2010:Harmonic Analysis - Revision history

CameronNonAdmin: /* References */

2011-06-04T11:58:04Z

‎References

CameronNonAdmin: /* References */

2011-05-21T17:02:19Z

‎References

CameronNonAdmin: /* References */

2011-05-20T11:22:06Z

‎References

CameronNonAdmin: /* References */

2011-04-28T22:50:49Z

‎References

CameronNonAdmin: /* Description */

2011-04-20T22:30:37Z

‎Description

Sutashu: /* Evaluation Measures */

2010-05-25T06:19:21Z

‎Evaluation Measures

Sutashu: /* Ground truth */

2010-05-25T06:12:26Z

‎Ground truth

Sutashu: /* Description */

2010-05-25T06:09:16Z

‎Description

Sutashu: /* References */

2010-05-15T09:20:29Z

‎References

Sutashu: /* Evaluation Measures */

2010-05-15T09:18:30Z

‎Evaluation Measures

@@ Line 134: / Line 134: @@
 # P. Kröger, A. Passos, M. Sampaio, G. de Cidra, [http://classes.berklee.edu/mbierylo/ICMC08/defevent/papers/cr1516.pdf “Rameau: a system for automatic harmonic analysis,”] Proc. of ICMC, 2008
 # C. S. Sapp, [http://www.ccarh.org/publications/cm/vol/15/cm15-06-sapp.pdf “Computational chord-root identification in symbolic musical data: rationale, methods and applications,”] Computing in Musicology 15, 2007
-# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.dissertationwritinghelp.co.uk/ Dissertation Help] [http://www.customwrittenpaper.com/ College Essay Help]
+# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.dissertationwritinghelp.co.uk/ Dissertation Help] [http://www.customwrittenpaper.com/ College Essay Help] [http://power-scooters.com/ power scooters]
 [http://www.logoinn.com/ Logo Design] [http://www.logoinn.com/web-design/ Web Design]

@@ Line 134: / Line 134: @@
 # P. Kröger, A. Passos, M. Sampaio, G. de Cidra, [http://classes.berklee.edu/mbierylo/ICMC08/defevent/papers/cr1516.pdf “Rameau: a system for automatic harmonic analysis,”] Proc. of ICMC, 2008
 # C. S. Sapp, [http://www.ccarh.org/publications/cm/vol/15/cm15-06-sapp.pdf “Computational chord-root identification in symbolic musical data: rationale, methods and applications,”] Computing in Musicology 15, 2007
-# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.customwrittenpaper.com/ College Essay Help]
+# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.dissertationwritinghelp.co.uk/ Dissertation Help] [http://www.customwrittenpaper.com/ College Essay Help]
 [http://www.logoinn.com/ Logo Design] [http://www.logoinn.com/web-design/ Web Design]

@@ Line 135: / Line 135: @@
 # C. S. Sapp, [http://www.ccarh.org/publications/cm/vol/15/cm15-06-sapp.pdf “Computational chord-root identification in symbolic musical data: rationale, methods and applications,”] Computing in Musicology 15, 2007
 # H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.customwrittenpaper.com/ College Essay Help]
 == Potential Participants ==
 == Discussion for 2010 ==

@@ Line 134: / Line 134: @@
 # P. Kröger, A. Passos, M. Sampaio, G. de Cidra, [http://classes.berklee.edu/mbierylo/ICMC08/defevent/papers/cr1516.pdf “Rameau: a system for automatic harmonic analysis,”] Proc. of ICMC, 2008
 # C. S. Sapp, [http://www.ccarh.org/publications/cm/vol/15/cm15-06-sapp.pdf “Computational chord-root identification in symbolic musical data: rationale, methods and applications,”] Computing in Musicology 15, 2007
-# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999
+# H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999 [http://www.customwrittenpaper.com/ College Essay Help]
 == Potential Participants ==
 == Discussion for 2010 ==

@@ Line 1: / Line 1: @@
 == Description ==
-This task is suggested for MIREX 2010. The general goal is to estimate the latent harmonic progression from score-like music data, a process typically called harmonic analysis or functional analysis in musicology (although this task cannot be really called functional analysis, functional analysis can be used to obtain the required output data). This typically involves estimating the musical key of a given piece of music and movement of the tonal center (modulations and tonicizations), as well as the progression of chords.
+This task is suggested for MIREX 2010. The general goal is to estimate the latent harmonic progression from score-like music data, a process typically called harmonic analysis or functional analysis in musicology (although this task cannot be really called functional analysis, functional analysis can be used to obtain the required output data). This typically involves estimating the musical key of a given piece of music and movement of the tonal center (modulations and tonicizations), as well as the progression of chords. [http://supremedissertation.co.uk/ Dissertation Proposal]
 === Time units ===

@@ Line 88: / Line 88: @@
 == Evaluation Measures ==
-Because of the large diversity in the dictionaries used by different researchers, we propose to use a weighted chord distance measure. In this measure, mistaking the root will cost much more than e.g. overlooking an added ninth and mistaking major for augmented chord will be less costly than e.g. interpreting major as minor. This should enable comparison of algorithms that work with different chord dictionaries: using a simplified dictionary will decrease the estimation accuracy, but not significantly so, and an algorithm using an accurate 24-chord dictionary should outperform a poor algorithm that deals with accurate chord labels.
+Evaluation is a difficult problem for 2 reasons:
 The following formula will be used to evaluate the performance of the submitted algorithms:

@@ Line 41: / Line 41: @@
   ...
-Each line will consist of harmony data corresponding to one line of the input data. Harmony is given by pitch class of the root tone followed by a list of all pitch classes belonging to the chord sorted by the frequencies of notes found in the data. Pitch classes are labeled with capital Latin letters (A through G) with flats represented by minus signs and sharps by plus signs. No harmony (e.g. a rest or upbeat) is marked with 'z'. So, in the example above, “E- E- G B-” means a chord with root in E♭ (which is the tonic) that consists of three notes: E♭, G and B♭, and so it is a major triad in its root position.
+Each line will consist of harmony data corresponding to one line of the input data. Harmony is given by pitch class of the root tone, followed by a list of all pitch classes belonging to the chord. Pitch classes are labeled with capital Latin letters (A through G) with flats represented by minus signs and sharps by plus signs. Non-harmony (e.g. a rest or upbeat) is marked with a special symbol 'z'. So, in the example above, “E- E- G B-” means a chord with root in E♭ (which is the tonic) that consists of three notes: E♭, G and B♭, and so it is a major triad (in its root position, although the order of the pitch classes play no role — it is straightforward to extract that from the input data).
-The tonal center (and therefore the structural meaning of a given chord) is not explicitly given in order to simplify the task and allow more participants to compete. Nevertheless, the algorithms will benefit much from considering e.g. tonal center movement and degrees of scale.
+The tonal center (and therefore the structural meaning of a given chord) is not explicitly given in order to simplify the task and allow more participants to compete. Nevertheless, the algorithms will benefit much from considering e.g. tonal center movement and degrees of scale (as in functional analysis).
 Labels for the entire RWC's classical portion will be made available for the MIREX participants (with some copying and usage restrictions). In order to allow cross-validation of the submitted algorithms, an independent set of labels will be created (to which the participants will have no access), and the algorithms will be executed on both sets.

@@ Line 1: / Line 1: @@
 == Description ==
-This task is suggested for MIREX 2010. The general goal is to estimate the latent harmonic progression from score-like music data, a process typically called harmonic analysis or functional analysis in musicology. This typically involves estimating the musical key of a given piece of music and movement of the tonal center (modulations and tonicizations), as well as the progression of chords.
+This task is suggested for MIREX 2010. The general goal is to estimate the latent harmonic progression from score-like music data, a process typically called harmonic analysis or functional analysis in musicology (although this task cannot be really called functional analysis, functional analysis can be used to obtain the required output data). This typically involves estimating the musical key of a given piece of music and movement of the tonal center (modulations and tonicizations), as well as the progression of chords.
 === Time units ===
-It is well known that harmony is strongly coupled with metric structure: progressing from one chord to another typically occurs at strong beats, while most non-chord tones are played on weak beats or off-beat. For this reason knowing the temporal structure of the piece of music is important when performing harmonic analysis, and in this task the harmony is estimated on a beat-based time basis from score-like symbolic data (as opposed to performance data).
+It is well known that harmony is strongly coupled with metric structure: progressing from one chord to another typically occurs at strong beats, while most non-chord tones are played on weak beats or off-beat. For this reason, knowing the temporal structure of the piece of music is important when performing harmonic analysis, and in this task the harmony is estimated on a beat-based time basis from score-like symbolic data (as opposed to performance data).
 === Chord dictionary ===
-Harmonic analysis can be performed with varying degree of preciseness, depending on the goal at hand. One might only want to extract general progression patterns to be used in harmonic similarity or harmonic queries, for example check if a song at hand contains the so called 50s progression: I–vi–IV–V. Another one might be interested in full musicological analysis of a score, i.e. obtaining information about root position, chord type (quality), added tones, invertion, etc. This person would for example say the the second bar of Chopin's Nocturne Op. 9 No. 2 (in the key of E♭-major) starts in the tonic I (E♭, G, B♭), then it progresses to the dominant ninth chord in the third inversion with the root removed and the tonic added in bass (E♭, A♭, D, F).
+Harmonic analysis can be performed with varying degree of precision, depending on the goal at hand. One might only want to extract general progression patterns to be used in harmonic similarity or harmonic queries, for example check if a song at hand contains the so called 50s progression: I–vi–IV–V. Another one might be interested in full musicological analysis of a score, i.e. obtaining information about root position, chord type (quality), added tones, invertion, etc. This person would for example say the the second bar of Chopin's Nocturne Op. 9 No. 2 (in the key of E♭-major) starts in the tonic I (E♭, G, B♭), then it progresses to the dominant ninth chord in the third inversion with the root removed and the tonic added in bass (E♭, A♭, D, F).
 [[File:Chopin 9 2.png||500px]]

@@ Line 122: / Line 122: @@
 == References ==
-# F. Lerdahl, “Tonal Pitch Space,” Oxford University Press, 2001
 # W. B. de Haas, R. C. Veltkamp, F. Wiering, [http://ismir2008.ismir.net/papers/ISMIR2008_252.pdf “Tonal pitch step distance: a similarity measure for chord progressions,”] Proc. of 9th ISMIR, 2008
 # J.-F. Paiement, D. Eck, S. Bengio, [http://publications.idiap.ch/downloads/reports/2005/rr-05-58.pdf “Chord representations for probabilistic models,”] IDIAP Research report, 2005
@@ Line 130: / Line 130: @@
 # C. S. Sapp, [http://www.ccarh.org/publications/cm/vol/15/cm15-06-sapp.pdf “Computational chord-root identification in symbolic musical data: rationale, methods and applications,”] Computing in Musicology 15, 2007
 # H. Taube, [http://www-camil.music.uiuc.edu/Software/cc/doc/cmj.pdf “Automatic tonal analysis: toward the implementation of a music theory workbench,”] Computer Music Journal vol. 23 nr. 4, 1999
 == Potential Participants ==
 == Discussion for 2010 ==

← Older revision		Revision as of 09:18, 15 May 2010
Line 107:		Line 107:

	where ''D'' is the distance between its arguments on the circle of fifths (minimal number of steps 0..6) and ''r'' is any chord root (0..11). ''W'' is normalized to 1.		where ''D'' is the distance between its arguments on the circle of fifths (minimal number of steps 0..6) and ''r'' is any chord root (0..11). ''W'' is normalized to 1.
		+
		+	[[File:Weights plot.png\|\|700px]]
		+
		+	The score will be equal to 1 if all the notes in the chord were identified correctly and between 0 and 1 for other cases (see pictures below).
		+
		+	[[File:Wrong_root.png\|\|400px]]
		+
		+	[[File:Wrong_type.png\|\|300px]]
		+	[[File:Wrong added.png\|\|300px]]
		+	[[File:Wrong removed.png\|\|300px]]

	== Test data ==		== Test data ==