Improving Music Classification Using Harmonic Complexity Ladislav - - PowerPoint PPT Presentation

▶
improving music classification using harmonic complexity
SMART_READER_LITE
LIVE PREVIEW

Improving Music Classification Using Harmonic Complexity Ladislav - - PowerPoint PPT Presentation

Nov 02, 2022 268 likes •567 views

Improving Music Classification Using Harmonic Complexity Ladislav Mark 1 , Jaroslav Pokorn 1 , Martin Ilk 2 1 Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic 2 Vienna University of Technology, Vienna,

slide-1
SLIDE 1

Improving Music Classification Using Harmonic Complexity

Ladislav Maršík1, Jaroslav Pokorný1, Martin Ilčík2

1 Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic 2 Vienna University of Technology, Vienna, Austria

slide-2
SLIDE 2

Categorization

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-3
SLIDE 3

 Motivation  Music harmony  Our music harmony model  Example analysis  Experiments: Music classification using our new feature

Outline

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-4
SLIDE 4

User preferences

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

recommends

slide-5
SLIDE 5

User preferences

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

recommends

Using what?

Tempo, Volume, Mood, Genre, Harmony, Melody, Author, Interpret, Music period, Instruments, ...

slide-6
SLIDE 6

 Determining genre / author / mood (or other category)

Music classification

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Using what?

Tempo, Volume, Harmony, Melody, Instruments, ...

slide-7
SLIDE 7

What we are working on?

 Finding a standard set of descriptors for music harmony  Motivation: there is no such descriptors yet

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-8
SLIDE 8

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

 1st step: Gathering low­level features using DFT, choosing tones with highest activation to

  • btain chords (harmonies)

 2nd step: Using our model, based on formal grammars, calculating „transition complexity“

between the successive chords (analogy to computational complexity)

 Example transitions:  Graph: 212 vertices, average degree ≈ 8

slide-9
SLIDE 9

Music harmony comparison

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-10
SLIDE 10

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

 Counting the mean transition complexity

slide-11
SLIDE 11

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Chord: Complexity: ∑:

F A C Eb

slide-12
SLIDE 12

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Chord: Complexity: ∑:

Bb Db F 3 3

slide-13
SLIDE 13

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Chord: Complexity: ∑:

Eb F G A 4 7

slide-14
SLIDE 14

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Chord: Complexity: ∑:

Bb Db F 4 11

slide-15
SLIDE 15

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Chord: Complexity: ∑:

Eb F G A 4 15

slide-16
SLIDE 16

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

∑:

# Transitions:

Average Transition Complexity: 3.16

60 19

slide-17
SLIDE 17

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

 Counting the mean transition complexity

slide-18
SLIDE 18

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Previous:

Now:

Next:

Transition: ∑:

G7 (G B D F)

C7 (C E G Bb)

slide-19
SLIDE 19

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Previous:

Now:

Next:

Transition: ∑:

G7 (G B D F)

C7 (C E G Bb)

F7 (F A C E)

3 3

slide-20
SLIDE 20

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Previous:

Now:

Next:

Transition: ∑:

C7 (C E G Bb)

F7 (F A C E)

Bb (Bb D F)

2 5

slide-21
SLIDE 21

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Previous:

Now:

Next:

Transition: ∑:

F7 (F A C E)

Bb (Bb D F)

G7 (G B D F)

1 6

slide-22
SLIDE 22

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

Previous:

Now:

Next:

Transition: ∑:

Bb (Bb D F)

G7 (G B D F)

C7 (C E G Bb)

2 8

slide-23
SLIDE 23

Harmonic complexity – useful harmony descriptor

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

∑:

# Transitions:

Average Transition Complexity: 2.08

25 12

slide-24
SLIDE 24

Supporting experiments

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

 Neural Network method  Parameters:  150­dimensional feature vector

MFCC, RMS Amplitude, Tempo, Transition probability matrix, Number of keys, Number of modulations, Number of similarity segments, Number of distinct chord roots with added mean Harmonic complexity

 20 hidden neurons  5 output classes

slide-25
SLIDE 25

Results: Without Harmonic complexity

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-26
SLIDE 26

Results: With Harmonic complexity

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

slide-27
SLIDE 27

Conclusion

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.

 Proposed a new descriptor for music analysis  Underlying grammar based model  Proved its usefulness for music classification problem

slide-28
SLIDE 28

Thank you for your attention

Maršík, Pokorný, Ilčík: Improving Music Classification Using Harmonic Complexity ITAT­DMUPW 2014, 28.9.