Dr Inventor ICCC Show Analogy

Analogy with Inferences ← Back to List for Target.

Inferences combine information from the target paper (blue) with the source paper (orange).

User-Controlling Expressed Emotions in Music with EDME (Year 2010) PDF Inferences Show Both PDFs Non-Conformant Harmonization: the Real Book in the Style of Take 6 (Year 2014) PDF
Emotion-Driven Music Engine software (EDME) expresses user-defined emotions with music and works in two stages. The first stage is done o✏ine and consists in emotionally classifying standard MIDI files in two dimensions: valence and arousal. The second stage works in real-time and uses classified files to produce musical sequences arranged in song patterns. First stage starts with the segmentation of MIDI files and proceeds to the extraction of features from the obtained segments. Classifiers for each emotional dimension use these features to label the segments, which are then stored in a music base. In the second stage, EDME starts by selecting the segments with emotional characteristics closer to the user-defined emotion. The software then uses a pattern-based approach to arrange selected segments into song-like structures. Segments are adapted, through transformations and sequencing, in order to match the tempo and pitch characteristics of given song patterns. Each pattern defines song structure and harmonic relations between the parts of each structure. The user interface of the application o↵ers three ways to define emotions: selection of discrete emotions from lists of emotions; graphical selection in a valence-arousal bi-dimensional space; or direct definition of valence-arousal values. While playing, EDME responds to input changes by quickly adapting the music to a new user-defined emotion. The user may also customize the music and pattern base. We intend to explore this possibility by challenging attendees to bring their own MIDI files and experiment the system. With this, we intend to allow a better understanding of the potential of EDME as a composition aid tool and get useful insights about further developments. References 1.
We will demonstrate a corpus of non-annotated MIDI files.

We will model the useful insights about further developments.

The works will are not adapted a task.

We will address the problem of automatically harmonizing a leadsheet.

We will demonstrate the validity of our approach.

The works will estimated a corpus of non-annotated MIDI files.

We address the problem of automatically harmonizing a leadsheet in the style of any arranger. We model the arranging style as a Markov model estimated from a corpus of non-annotated MIDI files. We consider a vertical approach to harmonization, in which chords are all taken from the ar-ranger corpus. We show that standard Markov models, using various vertical viewpoints are not adapted for such a task, because the problem is basically over constrained. We propose the concept of fioriture to better capture the subtleties of an arranging style. Fioritures are ornaments of the given melody during which the arranging style can be expressed more freely than for melody notes. Fioritures are defined as random walks with unary constraints and can be implemented with the technique of Markov constraints. We claim that fioritures lead to musically more interesting har-monizations than previous approaches and discuss why. We focus on the style of Take 6, arguably the most sophisticated arranging style in the jazz genre, and we demonstrate the validity of our approach by harmonizing a large corpus of standard leadsheets.

User-Controlling Expressed Emotions in Music with EDME (Year 2010) PDF	Inferences Show Both PDFs	Non-Conformant Harmonization: the Real Book in the Style of Take 6 (Year 2014) PDF
Emotion-Driven Music Engine software (EDME) expresses user-defined emotions with music and works in two stages. The first stage is done o✏ine and consists in emotionally classifying standard MIDI files in two dimensions: valence and arousal. The second stage works in real-time and uses classified files to produce musical sequences arranged in song patterns. First stage starts with the segmentation of MIDI files and proceeds to the extraction of features from the obtained segments. Classifiers for each emotional dimension use these features to label the segments, which are then stored in a music base. In the second stage, EDME starts by selecting the segments with emotional characteristics closer to the user-defined emotion. The software then uses a pattern-based approach to arrange selected segments into song-like structures. Segments are adapted, through transformations and sequencing, in order to match the tempo and pitch characteristics of given song patterns. Each pattern defines song structure and harmonic relations between the parts of each structure. The user interface of the application o↵ers three ways to define emotions: selection of discrete emotions from lists of emotions; graphical selection in a valence-arousal bi-dimensional space; or direct definition of valence-arousal values. While playing, EDME responds to input changes by quickly adapting the music to a new user-defined emotion. The user may also customize the music and pattern base. We intend to explore this possibility by challenging attendees to bring their own MIDI files and experiment the system. With this, we intend to allow a better understanding of the potential of EDME as a composition aid tool and get useful insights about further developments. References 1.	We will demonstrate a corpus of non-annotated MIDI files. We will model the useful insights about further developments. The works will are not adapted a task. We will address the problem of automatically harmonizing a leadsheet. We will demonstrate the validity of our approach. The works will estimated a corpus of non-annotated MIDI files.	We address the problem of automatically harmonizing a leadsheet in the style of any arranger. We model the arranging style as a Markov model estimated from a corpus of non-annotated MIDI files. We consider a vertical approach to harmonization, in which chords are all taken from the ar-ranger corpus. We show that standard Markov models, using various vertical viewpoints are not adapted for such a task, because the problem is basically over constrained. We propose the concept of fioriture to better capture the subtleties of an arranging style. Fioritures are ornaments of the given melody during which the arranging style can be expressed more freely than for melody notes. Fioritures are defined as random walks with unary constraints and can be implemented with the technique of Markov constraints. We claim that fioritures lead to musically more interesting har-monizations than previous approaches and discuss why. We focus on the style of Take 6, arguably the most sophisticated arranging style in the jazz genre, and we demonstrate the validity of our approach by harmonizing a large corpus of standard leadsheets.

Key Mapped Terms

Target Node Source Node
use use
work model
we we_our
intend show
intend propose
possibility concept
intend_challenge claim
intend_challenge consider
intend_get focus
insight style
understanding [unknown]-0
attendee [unknown]-2

Target Node	Source Node
use	use
work	model
we	we_our
intend	show
intend	propose
possibility	concept
intend_challenge	claim
intend_challenge	consider
intend_get	focus
insight	style
understanding	[unknown]-0
attendee	[unknown]-2

We would appreciate if you could give your thoughts on the analogy above by filling in the following evaluation form.

Explore this analogy in greater detail?

You may explore the analogy between larger portions of these papers, or even between the full papers. This may take up to 30 minutes to calculate as there maybe a lot of inferences and mappings.