Melody Features Summary

This table provides a comprehensive overview of all 282 melody features available in this package.

All Domains Pitch Rhythm Pitch & Rhythm All Categories Absolute Pitch Complexity Contour Corpus Expectation Inter-Onset Interval Lexical Diversity Metre Pitch Class Pitch Interval Timing Tonality All Implementations FANTASTIC IDyOM MIDI Toolbox MUST Novel Partitura SIMILE jSymbolic All Types Descriptor Sequence 0 features

Name	Pre-existing Implementations	Further References	Description	Type	Notes
Absolute Interval Range	FANTASTIC		The range between the largest and smallest absolute interval size.	Descriptor
Amount Of Arpeggiation	jSymbolic		The proportion of pitch intervals in the melody that constitute triadic movements.	Descriptor
Amount Of Staccato	jSymbolic		The proportion of notes with a duration shorter than 0.1 seconds.	Descriptor	Though this feature is named Amount Of Staccato, it is a fixed-duration cutoff statistic rather than symbolic articulation parsing.
Asym Index	MUST		The vertical mirror asymmetry index of a melody. Mirrors the MUST ``asymIndex`` implementation: the proportion of sampled time points at which pitch differs from its temporally mirrored counterpart.	Descriptor	In the MUST/MIDI Toolbox notematrix, onset (column 1) and duration (column 2) are in beats. The reference MATLAB code samples with `for t = 0:0.0001:T using that same unit. There is no separate millisecond grid. This implementation converts Melody` timing to beats and uses the same 0.0001 beat step. The real-time length of one step depends on tempo (e.g. 0.1 ms only at 60 BPM, 0.05 ms at 120 BPM).
Asym Total	MUST		The total vertical mirror asymmetry of a melody. Mirrors the MUST ``asymTotal`` implementation: build a sustained-pitch time series, compare each sample to its temporally reversed counterpart, and return the mean absolute pitch difference.	Descriptor	In the MUST/MIDI Toolbox notematrix, onset (column 1) and duration (column 2) are in beats. The reference MATLAB code samples with `for t = 0:0.0001:T using that same unit. There is no separate millisecond grid. This implementation converts Melody` timing to beats and uses the same 0.0001 beat step. The real-time length of one step depends on tempo (e.g. 0.1 ms only at 60 BPM, 0.05 ms at 120 BPM).
Av Abs Interval	MUST		The mean log-transformed absolute melodic interval size. Computed as the mean of ``log(abs(interval) + 1)`` over consecutive pitch pairs, where intervals are measured in semitones.	Descriptor
Av Local P1 Entropy	MUST		The average local zeroth-order pitch entropy across a melody. Computes Shannon entropy of the pitch distribution within sliding one-second windows advanced in 0.25-second steps, using an inclusive upper onset bound, then returns the mean entropy across windows.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
Average Interval Span By Melodic Arcs	jSymbolic		The average interval span of melodic arcs.	Descriptor
Average Length Of Melodic Arcs	jSymbolic		The average number of notes that separate peaks and troughs in melodic arcs.	Descriptor
Average Note Duration	jSymbolic		The average note duration in seconds.	Descriptor	This feature reports duration in seconds, unlike quarter-note duration means such as mean_duration and mean_rhythmic_value.
Basic Pitch Histogram	jSymbolic		A histogram of pitch values and their counts, with one pitch bin per non-zero count.	Sequence	We only return bins for pitches that have a count > 0. An implementation that is truer to the original jSymbolic implementation would return 128 bins (0-127) regardless of how any different pitches are present. However, we believe our approach is more concise and easier to understand for many purposes.
Bisect Unbalance	MUST		The bisect unbalance of a melody's temporal distribution of note onsets. Measures equilibrium between the first and second halves of the stimulus. Computed as ``1 - 4 * f1 * f2``, where ``f1`` and ``f2`` are the proportions of note onsets falling before and after the temporal midpoint, respectively. Values near 1 indicate balanced onset placement; lower values indicate concentration of events in one half.	Descriptor
Center Mass Offset	MUST		The center of mass offset of a melody's note-onset distribution. The absolute distance between the temporal center of mass (mean onset time, expressed as a proportion of total stimulus duration) and the geometric center (0.5). Values near 0 indicate a centrally concentrated onset distribution.	Descriptor
Chromatic Motion	jSymbolic		The proportion of chromatic motion in the melody. Chromatic motion is defined as a melodic interval of 1 semitone.	Descriptor
Comb Contour Matrix	MIDI Toolbox		The Marvin and Laprade comb contour matrix. For a melody with ``n`` notes, this feature returns an ``n x n`` lower-triangular binary matrix. Entry ``C[i][j]`` is ``1`` when note ``j`` is higher than note ``i`` and ``i >= j``; otherwise it is ``0``. The matrix encodes pairwise pitch-height relations in the melodic contour.	Sequence
Combined Strength Of Two Strongest Rhythmic Pulses	jSymbolic		Sum of the magnitudes of the two strongest rhythmic pulses.	Descriptor
Combined Strength Of Two Strongest Rhythmic Pulses Tempo Standardized	jSymbolic		Sum of the magnitudes of the two strongest rhythmic pulses using tempo-standardized histogram.	Descriptor
Complete Rests Fraction	jSymbolic		The fraction of the total duration during which no pitched notes are sounding.	Descriptor	This feature includes all complete silent runs (including those shorter than 0.1 quarter notes), whereas other complete-rest summary statistics apply a minimum 0.1 quarter-note threshold.
D1 Entropy	MUST		The zeroth-order duration entropy of a melody. Shannon entropy of the distribution of note durations in quarter-note beats. The final note duration is excluded, consistent with the MUST ``ddist1`` convention.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation. Durations are rounded to two decimal places before binning.
D2 Entropy	MUST		The first-order (2-tuple) duration entropy of a melody. Shannon entropy of the distribution of consecutive duration pairs in quarter-note beats. The final note duration is excluded.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation. Durations are rounded to two decimal places before binning.
D3 Entropy	MUST		The second-order (3-tuple) duration entropy of a melody. Shannon entropy of the distribution of consecutive duration triples in quarter-note beats. The final note duration is excluded.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation. Durations are rounded to two decimal places before binning.
Ddist1	MUST		Duration distribution in beats (MUST ``ddist1.m``). The final note duration is excluded, consistent with the MUST ``ddist*`` convention. Durations are rounded to two decimal places in beats.	Sequence
Ddist2	MUST		2-tuple duration distribution in beats (MUST ``ddist2.m``).	Sequence
Ddist3	MUST		3-tuple duration distribution in beats (MUST ``ddist3.m``).	Sequence
Difference Between Most Common Rhythmic Values	jSymbolic		The absolute difference in bins between most and second most common rhythmic values.	Descriptor
Direction Of Melodic Motion	jSymbolic		The proportion of upward melodic motions with regards to the total number of melodic motions.	Descriptor	This feature excludes unisons from its denominator and maps only to [0, 1].
Distance Between Most Prevalent Melodic Intervals	jSymbolic		The absolute difference between the two most common interval sizes.	Descriptor
Dominant Spread	jSymbolic		The longest sequence of pitch classes separated by perfect 5ths that each appear >9% of the time.	Descriptor
Dotted Duration Transitions	FANTASTIC	Steinbeck (1982)	The proportion of duration transitions that are dotted.	Descriptor
Dur Abruptness	MUST		The durational abruptness of pitch-direction changes in a melody. The proportion of total note duration (in seconds) accounted for by notes at which the pitch contour changes direction.	Descriptor
Duration Accent	MIDI Toolbox	Parncutt (1994)	Calculate duration accent for each note based on Parncutt (1994). Duration accent represents the perceptual salience of notes based on their duration.	Sequence	The MIDI toolbox implementation uses defaults of 0.5 for tau (saturation duration) and 2.0 for accent_index (minimum discriminable duration).
Duration Accent Std	MIDI Toolbox	Parncutt (1994)	The standard deviation of duration accents. Duration accent represents the perceptual salience of notes based on their duration, as defined by Parncutt (1994).	Descriptor
Duration Entropy	FANTASTIC		The zeroth-order base-2 entropy of the duration distribution in quarter notes.	Descriptor
Duration Histogram	jSymbolic		A histogram of note durations in quarter notes.	Sequence
Duration Range	FANTASTIC		The range between the longest and shortest note duration in quarter notes.	Descriptor
Duration Standard Deviation	Novel		The standard deviation of note durations in quarter notes.	Descriptor
Durdist1	MIDI Toolbox		Note duration distribution in nine log-spaced beat bins. Bin centers (in beats): 1/4, √2/4, 1/2, √2/2, 1, √2, 2, 2√2, 4.	Sequence
Durdist2	MIDI Toolbox		Second-order duration transition distribution (MIDI Toolbox ``durdist2.m``). Returns a 9×9 matrix of transition probabilities between log-spaced duration bins (same bin centres as ``durdist1``). Keys are ``(from_bin, to_bin)`` with bin indices 1–9.	Sequence
Equal Duration Transitions	FANTASTIC	Steinbeck (1982)	The proportion of duration transitions that are equal in length.	Descriptor
Event Density	MUST		The event density of a melody. The number of note events divided by total stimulus duration in seconds.	Descriptor
Event Heterogeneity	MUST		The event heterogeneity of a melody's temporal distribution of onsets. First computes a local unbalance curve using sliding windows sized to contain two note events (stepped at half the window length), then returns the distance-weighted mean squared deviation of that curve from unity.	Descriptor
Expectancy Complexity Optimal (Complebm)	MIDI Toolbox	Eerola & North (2000)	Expectancy-based melodic complexity calculated using an optimal combination of pitch patterns and rhythmic features, according to Eerola & North (2000). The complexity score is normalized against the Essen folksong collection, where a score of 5 represents average complexity.	Descriptor
Expectancy Complexity Pitch (Complebm)	MIDI Toolbox	Eerola & North (2000)	Expectancy-based melodic complexity calculated using pitch patterns only, according to Eerola & North (2000). The complexity score is normalized against the Essen folksong collection, where a score of 5 represents average complexity.	Descriptor
Expectancy Complexity Rhythm (Complebm)	MIDI Toolbox	Eerola & North (2000)	Expectancy-based melodic complexity calculated using rhythmic features only, according to Eerola & North (2000). The complexity score is normalized against the Essen folksong collection, where a score of 5 represents average complexity.	Descriptor
First Pitch	jSymbolic		The first pitch number in the melody.	Descriptor
First Pitch Class	jSymbolic		The first pitch class in the melody.	Descriptor
Folded Fifths Pitch Class Histogram	jSymbolic		A histogram of pitch classes arranged according to the circle of fifths.	Sequence
Global Duration	FANTASTIC jSymbolic		The total duration in seconds of the melody.	Descriptor	This feature is named duration_in_seconds (Duration In Seconds) in jSymbolic.
Gradus	MIDI Toolbox	Euler (1739)	The degree of melodiousness based on Euler's `gradus` suavitatis.	Descriptor
Half Duration Transitions	FANTASTIC	Steinbeck (1982)	The proportion of duration transitions that are halved or doubled.	Descriptor
Harmonicity Of Two Strongest Rhythmic Pulses	jSymbolic		The ratio of higher to lower bin index of the two strongest rhythmic pulses.	Descriptor	The first peak is selected from the raw beat histogram and the second peak is selected from the thresholded peak table (column 1), excluding the first peak bin.
Harmonicity Of Two Strongest Rhythmic Pulses Tempo Standardized	jSymbolic		The ratio of higher to lower bin index of the two strongest rhythmic pulses (120-BPM standardized histogram).	Descriptor	The first peak is selected from the standardized histogram values from the standardized histogram values and the second peak is selected from the standardized thresholded peak table (column 1), excluding the first peak bin.
Honore's H	FANTASTIC	Honoré (1979)	Honoré's H measure of m-type lexical richness. Honoré's H relates the total number of m-type tokens to the proportion of distinct m-types that occur exactly once (hapax legomena). It increases when a sequence contains many single-occurrence m-types relative to its overall m-type vocabulary.	Descriptor
Huron Contour Class Label	FANTASTIC	Huron (1996)	The Huron three-point contour classification for the melody. The melody is reduced to three pitch points: the first pitch, a rounded duration-weighted mean pitch, and the last pitch. Their relative ordering determines the contour class. For example, first < mean < last is ``"ascending"``, first > mean > last is ``"descending"``, first < mean > last is ``"convex"``, and first > mean < last is ``"concave"``.	Descriptor
I1 Entropy	MUST		The zeroth-order interval entropy of a melody. Shannon entropy of the distribution of consecutive melodic intervals, weighted by the underlying 2-tuple pitch distribution.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
I2 Entropy	MUST		The first-order (2-tuple) interval entropy of a melody. Shannon entropy of the distribution of consecutive interval pairs, weighted by the underlying 3-tuple pitch distribution.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
IOI	IDyOM		The sequence of inter-onset intervals. An inter-onset interval (IOI) is the elapsed time from one note onset to the next note onset. Unlike note duration, it includes any overlap or silence between consecutive notes because it depends only on onset times.	Sequence
IOI Contour	Novel		The sequence of IOI-ratio contour values (-1: shorter, 0: same, 1: longer).	Sequence	This contour is computed from ratios of consecutive IOIs, so it requires at least three onsets.
IOI Contour Mean	Novel		The arithmetic mean of ordinal IOI contour values. IOI contour values are ``-1`` for shorter, ``0`` for unchanged, and ``1`` for longer successive inter-onset intervals. The mean summarizes the balance of lengthening versus shortening onset gaps.	Descriptor
IOI Contour Standard Deviation	Novel		The sample standard deviation of ordinal IOI contour values. IOI contour values encode whether successive inter-onset intervals shorten, stay the same, or lengthen. This feature measures how variable those ordinal changes are across the melody.	Descriptor
IOI Histogram	Novel		A histogram of inter-onset intervals.	Sequence
IOI Mean	IDyOM jSymbolic		The arithmetic mean of inter-onset intervals.	Descriptor	This is called average_time_between_attacks in jSymbolic.
IOI Range	Novel		The range of inter-onset intervals.	Descriptor
IOI Ratio	IDyOM		The sequence of ratios between successive inter-onset intervals. First, consecutive onset times are converted to inter-onset intervals (IOIs). Each output value is then ``IOI[i] / IOI[i - 1]``. Values greater than ``1`` indicate that the current onset gap is longer than the previous one, values less than ``1`` indicate a shorter gap, and ``1`` indicates no change.	Sequence
IOI Ratio Mean	Novel		The arithmetic mean of successive inter-onset interval ratios. The ratio sequence is computed as ``IOI[i] / IOI[i - 1]`` for each pair of adjacent inter-onset intervals. This summary is above ``1`` when IOIs tend to lengthen, below ``1`` when they tend to shorten, and close to ``1`` when adjacent IOIs tend to have similar lengths.	Descriptor
IOI Ratio Standard Deviation	Novel		The sample standard deviation of successive inter-onset interval ratios. This feature measures the variability of ``IOI[i] / IOI[i - 1]`` across the melody. Larger values indicate less regular proportional change between neighboring onset gaps.	Descriptor
IOI Standard Deviation	IDyOM jSymbolic		The standard deviation of inter-onset intervals.	Descriptor	This is called variability_of_time_between_attacks in jSymbolic.
Idist1	MUST		Interval distribution marginalized from ``pdist2`` (MUST ``idist1.m``).	Sequence
Idist2	MUST		2-interval distribution marginalized from ``pdist3`` (MUST ``idist2.m``).	Sequence
Importance Of Bass Register	jSymbolic		The proportion of pitch numbers in the melody that are between 0 and 54.	Descriptor
Importance Of High Register	jSymbolic		The proportion of pitch numbers in the melody that are between 73 and 127.	Descriptor
Importance Of Middle Register	jSymbolic		The proportion of pitch numbers in the melody that are between 55 and 72.	Descriptor
Initial Tempo	FANTASTIC jSymbolic		The first tempo of the melody.	Descriptor
Inscale	IDyOM		For each pitch in the melody, returns 1 if the pitch is in the estimated key's scale, or 0 if it deviates from the scale.	Sequence
Interpolation Contour Class Label	FANTASTIC		A four-symbol categorical summary of interpolation-contour gradients. The interpolation contour is sampled at four equally spaced positions. Each sampled gradient is measured in semitones per second, rescaled to semitones per 0.25 seconds, and assigned to one of five ordered categories from strong downward motion to strong upward motion. The returned string preserves temporal order, so ``"ddbb"`` means that the first two sampled regions are upward and the last two are downward. Categories are defined as follows: - ``"a"``: strong downward motion; normalized gradient <= -1.45. - ``"b"``: downward motion; -1.45 < normalized gradient <= -0.45. - ``"c"``: approximately flat; -0.45 < normalized gradient < 0.45. - ``"d"``: upward motion; 0.45 <= normalized gradient < 1.45. - ``"e"``: strong upward motion; normalized gradient >= 1.45.	Descriptor
Interpolation Contour Direction Changes	FANTASTIC		The proportion of interpolation-contour transitions that reverse direction. The interpolation contour is a sequence of gradient samples. Consecutive samples with the same gradient belong to the same interpolation-gradient run, so they do not increase the denominator. The denominator is the number of transitions between distinct gradient runs; the numerator is the number of those transitions where the gradient changes sign. For example, the sampled contour ``[2, 2, 1, 1, -1, -1]`` has three gradient runs, ``[2, 1, -1]``. It has two transitions between runs, but only ``1 -> -1`` reverses direction, so the feature value is ``1 / 2 = 0.5``.	Descriptor
Interpolation Contour Global Direction	FANTASTIC		The overall direction of the interpolation-contour gradients. This feature sums the interpolation-contour gradient samples and returns the sign of that sum. A positive value indicates that upward gradients dominate overall, a negative value indicates that downward gradients dominate overall, and zero indicates that the sampled upward and downward gradients balance out.	Descriptor
Interpolation Contour Gradient Std	FANTASTIC		The sample standard deviation of interpolation-contour gradients. This feature measures how much the sampled interpolation-contour gradients vary around their mean. It uses Bessel's correction (``ddof=1``), matching the usual sample-standard-deviation convention. Contours with fewer than two gradient samples have no gradient variability and return ``0.0``.	Descriptor
Interpolation Contour Mean Gradient	FANTASTIC		The mean absolute gradient of the interpolation contour. The interpolation contour is represented as a sequence of local gradients between contour turning points. This feature takes the absolute value of each sampled gradient and averages the result, so upward and downward slopes contribute equally. Larger values indicate steeper pitch movement between turning points.	Descriptor
Interval Between Most Prevalent Pitch Classes	jSymbolic		The number of semitones between the two most prevalent pitch classes.	Descriptor
Interval Between Most Prevalent Pitches	jSymbolic		The number of semitones between the two most prevalent pitches.	Descriptor
Interval Direction	SIMILE		The sequence of interval directions in the melody, where 1 represents upward motion, 0 represents no motion, and -1 represents downward motion.	Sequence
Interval Direction Mean	Novel		The mean of the direction of each interval in the melody.	Descriptor	Unisons contribute 0 to both numerator and denominator because the mean is taken over the full direction sequence {-1, 0, 1}.
Interval Direction Std	Novel		The standard deviation of the direction of each interval in the melody.	Descriptor	Uses population variance (divide by N)
Interval Entropy	FANTASTIC		The zeroth-order base-2 entropy of the interval distribution.	Descriptor	Uses signed intervals rather than absolute interval sizes, consistent with FANTASTIC implementation.
Ivdirdist1	MIDI Toolbox		Directional interval bias for each interval size (1-12 semitones). Derived from duration-accent-weighted ``ivdist1`` (MIDI Toolbox ``ivdirdist1.m``).	Sequence
Ivdist1	MIDI Toolbox		Interval distribution weighted by duration accent.	Sequence
Ivdist2	MIDI Toolbox		Second-order interval transition distribution (MIDI Toolbox ``ivdist2.m``). Each note is assigned an interval class (incoming interval modulo octave, with 0 for the first note). Transition weights use the sum of Parncutt duration accents of the two notes. Keys are ``(from_interval, to_interval)`` in semitones (approximately -11 to 11).	Sequence
Ivsizedist1	MIDI Toolbox		Interval-size distribution (0-12 semitones) from ``ivdist1``. Derived from duration-accent-weighted ``ivdist1`` (MIDI Toolbox ``ivsizedist1.m``).	Sequence
Key	MIDI Toolbox	Krumhansl (1990)	The key of the melody, either read from the MIDI file or estimated using the specified key finding algorithm, depending on the key estimation strategy.	Sequence	This feature is named keyname in MIDI Toolbox.
Last Pitch	jSymbolic		The last pitch number in the melody.	Descriptor
Last Pitch Class	jSymbolic		The last pitch class in the melody.	Descriptor
Length	FANTASTIC jSymbolic		The total number of notes.	Descriptor	This feature is named total_number_of_notes (Total Number Of Notes) in jSymbolic.
Longest Complete Rest	jSymbolic		The longest uninterrupted complete rest in quarter-note units (ignoring rests shorter than 0.1 QN).	Descriptor
Longest Conjunct Scalar Passage	Novel		The longest sequence of consecutive notes that fit within the estimated key's scale. For example, a melody estimated to be in C major with notes C, D, E, F, G would have a longest conjunct scalar passage of 5.	Descriptor
Longest Monotonic Conjunct Scalar Passage	Novel		The longest sequence of consecutive notes that fit within the estimated key's scale that move in the same direction.	Descriptor
Longest Rhythmic Value	jSymbolic		The longest quantized rhythmic-bin value (in quarter notes).	Descriptor	This returns the ideal value of the longest occupied histogram bin, not the raw maximum note duration.
Max Log DF	FANTASTIC		log2 of the largest corpus document frequency among melody m-types. Highlights how frequent the most common pattern in the melody is relative to the corpus. Large values indicate that the melody contains at least one pattern that is very frequent in the corpus.	Descriptor
Maximum Note Duration	jSymbolic		The maximum note duration in seconds.	Descriptor
Mean Absolute Interval	FANTASTIC jSymbolic		The arithmetic mean of the absolute intervals in the melody.	Descriptor	This feature is named mean_melodic_interval (Mean Melodic Interval) in jSymbolic.
Mean Cloud Diameter	Partitura	Herremans & Chew (2016)	Mean cloud diameter from the tonal tension model. Cloud Diameter provides a measure of the maximal tonal distance of the notes in a chord, following the definition in Partitura.	Descriptor
Mean Cloud Momentum	Partitura	Herremans & Chew (2016)	Mean cloud momentum from the tonal tension model. Cloud momentum captures movement of pitch sets in the spiral array space, weighted by note durations, following the definition in Partitura.	Descriptor
Mean Complete Rest Duration	jSymbolic		The mean duration of complete rests in quarter-note units (ignoring rests shorter than 0.1 QN).	Descriptor
Mean Duration	Novel		The mean note duration in quarter notes, computed from the raw durations.	Descriptor	We use raw durations here (in the style of FANTASTIC), rather than jSymbolic's quantized rhythmic-value bins.
Mean Duration Accent	MIDI Toolbox	Parncutt (1994)	The mean duration accent across all notes. Duration accent represents the perceptual salience of notes based on their duration, as defined by Parncutt (1994).	Descriptor
Mean Entropy	FANTASTIC		Mean zeroth-order m-type entropy across counted n-gram orders. For each n-gram order, this feature treats the m-type counts as a discrete distribution and computes zeroth-order entropy. The returned value is the mean of those entropy values across the counted orders. Higher values indicate more even m-type distributions.	Descriptor
Mean Global Local Weight	FANTASTIC		Mean combined local-global weights for n-grams. The combined weight of an m-type is the product of local and global weights. In this implementation, unseen m-types (DF=0) receive a neutral global weight of 1.0. Higher values therefore indicate either high local frequency, higher global weight, or both. This relates to the percept of distinctiveness.	Descriptor
Mean Global Weight	FANTASTIC		Mean global weight across m-types. Higher values mean the m-types are less globally informative (more expected); lower values mean they are more globally informative (more distinctive).	Descriptor
Mean Log DF	FANTASTIC		Mean log2 corpus DF over melody m-types. Highlights how frequent the average pattern in the melody is relative to the corpus. Large values indicate that the melody contains patterns that are relatively frequent in the corpus. Small values indicate that the melody contains patterns that are relatively rare in the corpus.	Descriptor
Mean Log TFDF	FANTASTIC		Mean of log2-normalized TF × DF products over m-types. Higher values mean stronger alignment between within-melody usage and corpus document-frequency on the same m-types.	Descriptor
Mean Melodic Accent	MIDI Toolbox	Thomassen (1982)	The arithmetic mean of Thomassen melodic-accent values. Melodic accent values estimate local contour salience from three-note pitch windows. This feature averages those note-level salience values across the melody.	Descriptor
Mean Melodic Attraction	MIDI Toolbox		The arithmetic mean of Lerdahl melodic-attraction values.	Descriptor
Mean Mobility	MIDI Toolbox		The arithmetic mean of the `mobility` values across all notes.	Descriptor
Mean Pitch	jSymbolic		The arithmetic mean of the pitch numbers in the melody.	Descriptor
Mean Pitch Class	jSymbolic		The arithmetic mean of the pitch classes in the melody.	Descriptor	This is a linear (non-circular) mean over pitch classes. For example, pitch classes near the wraparound boundary (e.g., 11 and 0) are averaged numerically rather than on the unit circle.
Mean Productivity	FANTASTIC		The proportion of distinct m-types that occur only once. M-types that occur only once are hapax legomena. This feature divides the number of hapax m-types by the total number of distinct m-types, so higher values indicate that more of the melody's m-type vocabulary is used only once.	Descriptor
Mean Rhythmic Value	jSymbolic		The mean quantized rhythmic value in quarter notes.	Descriptor	Uses histogram-bin ideal values rather than raw note durations.
Mean Rhythmic Value Offset	jSymbolic		The mean quantized offset from the nearest ideal rhythmic value (in quarter notes).	Descriptor
Mean Rhythmic Value Run Length	jSymbolic		The mean run length of identical rhythmic values across the melody. Run length is the number of consecutive notes with the same rhythmic value. Returns 0.0 if there are fewer than 1 notes.	Descriptor
Mean Tempo	jSymbolic		The mean tempo of the melody.	Descriptor
Mean Tensile Strain	Partitura	Herremans & Chew (2016)	Mean tensile strain from the tonal tension model. Tensile strain provides a measure of the distance between the local and global tonal context, following the definition in Partitura.	Descriptor
Mean Tessitura	MIDI Toolbox		The arithmetic mean of local `tessitura` values.	Descriptor
Median Complete Rest Duration	jSymbolic		The median duration of complete rests in quarter-note units (ignoring rests shorter than 0.1 QN).	Descriptor
Median Rhythmic Value Offset	jSymbolic		The median quantized offset from the nearest ideal rhythmic value (in quarter notes).	Descriptor
Median Rhythmic Value Run Length	jSymbolic		The median run length of identical rhythmic values across the melody. Run length is the number of consecutive notes with the same rhythmic value.	Descriptor
Mel Abruptness	MUST		The melodic abruptness of pitch-direction changes in a melody. For each interior note where the pitch contour changes direction, accumulates the natural logarithm of the mean absolute interval size at that turning point, then normalizes by total stimulus duration in seconds.	Descriptor
Melodic Accent	MIDI Toolbox	Thomassen (1982)	Melodic accent salience for each note using Thomassen's contour model. Thomassen's model assigns accent strength from the melodic contour formed by three-note pitch windows. Notes at locally salient contour positions receive higher values. The implementation follows the MIDI Toolbox ``melaccent.m`` convention and returns values from ``0`` (no salience) to ``1`` (maximum salience).	Sequence
Melodic Accent Std	MIDI Toolbox	Thomassen (1982)	The sample standard deviation of Thomassen melodic-accent values. Melodic accent values estimate local contour salience from three-note pitch windows. This feature summarizes how unevenly that salience is distributed across the melody.	Descriptor
Melodic Attraction	MIDI Toolbox		Melodic attraction values following Lerdahl's tonal-attraction model.	Sequence
Melodic Attraction Std	MIDI Toolbox		The population standard deviation of Lerdahl melodic-attraction values.	Descriptor
Melodic Embellishment	jSymbolic		The proportion of melodic embellishments in the melody. Melodic embellishments are identified by notes that are surrounded on both sides by notes with durations at least 3 times longer than the central note.	Descriptor
Melodic Interval Histogram	jSymbolic		Histogram of absolute melodic interval sizes in semitones.	Sequence	We only return bins for intervals that have a count > 0. An implementation that is truer to the original jSymbolic implementation would return 128 bins (0-127) regardless of how any different intervals are present.
Melodic Large Intervals	jSymbolic		The proportion of intervals >= 13 semitones.	Descriptor
Melodic Octaves	jSymbolic		The proportion of intervals that are octaves (12 semitones).	Descriptor
Melodic Originality (Compltrans)	MIDI Toolbox	Simonton (1984)	The melodic originality measure, according to Simonton (1984). Calculated based on 2nd order pitch-class distribution derived from 15,618 classical music themes. Higher values indicate higher melodic originality (less predictable transitions).	Descriptor
Melodic Perfect Fifths	jSymbolic		The proportion of intervals that are perfect fifths (7 semitones).	Descriptor
Melodic Perfect Fourths	jSymbolic		The proportion of intervals that are perfect fourths (5 semitones).	Descriptor
Melodic Pitch Variety	jSymbolic		The average number of onset positions before a pitch is repeated.	Descriptor
Melodic Sevenths	jSymbolic		The proportion of intervals that are sevenths (10 or 11 semitones).	Descriptor
Melodic Sixths	jSymbolic		The proportion of intervals that are sixths (8 or 9 semitones).	Descriptor
Melodic Thirds	jSymbolic		The proportion of intervals that are thirds (3 or 4 semitones).	Descriptor
Melodic Tritones	jSymbolic		The proportion of intervals that are tritones (6 semitones).	Descriptor
Meter Accent	MIDI Toolbox		Phenomenal accent synchrony measure, calculated as the negative mean of the product of metric hierarchy, melodic accent, and durational accent for each note. Higher values indicate stronger accent synchrony. Implementation based on MIDI toolbox meteraccent.m.	Descriptor
Meter Denominator	jSymbolic		The denominator of the melody's active time signature. For a time signature written as ``numerator/denominator``, the denominator gives the note value that represents one beat: for example, ``4`` means a quarter-note beat and ``8`` means an eighth-note beat.	Descriptor
Meter Numerator	jSymbolic		The numerator of the melody's active time signature. For a time signature written as ``numerator/denominator``, the numerator is the number of beats in each notated bar. If a melody contains meter changes, this returns the meter stored on the melody object as its primary meter.	Descriptor
Metric Hierarchy	MIDI Toolbox		Metric hierarchy values for each note, indicating the strength of each note position within the known or estimated meter. Higher values indicate stronger metric positions (e.g., downbeat = 5, beat = 4, half-beat = 3, etc.). Implementation based on MIDI toolbox metrichierarchy.m.	Sequence
Min Log DF	FANTASTIC		log2 of the smallest corpus DF among melody m-types. Highlights how frequent the least common pattern in the melody is relative to the corpus. Small values indicate that the melody contains at least one pattern that is very rare in the corpus.	Descriptor
Minimum Note Duration	jSymbolic		The minimum note duration in seconds.	Descriptor
Minor Major Third Ratio	jSymbolic		The ratio of minor thirds to major thirds.	Descriptor	Instead of matching jSymbolic behavior (returning 0.0 when there are no major thirds), this returns NaN when there are no major thirds (including cases where minor thirds are present).
Mobility	MIDI Toolbox	von Hippel (2000)	The melodic `mobility` for each note based on von Hippel (2000). Mobility describes why melodies change direction after large skips by observing that they would otherwise run out of the comfortable melodic range. It uses lag-one autocorrelation between successive pitch heights.	Sequence
Mobility Std	MIDI Toolbox		The standard deviation of the `mobility` values across all notes.	Descriptor
Modal Duration	FANTASTIC jSymbolic		The modal raw note duration in quarter notes.	Descriptor	This computes the mode of raw quarter-note durations, so differs from jSymbolic most_common_rhythmic_value, which uses the modal bin in a 12-bin rhythmic-value histogram.
Modal Interval	FANTASTIC jSymbolic		The most common interval size in the melody.	Descriptor	This feature is named most_common_interval (Most Common Interval) in jSymbolic.
Mode	FANTASTIC		Calculate the mode (major/minor) of a melody, either read from the MIDI file or estimated using the specified key finding algorithm.	Sequence
Most Common Pitch	jSymbolic		The most frequently occurring pitch number in the melody.	Descriptor
Most Common Pitch Class	jSymbolic		The most frequently occurring pitch class in the melody.	Descriptor
Most Common Rhythmic Value	jSymbolic		The modal quantized rhythmic value (in quarter notes).	Descriptor	Uses rhythmic-value mode from a 12-bin histogram. Differs from modal_duration, which computes a raw-duration mode in quarter-note units.
NPVI	MIDI Toolbox	Patel & Daniele (2003)	The normalized Pairwise Variability Index (NPVI) of note durations in quarter notes. The NPVI measures the durational variability of events, originally developed for language research to distinguish stress-timed vs. syllable-timed languages. Applied to music by Patel & Daniele (2003) to study the prosodic influences on musical rhythm.	Descriptor
Narmour Closure	IDyOM MIDI Toolbox	Narmour (1990)	Narmour closure score for the final three-note pattern. The last three pitches define two successive intervals. One point is awarded when the second interval changes direction relative to the first. A second point is awarded when the second interval is at least two semitones smaller than the first in absolute size. The resulting score ranges from ``0`` to ``2``.	Descriptor
Narmour Intervallic Difference	IDyOM MIDI Toolbox	Narmour (1990)	Narmour intervallic-difference score for the final three notes. The last three pitches define an implicative interval followed by a realized interval. If the implicative interval is large (greater than a tritone), this feature returns ``1`` when the realized interval is sufficiently smaller: at least three semitones smaller in the same direction, or at least two semitones smaller after a direction change. If the implicative interval is small (smaller than a tritone), it returns ``1`` when the realized interval is similar in size, within the same margins. Otherwise it returns ``0``.	Descriptor
Narmour Proximity	IDyOM MIDI Toolbox	Narmour (1990)	Narmour proximity score for the final melodic interval. Proximity rewards small realized intervals. It is calculated as ``6 - d``, where ``d`` is the absolute semitone distance between the final two notes, and is clipped at ``0`` for intervals of a tritone or larger. Unisons therefore receive ``6``, whole tones receive ``4``, perfect fourths receive ``1``, and perfect fifths receive ``0``.	Descriptor
Narmour Registral Direction	IDyOM MIDI Toolbox	Narmour (1990)	Narmour registral-direction score for the final three notes. The last three pitches define an implicative interval followed by a realized interval. This feature returns ``1`` when a large implicative interval (greater than a tritone) is followed by a change of direction, or when a small implicative interval (smaller than a tritone) continues in the same direction. It returns ``0`` otherwise.	Descriptor
Narmour Registral Return	IDyOM MIDI Toolbox	Narmour (1990)	Narmour registral-return score for the final three-note pattern. Registral return measures whether the last three notes move away from a pitch and then return toward it. The contour must change direction and neither interval may be a repeated note. An exact return to the first pitch scores ``3``; returning within one semitone scores ``2``; returning within two semitones scores ``1``; all other patterns score ``0``.	Descriptor
Norm Log Dist	FANTASTIC		Mean absolute difference between log2-normalized TF and DF. Larger values mean the melody emphasizes different m-types than corpus prevalence; smaller values mean closer distributional match.	Descriptor
Note Density	FANTASTIC jSymbolic		The average number of notes per second.	Descriptor
Note Density Per Quarter Note	jSymbolic		The average number of note onsets per unit of time corresponding to an idealized quarter note duration based on the tempo.	Descriptor
Note Density Per Quarter Note Variability	jSymbolic		The standard deviation of note density per quarter note. Divides the melody into 8-quarter-note windows and calculates the standard deviation of note density across these windows.	Descriptor	Our tests indicate a certain discrepancy between our outputs and JSymbolic's outputs, which may be a consequence of JSymbolic's tick-based approach, or perhaps its idiosyncratic windowing approach.
Note Density Variability	jSymbolic		The standard deviation of note density across 5-second windows.	Descriptor	Our tests indicate a certain discrepancy between our outputs and JSymbolic's outputs, which may be a consequence of JSymbolic's tick-based approach, or perhaps its idiosyncratic windowing approach.
Number Of Common Melodic Intervals	jSymbolic		The number of intervals that appear in at least 9% of melodic transitions.	Descriptor
Number Of Common Pitch Classes	jSymbolic		The number of pitch classes that appear in at least 20% of total notes.	Descriptor	This feature is named number_of_common_pitches_classes (Number Of Common Pitches Classes) in local legacy export.
Number Of Common Pitches	jSymbolic		The number of unique pitch numbers that appear in at least 9% of total notes.	Descriptor
Number Of Common Rhythmic Values Present	jSymbolic		The number of rhythmic value bins with normalized proportion >= 0.15.	Descriptor
Number Of Different Rhythmic Values Present	jSymbolic		The number of distinct rhythmic value bins that are present in the melody (non-zero).	Descriptor
Number Of Moderate Rhythmic Pulses	jSymbolic		The number of beat histogram peaks with normalized magnitudes over 0.01.	Descriptor
Number Of Moderate Rhythmic Pulses Tempo Standardized	jSymbolic		The number of tempo-standardized beat histogram peaks with normalized magnitudes over 0.01.	Descriptor
Number Of Relatively Strong Rhythmic Pulses	jSymbolic		The number of peaks at least 30% of the max magnitude.	Descriptor
Number Of Relatively Strong Rhythmic Pulses Tempo Standardized	jSymbolic		The number of tempo-standardized peaks at least 30% of the max magnitude.	Descriptor
Number Of Strong Rhythmic Pulses	jSymbolic		The count of BPM bins with strong rhythmic pulses (> 0.1 in the underlying beat histogram).	Descriptor
Number Of Strong Rhythmic Pulses Tempo Standardized	jSymbolic		The count of BPM bins with strong pulses (> 0.1 in the underlying standardized beat histogram).	Descriptor
Number Of Unique Durations	Novel		The number of unique note durations, measured in quarter notes.	Descriptor
Number Of Unique Pitch Classes	jSymbolic		The number of unique pitch classes in the melody.	Descriptor
Number Of Unique Pitches	jSymbolic		The number of unique pitch numbers in the melody.	Descriptor
Number Of Unique Time Signatures	jSymbolic		The number of unique time signatures in the melody.	Descriptor	This feature is named Metrical Diversity in jSymbolic.
Onset Autocorr Peak	MIDI Toolbox		Maximum onset autocorrelation excluding lag 0.	Descriptor
Onset Autocorrelation	MIDI Toolbox	Parncutt (1994)	The autocorrelation function of onset times weighted by duration accents. This is calculated by weighting the onset times by the duration accents, as defined by Parncutt (1994). Onsets are converted to quarter-note beats using tempo before grid quantization.	Sequence
P1 Entropy	MUST		The zeroth-order pitch entropy of a melody. Shannon entropy of the marginal pitch distribution.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
P2 Entropy	MUST		The first-order (2-tuple) pitch entropy of a melody. Shannon entropy of the distribution of consecutive pitch pairs.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
P3 Entropy	MUST		The second-order (3-tuple) pitch entropy of a melody. Shannon entropy of the distribution of consecutive pitch triples.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
Pcdist1	MIDI Toolbox		Pitch-class distribution weighted by Parncutt duration accent.	Sequence
Pcdist2	MIDI Toolbox		Second-order pitch-class transition distribution (MIDI Toolbox ``pcdist2.m``). Transition weights are the product of Parncutt duration accents of the two notes. Keys are ``(from_pitch_class, to_pitch_class)`` with pitch classes 0–11 (C=0, …, B=11).	Sequence
Pdist1	MUST		Pitch distribution (MUST ``pdist1.m``). Returns normalized weights keyed by MIDI pitch.	Sequence
Pdist2	MUST		2-tuple pitch distribution (MUST ``pdist2.m``). Returns normalized weights keyed by consecutive pitch pairs.	Sequence
Pdist3	MUST		3-tuple pitch distribution (MUST ``pdist3.m``). Returns normalized weights keyed by consecutive pitch triples.	Sequence
Pitch Class Kurtosis	jSymbolic		The sample excess kurtosis of the pitch class histogram.	Descriptor
Pitch Class Kurtosis After Folding	jSymbolic		The sample excess kurtosis of the pitch class histogram, after arranging the pitch classes according to the circle of fifths.	Descriptor
Pitch Class Skewness	jSymbolic		The skewness of the pitch class histogram, using Pearson's median skewness formula.	Descriptor
Pitch Class Skewness After Folding	jSymbolic		The skewness of the pitch class histogram, using Pearson's median skewness formula, after arranging the pitch classes according to the circle of fifths.	Descriptor
Pitch Class Variability	jSymbolic		Standard deviation of all pitch classes in the melody.	Descriptor
Pitch Class Variability After Folding	jSymbolic		Standard deviation of all pitch classes after arranging the pitch classes by perfect fifths. Provides a measure of how close the pitch classes are as a whole from the mean pitch class from a dominant-tonic perspective.	Descriptor
Pitch Entropy	FANTASTIC		The zeroth-order base-2 entropy of the pitch distribution.	Descriptor
Pitch Interval	SIMILE		The intervals (in semitones) between consecutive pitches in the melody.	Sequence
Pitch Kurtosis	jSymbolic		The sample excess kurtosis of the pitch histogram.	Descriptor
Pitch LTM Mean Information Content	IDyOM	Pearce, M. (2005)	The average information content across all notes in a melody, calculated using IDyOM's long-term model (LTM). Information content is perceptually related to surprise, and can be calculated for pitches or rhythms.	Descriptor
Pitch Range	FANTASTIC MIDI Toolbox jSymbolic		Subtract the lowest pitch number in the melody from the highest.	Descriptor	This feature is named ambitus in MIDI Toolbox.
Pitch STM Mean Information Content	IDyOM	Pearce, M. (2005)	The average information content across all notes in a melody, calculated using IDyOM's prediction-by-partial-matching (PPM) algorithm. Information content is perceptually related to surprise, and can be calculated for pitches or rhythms.	Descriptor
Pitch Skewness	jSymbolic		The skewness of the pitch histogram, using Pearson's median skewness formula.	Descriptor
Pitch Spelling	Partitura	Meredith (2006)	Pitch spelling using the ps13s1 algorithm.	Sequence
Pitch Standard Deviation	FANTASTIC jSymbolic		Standard deviation of all pitch numbers in the melody.	Descriptor	This feature is named pitch_variability (Pitch Variability) in jSymbolic.
Polynomial Contour Coefficients	FANTASTIC		The first three non-constant polynomial-contour coefficients. Polynomial contour fits pitch as a polynomial function of centered onset time, then selects a model by Bayesian information criterion. This feature returns the linear, quadratic, and cubic terms ``[c1, c2, c3]`` from ``p = c0 + c1*t + c2*t**2 + c3*t**3``. The intercept ``c0`` is omitted because it represents absolute pitch height rather than contour shape.	Sequence
Polyrhythms	jSymbolic		The fraction of strong beat-histogram peaks related to the strongest peak. Among peaks at least 30% as tall as the maximum, returns the proportion whose bin is an integer multiple/factor of the strongest (multipliers 1, 2, 3, 4, 6, 8; ±3 bins).	Descriptor
Polyrhythms Tempo Standardized	jSymbolic		The fraction of strong beat-histogram peaks related to the strongest peak using the tempo-standardized beat histogram. Among peaks at least 30% as tall as the maximum, returns the proportion whose bin is an integer multiple/factor of the strongest (multipliers 1, 2, 3, 4, 6, 8; ±3 bins).	Descriptor
Prevalence Of Dotted Notes	jSymbolic		The sum of dotted rhythmic bins: 3, 5, 7, 9, 11.	Descriptor
Prevalence Of Long Rhythmic Values	jSymbolic		The sum of rhythmic bins 6 to 11 (half notes to dotted double whole notes or more).	Descriptor	Rhythmic-bin families overlap by construction in jSymbolic (e.g., short/medium/long), so these prevalence values are not mutually exclusive and can sum to more than 1.0.
Prevalence Of Medium Rhythmic Values	jSymbolic		The sum of rhythmic bins 2 to 6 (8th notes to half notes).	Descriptor	Rhythmic-bin families overlap by construction in jSymbolic (e.g., short/medium/long), so these prevalence values are not mutually exclusive and can sum to more than 1.0.
Prevalence Of Most Common Melodic Interval	jSymbolic		The proportion of intervals that are the most common interval.	Descriptor
Prevalence Of Most Common Pitch	jSymbolic		The proportion of pitches that are the most common pitch with regards to the total number of pitches in the melody.	Descriptor
Prevalence Of Most Common Pitch Class	jSymbolic		The proportion of pitch classes that are the most common pitch class with regards to the total number of pitch classes in the melody.	Descriptor
Prevalence Of Most Common Rhythmic Value	jSymbolic		The proportion (0.0 - 1.0) of the modal rhythmic bin.	Descriptor
Prevalence Of Short Rhythmic Values	jSymbolic		The sum of the three shortest rhythmic bins (indexes 0, 1, and 2).	Descriptor	Rhythmic-bin families overlap by construction in jSymbolic (e.g., short/medium/long), so these prevalence values are not mutually exclusive and can sum to more than 1.0.
Prevalence Of Very Long Rhythmic Values	jSymbolic		The sum of rhythmic bins 9 to 11 (dotted whole notes to dotted double whole notes or more).	Descriptor
Prevalence Of Very Short Rhythmic Values	jSymbolic		The sum of the two shortest rhythmic bins (indexes 0 and 1).	Descriptor
Proportion Conjunct Scalar	Novel		Longest conjunct scalar passage length divided by total note count.	Descriptor
Proportion Inscale	Novel		The proportion of notes in the melody that are in the scale of the estimated key.	Descriptor
Proportion Of Time In First Meter	Novel		The proportion of the melody's duration spent in its first time signature. The numerator and denominator of the first encountered time signature define the initial meter. This feature reports the fraction of total melody duration before any subsequent meter change. Melodies with no meter change therefore return ``1.0``.	Descriptor
Proportion Scalar	Novel		Longest monotonic conjunct scalar passage length divided by total notes.	Descriptor
Range Of Rhythmic Values	jSymbolic		The range of rhythmic values located within the 12-bin PPQN-based histogram. Durations are converted to quarter notes and mapped to 12 fixed rhythmic bins using midpoints. The returned value is the difference between the highest and lowest non-empty bins.	Descriptor
Referent	IDyOM		Calculate the `referent` (pitch-class root) of a melody.	Descriptor
Relative Prevalence Of Most Common Melodic Intervals	jSymbolic		The ratio of the frequency of the second most common interval to the frequency of the most common interval.	Descriptor
Relative Prevalence Of Most Common Rhythmic Values	jSymbolic		The ratio of the second-most-common rhythmic bin to the most common bin.	Descriptor
Relative Prevalence Of Top Pitch Classes	jSymbolic		The ratio of the frequency of the second most common pitch class to the frequency of the most common pitch class.	Descriptor
Relative Prevalence Of Top Pitches	jSymbolic		The ratio of the frequency of the second most common pitch to the frequency of the most common pitch.	Descriptor
Repeated Notes	jSymbolic		The proportion of repeated notes in the melody.	Descriptor
Rhythm Abruptness	MUST	Parncutt (1994)	The rhythmic abruptness of consecutive note durations. The mean ratio of consecutive beat durations after applying Parncutt (1994) duration accent, taking the larger-over-smaller ratio at each successive pair of notes.	Descriptor	Duration accent follows the MIDI Toolbox `duraccent defaults (tau=0.5, accent_index=2.0`).
Rhythm LTM Mean Information Content	IDyOM	Pearce, M. (2005)	The average rhythmic information content across all notes in a melody, calculated using IDyOM's long-term model (LTM). Information content is perceptually related to surprise, and can be calculated for pitches or rhythms.	Descriptor
Rhythm STM Mean Information Content	IDyOM	Pearce, M. (2005)	The average rhythmic information content across all notes in a melody, calculated using IDyOM's short-term model (STM). Information content is perceptually related to surprise, and can be calculated for pitches or rhythms.	Descriptor
Rhythmic Looseness	jSymbolic		The average width of beat histogram peaks. Width is defined as the distance between points at 30% of the peak height.	Descriptor
Rhythmic Looseness Tempo Standardized	jSymbolic		The average width of beat histogram peaks using tempo-standardized histogram. Width is defined as the distance between points at 30% of the peak height.	Descriptor
Rhythmic Variability	jSymbolic		The standard deviation of the beat histogram bin magnitudes, excluding the first 40 bins.	Descriptor
Rhythmic Variability Tempo Standardized	jSymbolic		The standard deviation of the tempo-standardized beat histogram bin magnitudes, excluding the first 40 bins.	Descriptor
Second Strongest Rhythmic Pulse	jSymbolic		The bin index (BPM) of the second-highest magnitude in the beat histogram.	Descriptor
Second Strongest Rhythmic Pulse Tempo Standardized	jSymbolic		The bin index (BPM) of the second-highest magnitude in the 120-BPM standardized beat histogram.	Descriptor
Shortest Rhythmic Value	jSymbolic		The shortest quantized (non-zero) rhythmic-bin value (in quarter notes).	Descriptor	This returns the ideal value of the shortest occupied histogram bin, not the raw minimum note duration.
Sichel's S	FANTASTIC	Sichel (1975)	The proportion of m-types that occur exactly twice. Sichel's S is the number of distinct m-types with frequency two divided by the total number of distinct m-types. Higher values indicate that more of the melody's m-type vocabulary consists of types that recur once.	Descriptor
Simpson's D	FANTASTIC	Simpson (1949)	Simpson's D measure of m-type concentration. Simpson's D is calculated from squared m-type frequencies. Higher values indicate a greater probability that two sampled tokens belong to the same m-type, and therefore a more concentrated or repetitive m-type vocabulary.	Descriptor
Standard Deviation Absolute Interval	FANTASTIC		The standard deviation of the absolute intervals in the melody.	Descriptor
Std Cloud Diameter	Partitura	Herremans & Chew (2016)	Standard deviation of cloud diameter from the tonal tension model. Cloud Diameter provides a measure of the maximal tonal distance of the notes in a chord, following the definition in Partitura.	Descriptor
Std Cloud Momentum	Partitura	Herremans & Chew (2016)	Standard deviation of cloud momentum from the tonal tension model. Cloud Momentum provides a measure of movement of pitch sets in the spiral array space, weighted by note durations, following the definition in Partitura.	Descriptor
Std Global Local Weight	FANTASTIC		Sample standard deviation of combined local-global weights for n-grams.	Descriptor
Std Global Weight	FANTASTIC		Sample standard deviation of global weights for m-types.	Descriptor
Std Tensile Strain	Partitura	Herremans & Chew (2016)	Standard deviation of tensile strain from the tonal tension model. Tensile strain provides a measure of the distance between the local and global tonal context, following the definition in Partitura.	Descriptor
Step Contour Global Direction	FANTASTIC		The overall ascending or descending tendency of the step contour. This is the Pearson correlation between the fixed-length step-contour pitch vector and an ascending linear ramp. Positive values indicate an overall upward trajectory, negative values indicate an overall downward trajectory, and values near zero indicate little linear pitch direction. A flat contour returns ``0.0``.	Descriptor
Step Contour Global Variation	FANTASTIC		The overall pitch variability of the step contour. The step contour is a fixed-length pitch vector in which each note's MIDI pitch is repeated in proportion to its duration. This feature is the standard deviation of that vector, so long notes influence the result more than short notes. Larger values indicate a wider duration-weighted pitch spread. The ``"amads"`` method uses population standard deviation (``ddof=0``), while ``"fantastic"`` uses sample standard deviation (``ddof=1``) for compatibility with the original FANTASTIC convention.	Descriptor
Step Contour Local Variation	FANTASTIC		The average adjacent-sample change in the step contour. This feature compares each neighboring pair of samples in the fixed-length step-contour pitch vector and averages the absolute pitch differences. Because adjacent samples inside the same sustained note have difference zero, the measure emphasizes local pitch changes after duration-weighted resampling rather than the raw sequence of note-to-note intervals.	Descriptor
Stepwise Motion	jSymbolic		The proportion of stepwise motion in the melody. Stepwise motion is defined as a melodic interval of 1 or 2 semitones.	Descriptor
Strength Of Second Strongest Rhythmic Pulse	jSymbolic		The magnitude of the beat histogram bin with the second-highest magnitude.	Descriptor
Strength Of Second Strongest Rhythmic Pulse Tempo Standardized	jSymbolic		The magnitude of the tempo-standardized beat histogram bin with the second-highest magnitude.	Descriptor
Strength Of Strongest Rhythmic Pulse	jSymbolic		The magnitude of the beat histogram bin with the highest magnitude.	Descriptor
Strength Of Strongest Rhythmic Pulse Tempo Standardized	jSymbolic		The magnitude of the tempo-standardized beat histogram bin with the highest magnitude.	Descriptor
Strength Ratio Of Two Strongest Rhythmic Pulses	jSymbolic		Ratio of the magnitude of the strongest to second-strongest rhythmic pulse.	Descriptor
Strength Ratio Of Two Strongest Rhythmic Pulses Tempo Standardized	jSymbolic		Ratio of the magnitude of the strongest to second-strongest rhythmic pulse (120-BPM standardized histogram).	Descriptor
Strong Tonal Centres	jSymbolic		Counts the number of isolated peaks in the pitch class histogram that each account for at least 9% of notes, arranged according to the circle of fifths.	Descriptor
Strongest Rhythmic Pulse	jSymbolic		The bin index (BPM) of the maximum beat histogram magnitude.	Descriptor
Strongest Rhythmic Pulse Tempo Standardized	jSymbolic		The bin index (BPM) of the maximum in the 120-BPM standardized beat histogram.	Descriptor
Syncopation	Novel	Longuet-Higgins & Lee (1984)	Calculate the mean `syncopation` value based on the Longuet-Higgins and Lee (1984) model. This `syncopation` model assigns metrical weights to each note position based on its position in the metric hierarchy. Syncopation occurs when a rest or tied note is preceded by a sounded note of lower metrical weight. The `syncopation` value is the difference between the rest weight and the preceding note weight.	Descriptor
Syncopicity	SIMILE		Calculates the sum `syncopicity` of a melody across metric levels. Syncopicity measures the degree to which notes occur off the main metrical grid but are long enough to span across metric boundaries. This calculates syncopations at four metric levels: 1) Half bar level 2) Beat level 3) First subdivision (half-beat) 4) Second subdivision (quarter-beat) An event is considered syncopated at a given level if: 1) It does not fall on a grid point of this level 2) It falls on a grid point of the next lower level 3) Its IOI extends beyond the lower level time unit (or it's the last note)	Descriptor	Grid durations are derived from the melody's initial meter and tempo and therefore assume constant meter/tempo over the analyzed passage.
TFDF Kendall	FANTASTIC		Kendall's tau rank correlation between melody TF and corpus DF for each m-type. Similar to ``tfdf_spearman``, but ordinal association is measured with Kendall's tau instead of Spearman's rho. Positive values mean higher within-melody usage tends to coincide with higher corpus-wide prevalence across m-types; negative values mean the opposite; near zero means little monotonic rank association.	Descriptor
TFDF Spearman	FANTASTIC		Spearman rank correlation between TF and DF over m-types. Positive values mean higher within-melody usage tends to coincide with higher corpus-wide prevalence across m-types; negative values mean the opposite; near zero means little monotonic rank association.	Descriptor
Tempo Variability	jSymbolic		The duration-weighted variability of tempo across the melody.	Descriptor
Tessitura	MIDI Toolbox	von Hippel (2000).	Tessitura is based on standard deviation from median pitch height. The median range of the melody tends to be favoured and thus more expected. Tessitura predicts whether listeners expect tones close to median pitch height. Higher `tessitura` values correspond to melodies that have a wider range of pitches.	Sequence
Tessitura Std	MIDI Toolbox		The standard deviation of the sequence of `tessitura` values.	Descriptor
Tonal Clarity	FANTASTIC	Temperley (2007)	The ratio between the top two key correlation values.	Descriptor
Tonal Entropy	Novel		Zeroth-order base-2 entropy of the 24-key Krumhansl-Schmuckler key correlation distribution. Normalizes the correlation values to a probability mass over all 24 major/minor keys, then computes Shannon entropy.	Descriptor
Tonal Spike	FANTASTIC		The ratio between the highest key correlation and the sum of all other correlations.	Descriptor
Tonal Tension	Partitura	Herremans & Chew (2016)	Computes tension ribbons using the tonal tension algorithm. Provides a means of comparing Chew's spiral array and the tonal tension profiles produced from Herremans and Chew's tension ribbons. This returns a dictionary containing the cloud diameter, cloud momentum, tensile strain, ordered by onset.	Sequence
Tonalness	FANTASTIC	Krumhansl (1990)	The magnitude of the highest correlation with a precomputed key profile. This key profile is established and elaborated on in Krumhansl (1990).	Descriptor
Tonalness Histogram	Novel	Krumhansl (1990)	Equal-width histogram of all 24 Krumhansl-Schmuckler key correlations.	Sequence
Variability In Rhythmic Value Run Lengths	jSymbolic		The standard deviation of rhythmic value run lengths. Run length is the number of consecutive notes with the same rhythmic value.	Descriptor
Variability Of Complete Rest Durations	jSymbolic		The standard deviation of complete rest durations in quarter notes (ignoring rests shorter than 0.1 QN).	Descriptor
Variability Of Note Durations	jSymbolic		The standard deviation of note durations in seconds.	Descriptor
Variability Of Rhythmic Value Offsets	jSymbolic		The standard deviation of rhythmic value offsets (in quarter notes).	Descriptor
Wp Entropy	MUST		The weighted permutation entropy of a melody's pitch sequence. Classifies each consecutive 3-note pitch window into one of 13 order signatures, weights each class by the standard deviation of the three pitches, and computes Shannon entropy over the resulting distribution.	Descriptor	Entropy is computed with the natural logarithm, consistent with the MUST Toolbox `shentropy` implementation.
Yule's K	FANTASTIC	Yule (1944)	Yule's K measure of m-type repetitiveness. This lexical-diversity feature is calculated from the frequency spectrum of m-types in the melody. Higher values indicate that a smaller set of m-types is repeated more often, whereas lower values indicate a more even or varied m-type vocabulary.	Descriptor

Sources

• FANTASTIC: Müllensiefen, D. (2009). Feature ANalysis Technology Accessing STatistics (In a Corpus): Technical Report v1.5
• jSymbolic: McKay, C., & Fujinaga, I. (2006). jSymbolic: A Feature Extractor for MIDI Files
• IDyOM: Pearce, M. T. (2005). The construction and evaluation of statistical models of melodic structure in music perception and composition
• MIDI Toolbox: Eerola, T., & Toiviainen, P. (2004). MIDI Toolbox: MATLAB Tools for Music Research
• MUST: Clemente, A., Vila-Vidal, M., Pearce, M. T., et al. (2020). A Set of 200 Musical Stimuli Varying in Balance, Contour, Symmetry, and Complexity
• Melsim: Silas, S., & Frieler, K. (n.d.). Melsim: Framework for calculating tons of melodic similarities
• Simile: Müllensiefen, D., & Frieler, K. (2004). The Simile algorithms documentation 0.3
• Novel: Custom features introduced in this package

Feature Types

• Descriptor: Returns a single scalar value (int, float, bool)
• Sequence: Returns a collection (list, tuple, dict, etc.)