Melodic Features and Retrieval ISMIR Graduate School, Barcelona 2004 Musicology 3-4 Frans Wiering, ICS, Utrecht University.

Slides:



Advertisements
Similar presentations
How we talk about music Especially 20th century music.
Advertisements

Music Introduction to Humanities. Music chapter 9 Music is one of the most powerful of the arts partly because sounds – more than any other sensory stimulus.
Music Retrieval and Analysis
Auditory scene analysis 2
Recovering Human Body Configurations: Combining Segmentation and Recognition Greg Mori, Xiaofeng Ren, and Jitentendra Malik (UC Berkeley) Alexei A. Efros.
Musical Performance Musical performance in broad sense Children’s play songs, hymn or folk singing, dancing to music All merit investigation in own right.
ALL MUSIC HAS VALUE TO SOMEBODY. What is Music? The Organization of Sound in Time.
Point-set algorithms for pattern discovery and pattern matching in music David Meredith Goldsmiths College University of London.
Sept. 6/11. - Sound Sounds may be perceived as pleasant or unpleasant. What are these sounds that we hear? What is "sound"? What causes it, and how do.
Melodic Similarity CS 275B/Music 254. "Natural history" of similarity  Concept of similarity fundamental to organization of most art music  Types of.
Unit Cell Characterization, Representation, and Assembly of 3D Porous Scaffolds Connie Gomez, M. Fatih Demirci, Craig Schroeder Drexel University 4/19/05.
Chapter 6 Melodic Organization.
What is music? Music is the deliberate organization of sounds by people for other people to hear.
Multidimensional timbre analysis of melody Dr. Deirdre Bolger CNRS-LMS,Paris,France Invited lecture, Institut für Elektronische Musik und Akustik, Kunstuniversität.
Pattern Recognition Pattern - complex composition of sensory stimuli that the human observer may recognize as being a member of a class of objects Issue.
T.Sharon 1 Internet Resources Discovery (IRD) Music IR.
Visual Cognition II Object Perception. Theories of Object Recognition Template matching models Feature matching Models Recognition-by-components Configural.
Tone and Voice: A Derivation of the Rules of Voice- Leading from Perceptual Principles DAVID HURON Music Perception 2001, 19, 1-64.
Gestalt Principles Visual and Musical Examples. Sensation and Perception Sensation is the process of receiving stimuli (e.g., light and sound) from the.
Treisman (1960) Used shadowing task Listen to one ear: I saw the girl / song was wishing (correct answer) Ignore other ear: me that bird / jumping in the.
Elements of Music (Continued) Melody. (General) the horizontal aspect of music; pitches heard one after another (Specific) a series of single tones that.
Polyphonic Queries A Review of Recent Research by Cory Mckay.
A Time Based Approach to Musical Pattern Discovery in Polyphonic Music Tamar Berman Graduate School of Library and Information Science University of Illinois.
Musical Expectancy Definition of expectancy
Audio Scene Analysis and Music Cognitive Elements of Music Listening
1 Melodic Similarity MUMT 611, March 2005 Assignment 4 Paul Kolesnik.
Cognitive Systems Foresight Brain Rhythms. Cognitive Systems Foresight Sensory Processing How does the brain build coherent perceptual accounts of sensory.
Perceptual Processes: Visual & Auditory Recognition Dr. Claudia J. Stanny EXP 4507 Memory & Cognition Spring 2009.
Aspects of Music Information Retrieval Will Meurer School of Information University of Texas.
MEMORY. Sensory Memory Sensory Memory: The sensory memory retains an exact copy of what is seen or heard (visual and auditory). It only lasts for a few.
Melodic Organization Motive Rhythmic Motive Melodic Motive
A year 1 musicianA year 2 musicianA year 3 musician I can use my voice to speak, sing and chant. I can use instruments to perform. I can clap short rhythmic.
The Principles of Design
ARTDIRECTION BASIC DESIGN PRINCIPLES. PRINCIPLES OF DESIGN The combination of design elements.
Melody The Basics.
The Elements of Music.
Melodic Search: Strategies and Formats CS 275B/Music 254.
A survey of different shape analysis techniques 1 A Survey of Different Shape Analysis Techniques -- Huang Nan.
Cognitive Theories of Learning Dr. K. A. Korb University of Jos.
Theories of Learning: Cognitive Theories Dr. K. A. Korb University of Jos 15 May 2009.
MIT 6.893; SMA 5508 Spring 2004 Larry Rudolph Lecture Introduction Sketching Interface.
Melodic Similarity Presenter: Greg Eustace. Overview Defining melody Introduction to melodic similarity and its applications Choosing the level of representation.
Area of Study 05: Structure and Form AQA GCSE Music.
Other Aspects of Musical Sound pp Texture  Texture describes the number of things that are going on at once in a piece of music.  Monophony-
Things to Consider When Writing Melodies Vital Elements  Two most vital elements - rhythm and melody.  Harmonic structure of your composition will.
Acoustic Illusions & Sound Segregation Reading Assignments for Quizette 3 Chapters 10 & 11.
Using Transportation Distances for Measuring Melodic Similarity Pichaya Tappayuthpijarn Qiang Wang.
 6 th Musical Literacy 1.1 All students will be able to use a steady tone when performing.
Discovering Musical Patterns through Perceptive Heuristics By Oliver Lartillot Presentation by Ananda Jacobs.
CognitiveViews of Learning Chapter 7. Overview n n The Cognitive Perspective n n Information Processing n n Metacognition n n Becoming Knowledgeable.
Audio Scene Analysis and Music Cognitive Elements of Music Listening Kevin D. Donohue Databeam Professor Electrical and Computer Engineering University.
Elements of Classical Period. Elements Transition to classical period: (pre-classical period) Shift to more homophonic textures. Pioneers in.
Harmonic Models CS 275B/Music 254.
An Introduction to Music as Social Experience
Musical Similarity: More perspectives and compound techniques
PRINCIPLES OF DESIGN BY ANDREW MCNALLY.
National Curriculum Requirements of Music at Key Stage 1
Weaving Music Knowledge, Skills and Understanding into the new National Curriculum Key Stage 1: Music Forest Academy.
Unit 2: Melodic analysis (part 1)
Aspects of Music Information Retrieval
Memory and Melodic Density : A Model for Melody Segmentation
Fine Arts section 1 pg.7-20 By david steen.
Integrating Segmentation and Similarity in Melodic Analysis
MUMT 611, March 2005 Assignment 4 Paul Kolesnik
Melodic Similarity CS 275B/Music 254.
(the represented system)
English Theme 8 Music and Memory “What we have done so far”
What is Sound?
Understanding Standards An overview of course assessment
Presentation transcript:

Melodic Features and Retrieval ISMIR Graduate School, Barcelona 2004 Musicology 3-4 Frans Wiering, ICS, Utrecht University

Outline yesterday’s assignment demo: MIR outside academia (7:20; 44:10) one-dimensional melody retrieval Gestalt view of melody advanced melody retrieval assignment

one-dimensional melody retrieval common assumption is (was?) pitch-only retrieval is sufficient  e.g. CCGGAAGGFFEEDDEC  mechanisms for fuzzy matching variants  interval (distance between 2 pitches)  pitch-contour same/up/down (Parson’s Code) RURURDRDRDRDRUD examples:  (Rainer Typke)  (CCARH)

Results from Musipedia query is ranked 3 other hits are very unlikely  unfortunately no notation/sound available Haydn: evident false positive  why?

Themefinder Several 1-dimensional search options, e.g.  pitch  interval  contour  rhythm wildcards each theme stored as a number of strings matching by regular expressions ca themes  Barlow and Morgenstern (1948)  ESAC encodings  Lincoln, 16 th Century Motet (DARMS project)

results from Themefinder Example from Byrd & Crawford (2001) other hits  not as far-fetched as musipedia’s  different rhythm  different meter  still not very similar is this what people have in mind? Query: +m2 +M2 P1 -M2 -m2 -M2

Nice one we’ve just discovered Pitch-only search of MIDI on the internet many more MIR systems in Rainer Typke’s survey. URL is in your mailbox

Why pitch-only retrieval is unsatisfactory information contribution of other 3 parameters (estimate for Western music; Byrd & Crawford 2001)  pitch: 50%  rhythm: 40%  timbre + dynamics: 10% melodic confounds (Selfridge-Field 1998):  rests  repeated notes  grace notes, ornamentation  Mozart example

Why pitch-only retrieval is unsatisfactory information contribution of other 3 parameters (estimate for Western music; Byrd & Crawford 2001)  pitch: 50%  rhythm: 40%  timbre + dynamics: 10% melodic confounds (Selfridge-Field 1998):  rests  repeated notes  grace notes, ornamentation  Mozart example

Gestalt and melody melody: coherent succession of pitches  from New Harvard Dictionary of Music coherence important for similarity: creates musical meaning  bottom-up (pitches and durations)  top-down: segmenting, Gestalt Gestalt theory of perception  late 19 th /early 20 th century, Germany, later US  perception of wholes rather than parts  explanations: Gestalt principles of grouping  application in visual and musical domain

Low-level Gestalt principles Snyder mentions:  proximity rhythmic intervallic  similarity duration articulation  continuity melodic these produce closure of wholes Example: Beethoven 5 th symphony: beginning 1 st movement  also illustrates high-level principles from Snyder (2001)

Low-level Gestalt principles Snyder mentions:  proximity rhythmic intervallic  similarity duration articulation  continuity melodic these produce closure of wholes Example: Beethoven  also illustrates high-level principles from Snyder (2001)

High-level Gestalt principles parallellism  very strong in Mozart, Ah vous, second half of melody intensification  important organisational principle in variations and improvisations  Mozart’s last variation from Snyder (2001)

Application in analysis and retrieval Gestalt reduces memory overload: we can ignore the details Analytical: Schering (1911)  14 th century Italian songs  basic melodic shape  might be nice for retrieval Problem with Gestalt principles:  many different formulations  overlap; no rules for conflict  intuitive, cannot be successfully formalized from New Grove, Music analysis

The cognitive interpretation: chunking what creates a boundary  interval leap  long duration  tonality (stable chords)  etc Example of quantification: Melucci & Orio (2004)  using local boundary detection (Cambouropoulos 1997) apply weight to intervals and durations boundary after maximum  chunks forther processed for indexing

Organising chunks STM problem: max. 5-7 different elements  very short span solution: hierarchical grouping melody schemas  contours of melody cf. Schering ex.  examples: axial, arch, gap- fill  Mozart begins with gap-fill next level: form  A-B-A from Snyder (2001)

mental model of a song Ah, vous dirai-je maman melody level phrase level chunk level subchunk level A A B analysis synthesis analysis: from ear to LTM  (sub) chunks created by similarity and continuity a lot of parallellism  boundaries by leaps and harmony chunks may have a harmonic aspect too (I, V, V->I) synthesis: from LTM to focus of attention  recollection using general characteristics of phrases and chunks  performance notes are reconstitued through some musical grammar

Problems of melody retrieval People remember high-level concepts, not notes  often confused with poor performance abilities  theme-intensive music (fugues) stimulate formation of such concepts melodic variability and change  transposition  augmentation/diminution  ornamentation  variation  compositional processes: inversion, retrograde other factors  polyphony  harmony

Set-based approaches to melody retrieval in polyphony General idea:  compare note sets: find supersets, calculate distance  usually take rhythm and pitch into account  hopefully more tolerant agains some of the problems of melodic variety Clausen, Engelbrecht, Meyer, Schmidt (2000):  PROMS  matches onset times; wildcards  elegant indexing Lemström, Mäkinen, Ukkonen, Turkia (several articles, )  C-Brahms  algorithms for matching line segments P1: onsets P2: partial match onset times P3: common shared time  attention to time complexity Typke, Veltkamp, Wiering ( )  Orpheus system

Earth Mover’s Distance The Earth Mover’s Distance (EMD) measures similarity by calculating a minimum flow that would match two set of weighted points. One set emits weight, the other one receives weight Y. Rubner (1998); S. Cohen (1999)

Application to music represent notes as weighted point sets in 2-dimensional space (pitch, time) weight represents duration  other possibilities contour/metric position etc other possible application: pitch event + acoustic feature(s)? here, the ‘earth’ is only moved along the temporal axis

Another example interesting properties  tolerant against melodic confounds  suitable for polyphony  continuous  partial matching disadvantage  triangle inequality doesn’t hold  less suitable for indexing: after alignment, the ‘earth’ is moved both along the temporal axis and along the pitch axis

Test on RISM A/II

Matching polyphony with the EMD EMD’s partial matching property is essential MIDI example used as query for RISM database gross errors in playing are ironed out

Proportional Transportation Distance (PTD) Giannopoulos & Veltkamp (2002) EMD, weigths of sets normalised to 1 suitable for indexing  triangle inequality holds no partial matching

Test on RISM A/II only hits with approximately same length need 4 queries to find all known items

False positive (EMD) problems arise when length and/or number of notes differs considerably

Segmenting overlapping segments of 6- 9 consecutive notes not musical units search results are combined better Recall-Precision averages

Example of new search

Concluding remarks about melodic retrieval lots of creativity go into melody; difficult to give rules  not a ‘basic musical structure’ (Temperley 2001) essential to use multiple features  pitch, rhythm  harmony segmentation  finding perceptually relevant chunks is not easy  finding complete melodies may be harder  arbitrary segments may also work indexing strategies for melody melodic change over time several projects have tentative results for polyphony  gut feeling: false positives are big issue  notion of salience (Byrd and Crawford)