A tutorial on onset detection in music signals
Citations
372 citations
298 citations
296 citations
Cites background from "A tutorial on onset detection in mu..."
...A survey by Aucouturier and Pachet (2003) describes a number of popular features for music similarity and classification, and research continues (e.g. Bello et al. (2005), Pampalk et al. (2005))....
[...]
284 citations
266 citations
Cites methods from "A tutorial on onset detection in mu..."
...The three performance measures used here are Precision (P), the ratio of Hits to Detected Changes and Recall (R), the ratio of hits to transcribed changes and the f-measure (F) which combines the two (see equation 5) [ 1 ]....
[...]
References
3,830 citations
"A tutorial on onset detection in mu..." refers methods in this paper
...The algorithms described in [30] are concerned with detecting this change of sign....
[...]
...1) Model-Based Change Point Detection Methods: A wellknown approach is based on the sequential probability ratio test [30]....
[...]
2,968 citations
1,956 citations
"A tutorial on onset detection in mu..." refers background in this paper
...Given the importance of musical events, it is clear that identifying and characterizing these events is an important aspect of this process....
[...]
1,659 citations
"A tutorial on onset detection in mu..." refers methods in this paper
...However, we will focus only on two processes that are consistently mentioned in the literature, and that appear to be of particular relevance to onset detection schemes, especially when simple reduction methods are implemented: separating the signal into multiple frequency bands, and…...
[...]
793 citations
"A tutorial on onset detection in mu..." refers methods in this paper
...By implementing a multiple-band scheme, the approach effectively avoids the constraints imposed by the use of a single reduction method, while having different time resolutions for different frequency bands....
[...]
Related Papers (5)
Frequently Asked Questions (9)
Q2. What are the contributions mentioned in the paper "A tutorial on onset detection in music signals" ?
The goal of this paper is to review, categorize, and compare some of the most commonly used techniques for onset detection, and to present possible enhancements. The authors discuss methods based on the use of explicitly predefined signal features: the signal ’ s amplitude envelope, spectral magnitudes and phases, time-frequency representations ; and methods based on probabilistic signal models: model-based change point detection, surprise signals, etc. Using a choice of test cases, the authors provide some guidelines for choosing the appropriate method for a given application.
Q3. What is the way to detect onsets?
For nontrivial sounds, onset detection schemes benefit from using richer representations of the signal (e.g., a time-frequency representation).
Q4. What is the expectation of the log-likelihood ratio?
Under model , the expectation is(15)If the authors assume that the signal initially follows model , and switches to model at some unknown time, then the short-time average of the log-likelihood ratio will change sign.
Q5. What is the scheme for detecting transients?
The scheme takes advantage of the correlations across scales of the coefficients: large wavelet coefficients, related to transients in the signal, are not evenly spread within the dyadic plane but rather form “structures”.
Q6. What is the procedure used in the majority of onset detection algorithms?
Fig. 2 illustrates the procedure employed in the majority of onset detection algorithms: from the original audio signal, which can be pre-processed to improve the performance of subsequent stages, a detection function is derived at a lower sampling rate, to which a peak-picking algorithm is applied to locate the onsets.
Q7. What is the alternative to the analysis of the temporal envelope of the signal and of Fourier?
An alternative to the analysis of the temporal envelope of the signal and of Fourier spectral coefficients, is the use of time-scale or timefrequency representations (TFR).
Q8. What is the general approach to detecting a frequency?
A more general approach based on changes in the spectrum is to formulate the detection function as a “distance” between successive short-term Fourier spectra, treating them as points in an -dimensional space.
Q9. What was the threshold for the peak-picking parameters?
All peak-picking parameters (e.g., filter’s cutoff frequency, ) were held constant, except for the threshold which was varied to trace out the performance curve.