Audio Structure Analysis

Transcription

1 Tutorial T3 A Basic Introduction to Audio-Related Music Information Retrieval Audio Structure Analysis Meinard Müller, Christof Weiß International Audio Laboratories Erlangen meinard.mueller@audiolabs-erlangen.de, christof.weiss@audiolabs-erlangen.de

2 Book: Fundamentals of Music Processing Meinard Müller Fundamentals of Music Processing Audio, Analysis, Algorithms, Applications 483 p., 249 illus., hardcover ISBN: Springer, 2015 Accompanying website:

3 Book: Fundamentals of Music Processing Meinard Müller Fundamentals of Music Processing Audio, Analysis, Algorithms, Applications 483 p., 249 illus., hardcover ISBN: Springer, 2015 Accompanying website:

4 Book: Fundamentals of Music Processing Meinard Müller Fundamentals of Music Processing Audio, Analysis, Algorithms, Applications 483 p., 249 illus., hardcover ISBN: Springer, 2015 Accompanying website:

5 Chapter 4: Music Structure Analysis 4.1 General Principles 4.2 Self-Similarity Matrices 4.3 Audio Thumbnailing 4.4 Novelty-Based Segmentation 4.5 Evaluation 4.6 Further Notes In Chapter 4, we address a central and well-researched area within MIR known as music structure analysis. Given a music recording, the objective is to identify important structural elements and to temporally segment the recording according to these elements. Within this scenario, we discuss fundamental segmentation principles based on repetitions, homogeneity, and novelty principles that also apply to other types of multimedia beyond music. As an important technical tool, we study in detail the concept of self-similarity matrices and discuss their structural properties. Finally, we briefly touch the topic of evaluation, introducing the notions of precision, recall, and F-measure.

6 Music Structure Analysis Example: Zager & Evans In The Year 2525 Time (seconds)

7 Music Structure Analysis Example: Zager & Evans In The Year 2525 Time (seconds)

8 Music Structure Analysis Example: Zager & Evans In The Year 2525 I V1 V2 V3 V4 V5 V6 V7 B V8 O

9 Music Structure Analysis Example: Brahms Hungarian Dance No. 5 (Ormandy) A1 A2 B1 B2 C A3 B3 B4 Time (seconds)

10 Music Structure Analysis Example: Folk Song Field Recording (Nederlandse Liederenbank) Time (seconds)

11 Music Structure Analysis Example: Weber, Song (No. 4) from Der Freischütz Introduction Stanzas Dialogues Kleiber Time (seconds) Ackermann Time (seconds)

12 Music Structure Analysis General goal: Divide an audio recording into temporal segments corresponding to musical parts and group these segments into musically meaningful categories. Examples: Stanzas of a folk song Intro, verse, chorus, bridge, outro sections of a pop song Exposition, development, recapitulation, coda of a sonata Musical form ABACADA of a rondo

13 Music Structure Analysis General goal: Divide an audio recording into temporal segments corresponding to musical parts and group these segments into musically meaningful categories. Challenge: There are many different principles for creating relationships that form the basis for the musical structure. Homogeneity: Novelty: Repetition: Consistency in tempo, instrumentation, key, Sudden changes, surprising elements Repeating themes, motives, rhythmic patterns,

14 Music Structure Analysis Novelty Homogeneity Repetition

15 Overview Introduction Feature Representations Self-Similarity Matrices Novelty-Based Segmentation Thanks: Clausen, Ewert, Kurth, Grohganz, Dannenberg, Goto Grosche, Jiang Paulus, Klapuri Peeters, Kaiser, Serra, Gómez, Smith, Fujinaga, Wiering, Wand, Sunkel, Jansen

16 Overview Introduction Feature Representations Self-Similarity Matrices Novelty-Based Segmentation Thanks: Clausen, Ewert, Kurth, Grohganz, Dannenberg, Goto Grosche, Jiang Paulus, Klapuri Peeters, Kaiser, Serra, Gómez, Smith, Fujinaga, Wiering, Wand, Sunkel, Jansen

17 Feature Representation General goal: Convert an audio recording into a mid-level representation that captures certain musical properties while supressing other properties. Timbre / Instrumentation Tempo / Rhythm Pitch / Harmony

18 Feature Representation General goal: Convert an audio recording into a mid-level representation that captures certain musical properties while supressing other properties. Timbre / Instrumentation Tempo / Rhythm Pitch / Harmony

19 Feature Representation Example: Brahms Hungarian Dance No. 5 (Ormandy) A1 A2 B1 B2 C A3 B3 B4 Time (seconds)

20 Feature Representation Example: Brahms Hungarian Dance No. 5 (Ormandy) Chroma (Harmony) Feature extraction A1 A2 B1 B2 C A3 B3 B4 Time (seconds)

21 Feature Representation Example: Brahms Hungarian Dance No. 5 (Ormandy) B b G D Chroma (Harmony) G minor Feature extraction G minor A1 A2 B1 B2 C A3 B3 B4 Time (seconds)

22 Feature Representation Example: Brahms Hungarian Dance No. 5 (Ormandy) B G B b G Chroma (Harmony) Feature extraction D D G minor G major G minor A1 A2 B1 B2 C A3 B3 B4 Time (seconds)

23 Overview Introduction Feature Representations Self-Similarity Matrices Novelty-Based Segmentation

24 Self-Similarity Matrix (SSM) General idea: Compare each element of the feature sequence with each other element of the feature sequence based on a suitable similarity measure. Quadratic self-similarity matrix

25 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

26 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

27 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

28 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

29 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

30 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

31 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy) G major G major

32 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

33 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

34 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

35 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy)

36 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy) Faster Slower

37 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy) Faster Slower

38 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy) Idealized SSM

39 Self-Similarity Matrix (SSM) Example: Brahms Hungarian Dance No. 5 (Ormandy) Idealized SSM Blocks: Homogeneity Paths: Repetition Corners: Novelty

40 SSM Enhancement Challenge: Presence of musical variations Fragmented paths and gaps Paths of poor quality Regions of constant (low) cost Curved paths Idea: Enhancement of path structure

41 SSM Enhancement Path Enhancement Time (samples) Time (samples)

42 SSM Enhancement Path Enhancement Diagonal smoothing Time (samples) Time (samples)

43 SSM Enhancement Path Enhancement Diagonal smoothing Multiple filtering Time (samples) Time (samples)

44 SSM Enhancement Path Enhancement Diagonal smoothing Multiple filtering Thresholding (relative) Scaling & penalty Time (samples) Time (samples)

45 SSM Enhancement Further Processing Path extraction Time (samples) Time (samples)

46 SSM Enhancement Further Processing Path extraction Pairwise relations Time (samples) Time (samples) Time (samples)

47 SSM Enhancement Further Processing Path extraction Pairwise relations Grouping (transitivity) Time (samples) Time (samples) Time (samples)

48 SSM Enhancement Further Processing Path extraction Pairwise relations Grouping (transitivity) Time (samples) Time (samples) Time (samples) Time (samples)

49 SSM Enhancement Example: Zager & Evans In The Year 2525 I V1 V2 V3 V4 V5 V6 V7 B V8 O

50 SSM Enhancement Example: Zager & Evans In The Year 2525 Missing relations because of transposed sections

51 SSM Enhancement Example: Zager & Evans In The Year 2525 Idea: Cyclic shift of one of the chroma sequences One semitone up

52 SSM Enhancement Example: Zager & Evans In The Year 2525 Idea: Cyclic shift of one of the chroma sequences Two semitones up

53 SSM Enhancement Example: Zager & Evans In The Year 2525 Idea: Overlay & Maximize Transposition-invariant SSM

54 Overview Introduction Feature Representations Self-Similarity Matrices Novelty-Based Segmentation

55 Novelty-Based Segmentation General goals: Find instances where musical changes occur. Find transition between subsequent musical parts. Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM.

56 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM.

57 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM.

58 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM.

59 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM.

60 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM. Novelty function using

61 Novelty-Based Segmentation Idea (Foote): Use checkerboard-like kernel function to detect corner points on main diagonal of SSM. Novelty function using Novelty function using

62 Conclusions Structure Analysis

63 Conclusions Score Audio MIDI Representations Structure Analysis

64 Conclusions Score Audio MIDI Representations Harmony Musical Aspects Timbre Structure Analysis Tempo

65 Conclusions Score Audio MIDI Representations Musical Aspects Structure Analysis Segmentation Principles Harmony Timbre Tempo Repetition Homogeneity Novelty

66 Conclusions Temporal and Hierarchical Context Score Audio MIDI Representations Musical Aspects Structure Analysis Segmentation Principles Harmony Timbre Tempo Repetition Homogeneity Novelty

67 Links SM Toolbox (MATLAB) MSAF: Music Structure Analysis Framework (Python) SALAMI Annotation Data LibROSA (Python) Evaluation: mir_eval (Python) Deep Learning: Boundary Detection Jan Schlüter (PhD thesis)