The features implemented here are based on those found in jSymbolic and defined in Cory McKay’s MA Thesis, “Automatic Genre Classification of MIDI Recordings”
The LGPL jSymbolic system can be found here: http://jmir.sourceforge.net/jSymbolic.html
music21.features.jSymbolic.getCompletionStats()¶>>> features.jSymbolic.getCompletionStats()
completion stats: 70/111 (0.6306...)
music21.features.jSymbolic.getExtractorByTypeAndNumber(extractorType, number)¶Typical usage:
>>> t5 = features.jSymbolic.getExtractorByTypeAndNumber('T', 5)
>>> t5.__name__
'VoiceEqualityNoteDurationFeature'
>>> bachExample = corpus.parse('bach/bwv66.6')
>>> fe = t5(bachExample)
Features unimplemented in jSymbolic but documented in the dissertation return None
>>> features.jSymbolic.getExtractorByTypeAndNumber('C', 20) is None
True
Totally unknown features return an exception:
>>> features.jSymbolic.getExtractorByTypeAndNumber('L', 900)
Traceback (most recent call last):
...
JSymbolicFeatureException: Could not find any jSymbolic features of type L
>>> features.jSymbolic.getExtractorByTypeAndNumber('C', 200)
Traceback (most recent call last):
...
JSymbolicFeatureException: jSymbolic features of type C do not have number 200
You could also find all the feature extractors this way:
>>> fs = features.jSymbolic.extractorsById
>>> for k in fs:
... for i in range(len(fs[k])):
... if fs[k][i] is not None:
... n = fs[k][i].__name__
... if fs[k][i] not in features.jSymbolic.featureExtractors:
... n += " (not implemented)"
... print("%s %d %s" % (k, i, n))
D 1 OverallDynamicRangeFeature (not implemented)
D 2 VariationOfDynamicsFeature (not implemented)
D 3 VariationOfDynamicsInEachVoiceFeature (not implemented)
D 4 AverageNoteToNoteDynamicsChangeFeature (not implemented)
I 1 PitchedInstrumentsPresentFeature
I 2 UnpitchedInstrumentsPresentFeature (not implemented)
I 3 NotePrevalenceOfPitchedInstrumentsFeature
I 4 NotePrevalenceOfUnpitchedInstrumentsFeature (not implemented)
I 5 TimePrevalenceOfPitchedInstrumentsFeature (not implemented)
I 6 VariabilityOfNotePrevalenceOfPitchedInstrumentsFeature
I 7 VariabilityOfNotePrevalenceOfUnpitchedInstrumentsFeature (not implemented)
I 8 NumberOfPitchedInstrumentsFeature
I 9 NumberOfUnpitchedInstrumentsFeature (not implemented)
I 10 PercussionPrevalenceFeature (not implemented)
I 11 StringKeyboardFractionFeature
I 12 AcousticGuitarFractionFeature
I 13 ElectricGuitarFractionFeature
I 14 ViolinFractionFeature
I 15 SaxophoneFractionFeature
I 16 BrassFractionFeature
I 17 WoodwindsFractionFeature
I 18 OrchestralStringsFractionFeature
I 19 StringEnsembleFractionFeature
I 20 ElectricInstrumentFractionFeature
M 1 MelodicIntervalHistogramFeature
M 2 AverageMelodicIntervalFeature
M 3 MostCommonMelodicIntervalFeature
M 4 DistanceBetweenMostCommonMelodicIntervalsFeature
M 5 MostCommonMelodicIntervalPrevalenceFeature
M 6 RelativeStrengthOfMostCommonIntervalsFeature
M 7 NumberOfCommonMelodicIntervalsFeature
M 8 AmountOfArpeggiationFeature
M 9 RepeatedNotesFeature
M 10 ChromaticMotionFeature
M 11 StepwiseMotionFeature
M 12 MelodicThirdsFeature
M 13 MelodicFifthsFeature
M 14 MelodicTritonesFeature
M 15 MelodicOctavesFeature
M 17 DirectionOfMotionFeature
M 18 DurationOfMelodicArcsFeature
M 19 SizeOfMelodicArcsFeature
P 1 MostCommonPitchPrevalenceFeature
P 2 MostCommonPitchClassPrevalenceFeature
P 3 RelativeStrengthOfTopPitchesFeature
P 4 RelativeStrengthOfTopPitchClassesFeature
P 5 IntervalBetweenStrongestPitchesFeature
P 6 IntervalBetweenStrongestPitchClassesFeature
P 7 NumberOfCommonPitchesFeature
P 8 PitchVarietyFeature
P 9 PitchClassVarietyFeature
P 10 RangeFeature
P 11 MostCommonPitchFeature
P 12 PrimaryRegisterFeature
P 13 ImportanceOfBassRegisterFeature
P 14 ImportanceOfMiddleRegisterFeature
P 15 ImportanceOfHighRegisterFeature
P 16 MostCommonPitchClassFeature
P 17 DominantSpreadFeature (not implemented)
P 18 StrongTonalCentresFeature (not implemented)
P 19 BasicPitchHistogramFeature
P 20 PitchClassDistributionFeature
P 21 FifthsPitchHistogramFeature
P 22 QualityFeature
P 23 GlissandoPrevalenceFeature (not implemented)
P 24 AverageRangeOfGlissandosFeature (not implemented)
P 25 VibratoPrevalenceFeature (not implemented)
R 1 StrongestRhythmicPulseFeature (not implemented)
R 2 SecondStrongestRhythmicPulseFeature (not implemented)
R 3 HarmonicityOfTwoStrongestRhythmicPulsesFeature (not implemented)
R 4 StrengthOfStrongestRhythmicPulseFeature (not implemented)
R 5 StrengthOfSecondStrongestRhythmicPulseFeature (not implemented)
R 6 StrengthRatioOfTwoStrongestRhythmicPulsesFeature (not implemented)
R 7 CombinedStrengthOfTwoStrongestRhythmicPulsesFeature (not implemented)
R 8 NumberOfStrongPulsesFeature (not implemented)
R 9 NumberOfModeratePulsesFeature (not implemented)
R 10 NumberOfRelativelyStrongPulsesFeature (not implemented)
R 11 RhythmicLoosenessFeature (not implemented)
R 12 PolyrhythmsFeature (not implemented)
R 13 RhythmicVariabilityFeature (not implemented)
R 14 BeatHistogramFeature (not implemented)
R 15 NoteDensityFeature
R 17 AverageNoteDurationFeature
R 18 VariabilityOfNoteDurationFeature (not implemented)
R 19 MaximumNoteDurationFeature
R 20 MinimumNoteDurationFeature
R 21 StaccatoIncidenceFeature
R 22 AverageTimeBetweenAttacksFeature
R 23 VariabilityOfTimeBetweenAttacksFeature
R 24 AverageTimeBetweenAttacksForEachVoiceFeature
R 25 AverageVariabilityOfTimeBetweenAttacksForEachVoiceFeature
R 30 InitialTempoFeature
R 31 InitialTimeSignatureFeature
R 32 CompoundOrSimpleMeterFeature
R 33 TripleMeterFeature
R 34 QuintupleMeterFeature
R 35 ChangesOfMeterFeature
T 1 MaximumNumberOfIndependentVoicesFeature
T 2 AverageNumberOfIndependentVoicesFeature
T 3 VariabilityOfNumberOfIndependentVoicesFeature
T 4 VoiceEqualityNumberOfNotesFeature (not implemented)
T 5 VoiceEqualityNoteDurationFeature (not implemented)
T 6 VoiceEqualityDynamicsFeature (not implemented)
T 7 VoiceEqualityMelodicLeapsFeature (not implemented)
T 8 VoiceEqualityRangeFeature (not implemented)
T 9 ImportanceOfLoudestVoiceFeature (not implemented)
T 10 RelativeRangeOfLoudestVoiceFeature (not implemented)
T 12 RangeOfHighestLineFeature (not implemented)
T 13 RelativeNoteDensityOfHighestLineFeature (not implemented)
T 15 MelodicIntervalsInLowestLineFeature (not implemented)
T 20 VoiceSeparationFeature (not implemented)
music21.features.jSymbolic.AcousticGuitarFractionFeature(dataOrStream=None, *arguments, **keywords)¶A feature exractor that extracts the fraction of all Note Ons belonging to acoustic guitar patches (General MIDI patches 25 and 26).
>>> s1 = stream.Stream()
>>> s1.append(instrument.AcousticGuitar())
>>> s1.repeatAppend(note.Note(), 3)
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.AcousticGuitarFractionFeature(s1)
>>> fe.extract().vector
[0.75]
AcousticGuitarFractionFeature bases
AcousticGuitarFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AmountOfArpeggiationFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of horizontal intervals that are repeated notes, minor thirds, major thirds, perfect fifths, minor sevenths, major sevenths, octaves, minor tenths or major tenths.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AmountOfArpeggiationFeature(s)
>>> f = fe.extract()
>>> f.name
'Amount of Arpeggiation'
>>> f.vector
[0.333...]
AmountOfArpeggiationFeature bases
AmountOfArpeggiationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageMelodicIntervalFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageMelodicIntervalFeature(s)
>>> f = fe.extract()
>>> f.vector
[2.44...]
AverageMelodicIntervalFeature bases
AverageMelodicIntervalFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageNoteDurationFeature(dataOrStream=None, *arguments, **keywords)¶Average duration of notes in seconds.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageNoteDurationFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.441717...]
>>> s.insert(0, tempo.MetronomeMark(number=240))
>>> fe = features.jSymbolic.AverageNoteDurationFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.220858...]
AverageNoteDurationFeature bases
AverageNoteDurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageNoteToNoteDynamicsChangeFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
Average change of loudness from one note to the next note in the same channel (in MIDI velocity units).
TODO: implement
AverageNoteToNoteDynamicsChangeFeature bases
AverageNoteToNoteDynamicsChangeFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageNumberOfIndependentVoicesFeature(dataOrStream=None, *arguments, **keywords)¶Average number of different channels in which notes have sounded simultaneously. Rests are not included in this calculation. Here, Parts are treated as voices
>>> s = corpus.parse('handel/rinaldo/lascia_chio_pianga')
>>> fe = features.jSymbolic.AverageNumberOfIndependentVoicesFeature(s)
>>> f = fe.extract()
>>> f.vector
[2.1...]
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageNumberOfIndependentVoicesFeature(s)
>>> f = fe.extract()
>>> f.vector
[3.96...]
AverageNumberOfIndependentVoicesFeature bases
AverageNumberOfIndependentVoicesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageRangeOfGlissandosFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented in music21
Average range of MIDI Pitch Bends, where “range” is defined as the greatest value of the absolute difference between 64 and the second data byte of all MIDI Pitch Bend messages falling between the Note On and Note Off messages of any note
AverageRangeOfGlissandosFeature bases
AverageRangeOfGlissandosFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageTimeBetweenAttacksFeature(dataOrStream=None, *arguments, **keywords)¶Average time in seconds between Note On events (regardless of channel).
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageTimeBetweenAttacksFeature(s)
>>> f = fe.extract()
>>> print(round(f.vector[0], 2))
0.35
AverageTimeBetweenAttacksFeature bases
AverageTimeBetweenAttacksFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageTimeBetweenAttacksForEachVoiceFeature(dataOrStream=None, *arguments, **keywords)¶Average of average times in seconds between Note On events on individual channels that contain at least one note.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageTimeBetweenAttacksForEachVoiceFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.4428...]
AverageTimeBetweenAttacksForEachVoiceFeature bases
AverageTimeBetweenAttacksForEachVoiceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.AverageVariabilityOfTimeBetweenAttacksForEachVoiceFeature(dataOrStream=None, *arguments, **keywords)¶Average standard deviation, in seconds, of time between Note On events on individual channels that contain at least one note.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.AverageVariabilityOfTimeBetweenAttacksForEachVoiceFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.1773926...]
AverageVariabilityOfTimeBetweenAttacksForEachVoiceFeature bases
AverageVariabilityOfTimeBetweenAttacksForEachVoiceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.BasicPitchHistogramFeature(dataOrStream=None, *arguments, **keywords)¶A feature exractor that finds a features array with bins corresponding to the values of the basic pitch histogram.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.BasicPitchHistogramFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.052631578..., 0.0, 0.0, 0.052631578...,
0.05263157894..., 0.2631578..., 0.0, 0.3157894..., 0.1052631...,
0.0, 0.052631..., 0.157894736..., 0.5263157..., 0.0, 0.368421052...,
0.6315789473..., 0.105263157..., 0.78947368..., 0.0, 1.0, 0.52631578...,
0.052631578..., 0.736842105..., 0.1578947..., 0.9473684..., 0.0,
0.36842105..., 0.47368421..., 0.0, 0.42105263..., 0.0, 0.36842105...,
0.0, 0.0, 0.052631578...,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
TODO: Better doctest...
BasicPitchHistogramFeature bases
BasicPitchHistogramFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.BeatHistogramFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
A feature exractor that finds a feature array with entries corresponding to the frequency values of each of the bins of the beat histogram (except the first 40 empty ones).
BeatHistogramFeature bases
BeatHistogramFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.BrassFractionFeature(dataOrStream=None, *arguments, **keywords)¶A feature exractor that extracts the fraction of all Note Ons belonging to brass patches (General MIDI patches 57 through 68).
TODO: Conflict in source: only does 57-62?
>>> s1 = stream.Stream()
>>> s1.append(instrument.SopranoSaxophone())
>>> s1.repeatAppend(note.Note(), 6)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 4)
>>> fe = features.jSymbolic.BrassFractionFeature(s1)
>>> print(fe.extract().vector[0])
0.4
BrassFractionFeature bases
BrassFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ChangesOfMeterFeature(dataOrStream=None, *arguments, **keywords)¶A feature exractor that sets the feature to 1 if the time signature is changed one or more times during the recording.
>>> s1 = stream.Stream()
>>> s1.append(meter.TimeSignature('3/4'))
>>> s2 = stream.Stream()
>>> s2.append(meter.TimeSignature('3/4'))
>>> s2.append(meter.TimeSignature('4/4'))
>>> fe = features.jSymbolic.ChangesOfMeterFeature(s1)
>>> fe.extract().vector
[0]
>>> fe.setData(s2) # change the data
>>> fe.extract().vector
[1]
ChangesOfMeterFeature bases
ChangesOfMeterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ChromaticMotionFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals corresponding to a semitone.
ChromaticMotionFeature bases
ChromaticMotionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.CombinedStrengthOfTwoStrongestRhythmicPulsesFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
The sum of the frequencies of the two beat bins of the peaks with the highest frequencies.
CombinedStrengthOfTwoStrongestRhythmicPulsesFeature bases
CombinedStrengthOfTwoStrongestRhythmicPulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.CompoundOrSimpleMeterFeature(dataOrStream=None, *arguments, **keywords)¶Set to 1 if the initial meter is compound (numerator of time signature is greater than or equal to 6 and is evenly divisible by 3) and to 0 if it is simple (if the above condition is not fulfilled).
>>> s1 = stream.Stream()
>>> s1.append(meter.TimeSignature('3/4'))
>>> s2 = stream.Stream()
>>> s2.append(meter.TimeSignature('9/8'))
>>> fe = features.jSymbolic.CompoundOrSimpleMeterFeature(s1)
>>> fe.extract().vector
[0]
>>> fe.setData(s2) # change the data
>>> fe.extract().vector
[1]
CompoundOrSimpleMeterFeature bases
CompoundOrSimpleMeterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.DirectionOfMotionFeature(dataOrStream=None, *arguments, **keywords)¶Returns the fraction of melodic intervals that are rising rather than falling. Unisons are omitted.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.DirectionOfMotionFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.47...]
DirectionOfMotionFeature bases
DirectionOfMotionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.DistanceBetweenMostCommonMelodicIntervalsFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.DistanceBetweenMostCommonMelodicIntervalsFeature(s)
>>> f = fe.extract()
>>> f.vector
[1]
DistanceBetweenMostCommonMelodicIntervalsFeature bases
DistanceBetweenMostCommonMelodicIntervalsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.DominantSpreadFeature(dataOrStream=None, *arguments, **keywords)¶Largest number of consecutive pitch classes separated by perfect 5ths that accounted for at least 9% each of the notes.
DominantSpreadFeature bases
DominantSpreadFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.DurationFeature(dataOrStream=None, *arguments, **keywords)¶A feature extractor that extracts the duration in seconds.
DurationFeature bases
DurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.DurationOfMelodicArcsFeature(dataOrStream=None, *arguments, **keywords)¶Average number of notes that separate melodic peaks and troughs in any channel.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.DurationOfMelodicArcsFeature(s)
>>> f = fe.extract()
>>> f.vector
[10.28...]
DurationOfMelodicArcsFeature bases
DurationOfMelodicArcsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ElectricGuitarFractionFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.ElectricGuitar())
>>> s1.repeatAppend(note.Note(), 4)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 4)
>>> fe = features.jSymbolic.ElectricGuitarFractionFeature(s1)
>>> fe.extract().vector
[0.5]
ElectricGuitarFractionFeature bases
ElectricGuitarFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ElectricInstrumentFractionFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.ElectricOrgan())
>>> s1.repeatAppend(note.Note(), 8)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 2)
>>> fe = features.jSymbolic.ElectricInstrumentFractionFeature(s1)
>>> print(fe.extract().vector[0])
0.8
ElectricInstrumentFractionFeature bases
ElectricInstrumentFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.FifthsPitchHistogramFeature(dataOrStream=None, *arguments, **keywords)¶A feature array with bins corresponding to the values of the 5ths pitch class histogram.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.FifthsPitchHistogramFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.0, 0.0, 0.375, 0.6875, 0.5, 0.875, 0.90625, 1.0, 0.4375, 0.03125, 0.09375, 0.1875]
FifthsPitchHistogramFeature bases
FifthsPitchHistogramFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.GlissandoPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented in music21
Number of Note Ons that have at least one MIDI Pitch Bend associated with them divided by total number of pitched Note Ons.
GlissandoPrevalenceFeature bases
GlissandoPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.HarmonicityOfTwoStrongestRhythmicPulsesFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented.
The bin label of the higher (in terms of bin label) of the two beat bins of the peaks with the highest frequency divided by the bin label of the lower.
HarmonicityOfTwoStrongestRhythmicPulsesFeature bases
HarmonicityOfTwoStrongestRhythmicPulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ImportanceOfBassRegisterFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of Notes between MIDI pitches 0 and 54.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.ImportanceOfBassRegisterFeature(s)
>>> fe.extract().vector
[0.18...]
ImportanceOfBassRegisterFeature bases
ImportanceOfBassRegisterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ImportanceOfHighRegisterFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of Notes between MIDI pitches 73 and 127.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.ImportanceOfHighRegisterFeature(s)
>>> fe.extract().vector
[0.049...]
ImportanceOfHighRegisterFeature bases
ImportanceOfHighRegisterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ImportanceOfLoudestVoiceFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
ImportanceOfLoudestVoiceFeature bases
ImportanceOfLoudestVoiceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ImportanceOfMiddleRegisterFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of Notes between MIDI pitches 55 and 72
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.ImportanceOfMiddleRegisterFeature(s)
>>> fe.extract().vector
[0.766...]
ImportanceOfMiddleRegisterFeature bases
ImportanceOfMiddleRegisterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.InitialTempoFeature(dataOrStream=None, *arguments, **keywords)¶Tempo in beats per minute at the start of the recording.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.InitialTempoFeature(s)
>>> f = fe.extract()
>>> f.vector # a default
[120.0]
InitialTempoFeature bases
InitialTempoFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.InitialTimeSignatureFeature(dataOrStream=None, *arguments, **keywords)¶A feature array with two elements. The first is the numerator of the first occurring time signature and the second is the denominator of the first occurring time signature. Both are set to 0 if no time signature is present.
>>> s1 = stream.Stream()
>>> s1.append(meter.TimeSignature('3/4'))
>>> fe = features.jSymbolic.InitialTimeSignatureFeature(s1)
>>> fe.extract().vector
[3, 4]
InitialTimeSignatureFeature bases
InitialTimeSignatureFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.InstrumentFractionFeature(dataOrStream=None, *arguments, **keywords)¶This subclass is in-turn subclassed by all FeatureExtractors that look at the proportional usage of an Insutrment
InstrumentFractionFeature bases
InstrumentFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.IntervalBetweenStrongestPitchClassesFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.IntervalBetweenStrongestPitchClassesFeature(s)
>>> fe.extract().vector
[5]
IntervalBetweenStrongestPitchClassesFeature bases
IntervalBetweenStrongestPitchClassesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.IntervalBetweenStrongestPitchesFeature(dataOrStream=None, *arguments, **keywords)¶Absolute value of the difference between the pitches of the two most common MIDI pitches.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.IntervalBetweenStrongestPitchesFeature(s)
>>> fe.extract().vector
[5]
IntervalBetweenStrongestPitchesFeature bases
IntervalBetweenStrongestPitchesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MaximumNoteDurationFeature(dataOrStream=None, *arguments, **keywords)¶Duration of the longest note (in seconds).
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MaximumNoteDurationFeature(s)
>>> f = fe.extract()
>>> f.vector
[1.0]
MaximumNoteDurationFeature bases
MaximumNoteDurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MaximumNumberOfIndependentVoicesFeature(dataOrStream=None, *arguments, **keywords)¶Maximum number of different channels in which notes have sounded simultaneously.
Here, Parts are treated as channels.
>>> s = corpus.parse('handel/rinaldo/lascia_chio_pianga')
>>> fe = features.jSymbolic.MaximumNumberOfIndependentVoicesFeature(s)
>>> f = fe.extract()
>>> f.vector
[3]
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MaximumNumberOfIndependentVoicesFeature(s)
>>> f = fe.extract()
>>> f.vector
[4]
MaximumNumberOfIndependentVoicesFeature bases
MaximumNumberOfIndependentVoicesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicFifthsFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that are perfect fifths
MelodicFifthsFeature bases
MelodicFifthsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicIntervalHistogramFeature(dataOrStream=None, *arguments, **keywords)¶A features array with bins corresponding to the values of the melodic interval histogram.
128 dimensions
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MelodicIntervalHistogramFeature(s)
>>> f = fe.extract()
>>> f.vector[0:5]
[0.14..., 0.22..., 0.36..., 0.06..., 0.05...]
MelodicIntervalHistogramFeature bases
MelodicIntervalHistogramFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicIntervalsInLowestLineFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
MelodicIntervalsInLowestLineFeature bases
MelodicIntervalsInLowestLineFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicOctavesFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that are octaves
MelodicOctavesFeature bases
MelodicOctavesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicThirdsFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that are major or minor thirds
MelodicThirdsFeature bases
MelodicThirdsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MelodicTritonesFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that are tritones
MelodicTritonesFeature bases
MelodicTritonesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MinimumNoteDurationFeature(dataOrStream=None, *arguments, **keywords)¶Duration of the shortest note (in seconds).
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MinimumNoteDurationFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.25]
MinimumNoteDurationFeature bases
MinimumNoteDurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonMelodicIntervalFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonMelodicIntervalFeature(s)
>>> f = fe.extract()
>>> f.vector
[2]
MostCommonMelodicIntervalFeature bases
MostCommonMelodicIntervalFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonMelodicIntervalPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that belong to the most common interval.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonMelodicIntervalPrevalenceFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.364...]
MostCommonMelodicIntervalPrevalenceFeature bases
MostCommonMelodicIntervalPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonPitchClassFeature(dataOrStream=None, *arguments, **keywords)¶Bin label of the most common pitch class.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonPitchClassFeature(s)
>>> fe.extract().vector
[1]
MostCommonPitchClassFeature bases
MostCommonPitchClassFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonPitchClassPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of Notes corresponding to the most common pitch class.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonPitchClassPrevalenceFeature(s)
>>> fe.extract().vector
[0.19...]
MostCommonPitchClassPrevalenceFeature bases
MostCommonPitchClassPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonPitchFeature(dataOrStream=None, *arguments, **keywords)¶Bin label of the most common pitch divided by the number of possible pitches.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonPitchFeature(s)
>>> fe.extract().vector
[0.47...]
MostCommonPitchFeature bases
MostCommonPitchFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.MostCommonPitchPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of Notes corresponding to the most common pitch.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.MostCommonPitchPrevalenceFeature(s)
>>> fe.extract().vector[0]
0.11...
MostCommonPitchPrevalenceFeature bases
MostCommonPitchPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NoteDensityFeature(dataOrStream=None, *arguments, **keywords)¶Gives the Average number of notes per second, taking into account the tempo at any moment in the piece. N.B. unlike the jSymbolic version, music21’s Feature Extraction methods can run on a subset of the entire piece (measures, certain parts, etc.). However, unlike jSymbolic, music21 quantizes notes from midi somewhat before running this test, so it is better run on encoded midi scores than recorded midi performances.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.NoteDensityFeature(s)
>>> f = fe.extract()
>>> f.vector
[12.368421...]
NoteDensityFeature bases
NoteDensityFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NotePrevalenceOfPitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.AcousticGuitar())
>>> s1.repeatAppend(note.Note(), 4)
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.NotePrevalenceOfPitchedInstrumentsFeature(s1)
>>> fe.extract().vector
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0.8..., 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.2...,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
NotePrevalenceOfPitchedInstrumentsFeature bases
NotePrevalenceOfPitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NotePrevalenceOfUnpitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
NotePrevalenceOfUnpitchedInstrumentsFeature bases
NotePrevalenceOfUnpitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfCommonMelodicIntervalsFeature(dataOrStream=None, *arguments, **keywords)¶Number of melodic intervals that represent at least 9% of all melodic intervals.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.NumberOfCommonMelodicIntervalsFeature(s)
>>> f = fe.extract()
>>> f.vector
[3]
NumberOfCommonMelodicIntervalsFeature bases
NumberOfCommonMelodicIntervalsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfCommonPitchesFeature(dataOrStream=None, *arguments, **keywords)¶Number of pitches that account individually for at least 9% of all notes.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.NumberOfCommonPitchesFeature(s)
>>> fe.extract().vector
[3]
NumberOfCommonPitchesFeature bases
NumberOfCommonPitchesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfModeratePulsesFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Number of beat peaks with normalized frequencies over 0.01.
NumberOfModeratePulsesFeature bases
NumberOfModeratePulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfPitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.AcousticGuitar())
>>> s1.append(note.Note())
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.NumberOfPitchedInstrumentsFeature(s1)
>>> fe.extract().vector
[2]
NumberOfPitchedInstrumentsFeature bases
NumberOfPitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfRelativelyStrongPulsesFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented
Number of beat peaks with frequencies at least 30% as high as the frequency of the bin with the highest frequency.
NumberOfRelativelyStrongPulsesFeature bases
NumberOfRelativelyStrongPulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfStrongPulsesFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Number of beat peaks with normalized frequencies over 0.1.
NumberOfStrongPulsesFeature bases
NumberOfStrongPulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.NumberOfUnpitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
Number of distinct MIDI Percussion Key Map patches that were used to play at least one note. It should be noted that only instruments 35 to 81 are included here, as they are the ones that are included in the official standard.
TODO: implement
NumberOfUnpitchedInstrumentsFeature bases
NumberOfUnpitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.OrchestralStringsFractionFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.Violoncello())
>>> s1.repeatAppend(note.Note(), 4)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 6)
>>> fe = features.jSymbolic.OrchestralStringsFractionFeature(s1)
>>> print(fe.extract().vector[0])
0.4
OrchestralStringsFractionFeature bases
OrchestralStringsFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.OverallDynamicRangeFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
The maximum loudness minus the minimum loudness value.
TODO: implement
OverallDynamicRangeFeature bases
OverallDynamicRangeFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PercussionPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
PercussionPrevalenceFeature bases
PercussionPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PitchClassDistributionFeature(dataOrStream=None, *arguments, **keywords)¶A feature array with 12 entries where the first holds the frequency of the bin of the pitch class histogram with the highest frequency, and the following entries holding the successive bins of the histogram, wrapping around if necessary.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.PitchClassDistributionFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.0, 1.0, 0.375, 0.03125, 0.5, 0.1875, 0.90625, 0.0, 0.4375, 0.6875, 0.09375, 0.875]
PitchClassDistributionFeature bases
PitchClassDistributionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PitchClassVarietyFeature(dataOrStream=None, *arguments, **keywords)¶Number of pitch classes used at least once.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.PitchClassVarietyFeature(s)
>>> fe.extract().vector
[10]
PitchClassVarietyFeature bases
PitchClassVarietyFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PitchVarietyFeature(dataOrStream=None, *arguments, **keywords)¶Number of pitches used at least once.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.PitchVarietyFeature(s)
>>> fe.extract().vector
[24]
PitchVarietyFeature bases
PitchVarietyFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PitchedInstrumentsPresentFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.AcousticGuitar())
>>> s1.append(note.Note())
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.PitchedInstrumentsPresentFeature(s1)
>>> fe.extract().vector
[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
PitchedInstrumentsPresentFeature bases
PitchedInstrumentsPresentFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PolyrhythmsFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Number of beat peaks with frequencies at least 30% of the highest frequency whose bin labels are not integer multiples or factors (using only multipliers of 1, 2, 3, 4, 6 and 8) (with an accepted error of +/- 3 bins) of the bin label of the peak with the highest frequency. This number is then divided by the total number of beat bins with frequencies over 30% of the highest frequency.
PolyrhythmsFeature bases
PolyrhythmsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.PrimaryRegisterFeature(dataOrStream=None, *arguments, **keywords)¶Average MIDI pitch.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.PrimaryRegisterFeature(s)
>>> fe.extract().vector
[58.58...]
PrimaryRegisterFeature bases
PrimaryRegisterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.QualityFeature(dataOrStream=None, *arguments, **keywords)¶Set to 0 if the key signature indicates that a recording is major, set to 1 if it indicates that it is minor. In jSymbolic, this is set to 0 if key signature is unknown.
See features.native.QualityFeature for a music21 improvement on this method
Example: Handel, Rinaldo Aria (musicxml) is explicitly encoded as being in Major:
>>> s = corpus.parse('handel/rinaldo/lascia_chio_pianga')
>>> fe = features.jSymbolic.QualityFeature(s)
>>> f = fe.extract()
>>> f.vector
[0]
QualityFeature bases
QualityFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.QuintupleMeterFeature(dataOrStream=None, *arguments, **keywords)¶Set to 1 if numerator of initial time signature is 5, set to 0 otherwise.
>>> s1 = stream.Stream()
>>> s1.append(meter.TimeSignature('5/4'))
>>> s2 = stream.Stream()
>>> s2.append(meter.TimeSignature('3/4'))
>>> fe = features.jSymbolic.QuintupleMeterFeature(s1)
>>> fe.extract().vector
[1]
>>> fe.setData(s2) # change the data
>>> fe.extract().vector
[0]
QuintupleMeterFeature bases
QuintupleMeterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RangeFeature(dataOrStream=None, *arguments, **keywords)¶Difference between highest and lowest pitches. In semitones
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.RangeFeature(s)
>>> fe.extract().vector
[34]
RangeFeature bases
RangeFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RangeOfHighestLineFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
RangeOfHighestLineFeature bases
RangeOfHighestLineFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RelativeNoteDensityOfHighestLineFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
RelativeNoteDensityOfHighestLineFeature bases
RelativeNoteDensityOfHighestLineFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RelativeRangeOfLoudestVoiceFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
RelativeRangeOfLoudestVoiceFeature bases
RelativeRangeOfLoudestVoiceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RelativeStrengthOfMostCommonIntervalsFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.RelativeStrengthOfMostCommonIntervalsFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.60...]
RelativeStrengthOfMostCommonIntervalsFeature bases
RelativeStrengthOfMostCommonIntervalsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RelativeStrengthOfTopPitchClassesFeature(dataOrStream=None, *arguments, **keywords)¶>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.RelativeStrengthOfTopPitchClassesFeature(s)
>>> fe.extract().vector
[0.90...]
RelativeStrengthOfTopPitchClassesFeature bases
RelativeStrengthOfTopPitchClassesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RelativeStrengthOfTopPitchesFeature(dataOrStream=None, *arguments, **keywords)¶The frequency of the 2nd most common pitch divided by the frequency of the most common pitch.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.RelativeStrengthOfTopPitchesFeature(s)
>>> fe.extract().vector
[0.94...]
RelativeStrengthOfTopPitchesFeature bases
RelativeStrengthOfTopPitchesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RepeatedNotesFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of notes that are repeated melodically
RepeatedNotesFeature bases
RepeatedNotesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RhythmicLoosenessFeature(dataOrStream=None, *arguments, **keywords)¶TODO: implement
Average width of beat histogram peaks (in beats per minute). Width is measured for all peaks with frequencies at least 30% as high as the highest peak, and is defined by the distance between the points on the peak in question that are 30% of the height of the peak.
RhythmicLoosenessFeature bases
RhythmicLoosenessFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.RhythmicVariabilityFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Standard deviation of the bin values (except the first 40 empty ones).
RhythmicVariabilityFeature bases
RhythmicVariabilityFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.SaxophoneFractionFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.SopranoSaxophone())
>>> s1.repeatAppend(note.Note(), 6)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 4)
>>> fe = features.jSymbolic.SaxophoneFractionFeature(s1)
>>> print(fe.extract().vector[0])
0.6
SaxophoneFractionFeature bases
SaxophoneFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.SecondStrongestRhythmicPulseFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented
Bin label of the beat bin of the peak with the second highest frequency.
SecondStrongestRhythmicPulseFeature bases
SecondStrongestRhythmicPulseFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.SizeOfMelodicArcsFeature(dataOrStream=None, *arguments, **keywords)¶Average melodic interval separating the top note of melodic peaks and the bottom note of melodic troughs.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.SizeOfMelodicArcsFeature(s)
>>> f = fe.extract()
>>> f.vector
[14.5]
SizeOfMelodicArcsFeature bases
SizeOfMelodicArcsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StaccatoIncidenceFeature(dataOrStream=None, *arguments, **keywords)¶Number of notes with durations of less than a 10th of a second divided by the total number of notes in the recording.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.StaccatoIncidenceFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.0]
StaccatoIncidenceFeature bases
StaccatoIncidenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StepwiseMotionFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of melodic intervals that corresponded to a minor or major second
StepwiseMotionFeature bases
StepwiseMotionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StrengthOfSecondStrongestRhythmicPulseFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented
Frequency of the beat bin of the peak with the second highest frequency.
StrengthOfSecondStrongestRhythmicPulseFeature bases
StrengthOfSecondStrongestRhythmicPulseFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StrengthOfStrongestRhythmicPulseFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented
Frequency of the beat bin with the highest frequency.
StrengthOfStrongestRhythmicPulseFeature bases
StrengthOfStrongestRhythmicPulseFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StrengthRatioOfTwoStrongestRhythmicPulsesFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
The frequency of the higher (in terms of frequency) of the two beat bins corresponding to the peaks with the highest frequency divided by the frequency of the lower.
StrengthRatioOfTwoStrongestRhythmicPulsesFeature bases
StrengthRatioOfTwoStrongestRhythmicPulsesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StringEnsembleFractionFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of all Note Ons belonging to string ensemble patches(General MIDI patches 49 to 52)
StringEnsembleFractionFeature bases
StringEnsembleFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StringKeyboardFractionFeature(dataOrStream=None, *arguments, **keywords)¶Fraction of all Note Ons belonging to string keyboard patches (GeneralMIDI patches 1 to 8).
>>> s1 = stream.Stream()
>>> s1.append(instrument.Piano())
>>> s1.repeatAppend(note.Note(), 9)
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.StringKeyboardFractionFeature(s1)
>>> fe.extract().vector
[0.9...]
StringKeyboardFractionFeature bases
StringKeyboardFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StrongTonalCentresFeature(dataOrStream=None, *arguments, **keywords)¶Number of peaks in the fifths pitch histogram that each account for at least 9% of all Note Ons.
StrongTonalCentresFeature bases
StrongTonalCentresFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.StrongestRhythmicPulseFeature(dataOrStream=None, *arguments, **keywords)¶not yet implemented
Bin label of the beat bin with the highest frequency.
StrongestRhythmicPulseFeature bases
StrongestRhythmicPulseFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.TimePrevalenceOfPitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
The fraction of the total time of the recording in which a note was sounding for each (pitched) General MIDI Instrument. There is one entry for each instrument, which is set to the total time in seconds during which a given instrument was sounding one or more notes divided by the total length in seconds of the piece.’
TODO: implement
TimePrevalenceOfPitchedInstrumentsFeature bases
TimePrevalenceOfPitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.TripleMeterFeature(dataOrStream=None, *arguments, **keywords)¶Set to 1 if numerator of initial time signature is 3, set to 0 otherwise.
>>> s1 = stream.Stream()
>>> s1.append(meter.TimeSignature('5/4'))
>>> s2 = stream.Stream()
>>> s2.append(meter.TimeSignature('3/4'))
>>> fe = features.jSymbolic.TripleMeterFeature(s1)
>>> fe.extract().vector
[0]
>>> fe.setData(s2) # change the data
>>> fe.extract().vector
[1]
TripleMeterFeature bases
TripleMeterFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.UnpitchedInstrumentsPresentFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Which unpitched MIDI Percussion Key Map instruments are present. There is one entry for each instrument, which is set to 1.0 if there is at least one Note On in the recording corresponding to the instrument and to 0.0 if there is not. It should be noted that only instruments 35 to 81 are included here, as they are the ones that meet the official standard. They are numbered in this array from 0 to 46.
TODO: implement
UnpitchedInstrumentsPresentFeature bases
UnpitchedInstrumentsPresentFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariabilityOfNoteDurationFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented
Standard deviation of note durations in seconds.
VariabilityOfNoteDurationFeature bases
VariabilityOfNoteDurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariabilityOfNotePrevalenceOfPitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶Standard deviation of the fraction of Note Ons played by each (pitched) General MIDI instrument that is used to play at least one note.
>>> s1 = stream.Stream()
>>> s1.append(instrument.AcousticGuitar())
>>> s1.repeatAppend(note.Note(), 5)
>>> s1.append(instrument.Tuba())
>>> s1.append(note.Note())
>>> fe = features.jSymbolic.VariabilityOfNotePrevalenceOfPitchedInstrumentsFeature(s1)
>>> fe.extract().vector
[0.33333...]
VariabilityOfNotePrevalenceOfPitchedInstrumentsFeature bases
VariabilityOfNotePrevalenceOfPitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariabilityOfNotePrevalenceOfUnpitchedInstrumentsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
Standard deviation of the fraction of Note Ons played by each (unpitched) MIDI Percussion Key Map instrument that is used to play at least one note. It should be noted that only instruments 35 to 81 are included here, as they are the ones that are included in the official standard.
TODO: implement
VariabilityOfNotePrevalenceOfUnpitchedInstrumentsFeature bases
VariabilityOfNotePrevalenceOfUnpitchedInstrumentsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariabilityOfNumberOfIndependentVoicesFeature(dataOrStream=None, *arguments, **keywords)¶Standard deviation of number of different channels in which notes have sounded simultaneously. Rests are not included in this calculation.
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.VariabilityOfNumberOfIndependentVoicesFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.19...]
VariabilityOfNumberOfIndependentVoicesFeature bases
VariabilityOfNumberOfIndependentVoicesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariabilityOfTimeBetweenAttacksFeature(dataOrStream=None, *arguments, **keywords)¶Standard deviation of the times, in seconds, between Note On events (regardless of channel).
>>> s = corpus.parse('bwv66.6')
>>> fe = features.jSymbolic.VariabilityOfTimeBetweenAttacksFeature(s)
>>> f = fe.extract()
>>> f.vector
[0.15000...]
VariabilityOfTimeBetweenAttacksFeature bases
VariabilityOfTimeBetweenAttacksFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariationOfDynamicsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
Standard deviation of loudness levels of all notes.
TODO: implement
VariationOfDynamicsFeature bases
VariationOfDynamicsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VariationOfDynamicsInEachVoiceFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
The average of the standard deviations of loudness levels within each channel that contains at least one note.
TODO: implement
VariationOfDynamicsInEachVoiceFeature bases
VariationOfDynamicsInEachVoiceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VibratoPrevalenceFeature(dataOrStream=None, *arguments, **keywords)¶Not yet implemented in music21
Number of notes for which Pitch Bend messages change direction at least twice divided by total number of notes that have Pitch Bend messages associated with them.
VibratoPrevalenceFeature bases
VibratoPrevalenceFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.ViolinFractionFeature(dataOrStream=None, *arguments, **keywords)¶>>> s1 = stream.Stream()
>>> s1.append(instrument.Violin())
>>> s1.repeatAppend(note.Note(), 2)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 8)
>>> fe = features.jSymbolic.ViolinFractionFeature(s1)
>>> fe.extract().vector
[0.2...]
ViolinFractionFeature bases
ViolinFractionFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceEqualityDynamicsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
VoiceEqualityDynamicsFeature bases
VoiceEqualityDynamicsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceEqualityMelodicLeapsFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
VoiceEqualityMelodicLeapsFeature bases
VoiceEqualityMelodicLeapsFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceEqualityNoteDurationFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
VoiceEqualityNoteDurationFeature bases
VoiceEqualityNoteDurationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceEqualityNumberOfNotesFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
Standard deviation of the total number of Note Ons in each channel that contains at least one note.
VoiceEqualityNumberOfNotesFeature bases
VoiceEqualityNumberOfNotesFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceEqualityRangeFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
VoiceEqualityRangeFeature bases
VoiceEqualityRangeFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.VoiceSeparationFeature(dataOrStream=None, *arguments, **keywords)¶Not implemented
TODO: implement
VoiceSeparationFeature bases
VoiceSeparationFeature methods
Methods inherited from FeatureExtractor:
music21.features.jSymbolic.WoodwindsFractionFeature(dataOrStream=None, *arguments, **keywords)¶TODO: Conflict in source: does 69-79?
>>> s1 = stream.Stream()
>>> s1.append(instrument.Flute())
>>> s1.repeatAppend(note.Note(), 3)
>>> s1.append(instrument.Tuba())
>>> s1.repeatAppend(note.Note(), 7)
>>> fe = features.jSymbolic.WoodwindsFractionFeature(s1)
>>> print(fe.extract().vector[0])
0.3
WoodwindsFractionFeature bases
WoodwindsFractionFeature methods
Methods inherited from FeatureExtractor: