaced with a more robust onset detector, and several new features have been added, such as annotation of multiple metrical levels and phrase boundaries, and improvements in the user interface.
Tonal Description of Polyphonic Audio for Music Content Processing
  • E. Gómez
  • Computer Science
    INFORMS J. Comput.
  • 2006
TLDR
A method to extract a description of the tonal aspects of music from polyphonic audio signals using different levels of abstraction, differentiating between low-level signal descriptors and high-level textual labels.
Adaptive whitening for Improved Real-Time audio onset Detection
TLDR
Results demonstrate that adaptive whitening can significantly improve the performance of various STFT-based onset detection functions, including functions based on the power, spectral flux, phase deviation, and complex deviation measures.
MULTIPLE-F 0 ESTIMATION AND NOTE TRACKING FOR MIREX 2012 USING A SHIFT-INVARIANT LATENT VARIABLE MODEL Emmanouil Benetos
TLDR
This submission for MIREX 2012 utilizes a shift-invariant latent variable model for multiple-F0 estimation and note tracking that extends the SI-PLCA method and employs several note templates from multiple orchestral instruments.
Approximate Note Transcription for the Improved Identification of Difficult Chords
TLDR
This paper seeks to find chroma features that are more suitable for usage in a musically-motivated model by performing a prior approximate transcription using an existing technique to solve non-negative least squares problems (NNLS).
Using Musical Structure to Enhance Automatic Chord Transcription
TLDR
The present work shows that additional exploitation of the repetitive structure of songs can enhance chord extraction, by combining chroma information from multiple occurrences of the same segment type.
A Shift-Invariant Latent Variable Model for Automatic Music Transcription
TLDR
Results demonstrate that the proposed probabilistic model for multiple-instrument automatic music transcription outperforms leading approaches from the transcription literature, using several error metrics.
Context-Dependent Beat Tracking of Musical Audio
TLDR
A simple and efficient method for beat tracking of musical audio using a two state model that replicates the human ability of tapping in time to music at significantly reduced computational cost.
Automatic chord transcription from audio using computational models of musical context
TLDR
A novel dynamic Bayesian network (DBN) is presented which integrates models of metric position, key, chord, bass note and two beat-synchronous audio features into a single high-level musical context model.
...
...