Notes on first meeting » History » Version 13

Gyorgy Fazekas, 2012-02-23 11:46 AM

1 6 Gyorgy Fazekas
These notes are based on our initial meeting on 24 January 2012. The aim was to collect some use cases and have an initial idea on what needs to be done to extend or revise the existing Audio Features Ontology.
2 1 Gyorgy Fazekas
3 4 Gyorgy Fazekas
h1. Topics discussed (roughly)
4 1 Gyorgy Fazekas
5 1 Gyorgy Fazekas
** What are the main research use cases for an Audio Features Ontology (AF) ?
6 2 Gyorgy Fazekas
** Are they served well by the existing AF ? 
7 1 Gyorgy Fazekas
** If not, what are the most important extensions we need to do?
8 1 Gyorgy Fazekas
** Does the fundamental structure of the ontology need to be changed?
9 1 Gyorgy Fazekas
** What is the relation of AF to existing software, including:
10 10 Gyorgy Fazekas
11 8 Gyorgy Fazekas
 * software like: Sonic Annotator, Sonic Visualiser, SAWA, AudioDB other tools... 
12 1 Gyorgy Fazekas
 * and projects like: OMRAS2, EASAIER, SALAMI, new Semantic Media/Semantic Audio grants...
13 7 Gyorgy Fazekas
** Personal Objectives: what are we going to do with a modified/re-engineered ontology?
14 1 Gyorgy Fazekas
15 2 Gyorgy Fazekas
16 4 Gyorgy Fazekas
h1. Use cases:
17 1 Gyorgy Fazekas
18 12 Gyorgy Fazekas
Use cases discussed so far:
19 12 Gyorgy Fazekas
20 12 Gyorgy Fazekas
+Thomas:+
21 1 Gyorgy Fazekas
  
22 1 Gyorgy Fazekas
** drive audio effects -> adaptive effect (controlling effects)
23 1 Gyorgy Fazekas
** KM like use case: association of audio effects and audio features e.g. pitch shifter won’t change onsets
24 1 Gyorgy Fazekas
** part of the AFX ontology
25 1 Gyorgy Fazekas
** more audio features
26 1 Gyorgy Fazekas
** technical classification of audio effects
27 1 Gyorgy Fazekas
28 1 Gyorgy Fazekas
29 12 Gyorgy Fazekas
+Steve:+
30 1 Gyorgy Fazekas
** Finding structure, repeated sequences of features
31 1 Gyorgy Fazekas
** Beat related stuff, BPM (tempo, major/minor is it an audio feature, not necessarilty)
32 1 Gyorgy Fazekas
** Chords => Chord ontology
33 1 Gyorgy Fazekas
** Melody and notes
34 1 Gyorgy Fazekas
35 1 Gyorgy Fazekas
36 12 Gyorgy Fazekas
+George:+
37 1 Gyorgy Fazekas
** Improve SAWA
38 1 Gyorgy Fazekas
** Facilitate the development of intelligent music production systems
39 1 Gyorgy Fazekas
** Release large content based metadata repositories in RDF
40 1 Gyorgy Fazekas
** Re-release the MSD in RDF (??)
41 1 Gyorgy Fazekas
** Deploy a knowledge based environment for content-based audio analysis based on the concept of the Knowledge Machine that can combine multiple modalities
42 1 Gyorgy Fazekas
** Research reproducibility using Ontologies as a model to exchange research data.
43 1 Gyorgy Fazekas
44 6 Gyorgy Fazekas
45 1 Gyorgy Fazekas
h1. Open issues:
46 6 Gyorgy Fazekas
47 1 Gyorgy Fazekas
Some important questions to be decided on:
48 1 Gyorgy Fazekas
49 12 Gyorgy Fazekas
h2. Domain boundaries and scope:
50 1 Gyorgy Fazekas
51 1 Gyorgy Fazekas
** Are Musicological concepts outside the scope of an AF ?
52 12 Gyorgy Fazekas
** How about Physical features:
53 12 Gyorgy Fazekas
 
54 12 Gyorgy Fazekas
 *Acoustic features, 
55 12 Gyorgy Fazekas
 *Perceptual Features, 
56 12 Gyorgy Fazekas
 *DSP type feature, 
57 12 Gyorgy Fazekas
 *Musical Features (musically meaningful features related to acoustics) 
58 11 Gyorgy Fazekas
59 11 Gyorgy Fazekas
60 12 Gyorgy Fazekas
h2. Fundamental structure of the existing AF Ontology:
61 11 Gyorgy Fazekas
62 12 Gyorgy Fazekas
The Audio Features Ontology currently provides a core model 
63 11 Gyorgy Fazekas
64 13 Gyorgy Fazekas
!http://isophonics.net/sites/isophonics.net/files/AF_ontology_small.png!
65 11 Gyorgy Fazekas
66 11 Gyorgy Fazekas
67 9 Gyorgy Fazekas
!http://isophonics.net/sites/isophonics.net/files/FeatureConceptualisations.png!
68 6 Gyorgy Fazekas
69 6 Gyorgy Fazekas
The main scope of the ontology is to provide a framework for communication, feature representation, and describe the association of features and audio signals. Therefore it does not classify features, describe their interrelationships or their computation. With re- gards to the different conceptualisations of feature representations presented in table 4.2 (see §4.2.5.1), the Audio Features Ontology deals with data density, and temporal characteristics. It differentiates between dense, signal-like features of various dimensionality, for instance a chromagrams and detection functions, and sparse features that are scattered across the signal timeline, for instance, notes, or onsets.
70 6 Gyorgy Fazekas
71 6 Gyorgy Fazekas
** Does it serve us well?
72 1 Gyorgy Fazekas
** For example, loudness is defined as a segment in AF, and it does not fit a perceptual attribute well.
73 1 Gyorgy Fazekas
** What depth do we want ? (both in terms of scope and the level of detail in describing a feature extraction workflow)
74 1 Gyorgy Fazekas
** How AF relates to the DSP workflows used in extracting them?
75 1 Gyorgy Fazekas
76 1 Gyorgy Fazekas
77 1 Gyorgy Fazekas
h2. Existing resources :
78 1 Gyorgy Fazekas
79 2 Gyorgy Fazekas
h3. Some work related to Steve's use cases, segmentation and Ontologies:
80 1 Gyorgy Fazekas
81 2 Gyorgy Fazekas
** SALAMI Project: Kevin Page, DaveDeRoure http://salami.music.mcgill.ca/
82 2 Gyorgy Fazekas
** The Segment Ontology: http://users.ox.ac.uk/~oerc0033/preprints/admire2011.pdf
83 2 Gyorgy Fazekas
** PopStructure Ontology: Kurt Jacobson Unpublished. 
84 2 Gyorgy Fazekas
(Example available: http://wiki.musicontology.com/index.php/Structural_annotations_of_%22Can%27t_buy_me_love%22_by_the_Beatles) 
85 2 Gyorgy Fazekas
** Similarity Ontology: Kurt Jacobson http://grasstunes.net/ontology/musim/musim.html
86 1 Gyorgy Fazekas
87 1 Gyorgy Fazekas
88 2 Gyorgy Fazekas
h2. Ideas/resources for new Ontologies:
89 1 Gyorgy Fazekas
90 2 Gyorgy Fazekas
** Steve has worked on Acoustics related ontology
91 1 Gyorgy Fazekas
92 2 Gyorgy Fazekas
** Creating a DSP ontology:
93 2 Gyorgy Fazekas
** include processing steps down to math operations 
94 2 Gyorgy Fazekas
  (this can take advantage to the math:namespace in CWM: http://www.w3.org/DesignIssues/Notation3.html)
95 2 Gyorgy Fazekas
** describe common DSP parameters
96 2 Gyorgy Fazekas
97 2 Gyorgy Fazekas
** create an Acoustics Ontology
98 2 Gyorgy Fazekas
** describe Musicological concepts
99 2 Gyorgy Fazekas
** describe concepts related to cognitive and perceptual issues
100 2 Gyorgy Fazekas
101 2 Gyorgy Fazekas
102 1 Gyorgy Fazekas
h2. Currently missing features
103 1 Gyorgy Fazekas
104 2 Gyorgy Fazekas
** MFCC-s
105 1 Gyorgy Fazekas
** Rythmogram
106 1 Gyorgy Fazekas
** RMS energy
107 2 Gyorgy Fazekas
** combined features, e.g. weighted combinations or statistical averages over features
108 1 Gyorgy Fazekas
109 1 Gyorgy Fazekas
110 1 Gyorgy Fazekas
h2. Development issues
111 1 Gyorgy Fazekas
112 1 Gyorgy Fazekas
** chaining, combination, weighting
113 1 Gyorgy Fazekas
** how you associate features with arbitrary data
114 2 Gyorgy Fazekas
** summary feature types 
115 1 Gyorgy Fazekas
** SM (similarity matrix) are they part of the ontoogy?
116 2 Gyorgy Fazekas
** how to describe salience, can you hear it, can you perceive, is there an agreement
117 2 Gyorgy Fazekas
** how to describe weighting, confidence
118 1 Gyorgy Fazekas
** mood, music psychology, cognition, emotion, (perception ?)
119 1 Gyorgy Fazekas
** provenance => music provenance
120 2 Gyorgy Fazekas
** deprecation and versioning
121 1 Gyorgy Fazekas
122 1 Gyorgy Fazekas
123 1 Gyorgy Fazekas
h2. Long term objectives:
124 1 Gyorgy Fazekas
125 1 Gyorgy Fazekas
Some concrete tasks that can be done as the outcome of the collaboration:
126 1 Gyorgy Fazekas
127 2 Gyorgy Fazekas
** A version of Sonic Annotator that produces output adhering the new ontology
128 1 Gyorgy Fazekas
** Are we making people happier by doing so?
129 2 Gyorgy Fazekas
** gradual transition period?
130 2 Gyorgy Fazekas
** extend other software toolkits; e.g. a verison of Marsyas in C++
131 2 Gyorgy Fazekas
** multitrack processing using Sonic Annotator (this feature might come along soon)
132 1 Gyorgy Fazekas
133 1 Gyorgy Fazekas
134 2 Gyorgy Fazekas
h2. Some immediate tasks (TODO):
135 1 Gyorgy Fazekas
136 2 Gyorgy Fazekas
** collect more resources 
137 2 Gyorgy Fazekas
** Verify the relationship between AF as is, and other feature/segmentation Ontologies
138 2 Gyorgy Fazekas
** what other software uses it?
139 2 Gyorgy Fazekas
** papers and literature review
140 2 Gyorgy Fazekas
** relation to projects e.g. SIEMAC
141 2 Gyorgy Fazekas
** collect features that we need
142 2 Gyorgy Fazekas
** define scope (extend the diagram of the set of ontologies: )
143 2 Gyorgy Fazekas
** collect specific application examples from existing processing chain / workflow
144 1 Gyorgy Fazekas
145 2 Gyorgy Fazekas
collect software/projects that use/produce audio features:
146 1 Gyorgy Fazekas
147 1 Gyorgy Fazekas
** plugins, LADSPA, VAMP, Marsyas, CLAM, libextract, COMirva, MIRtoolbox, Supercollider, other frameworks
148 3 Gyorgy Fazekas
149 3 Gyorgy Fazekas
150 3 Gyorgy Fazekas
!http://www.isophonics.net/sites/isophonics.net/files/combined-frameworks.png!