Guitar Fingering For Music Performance Page 2

ADVERTISEMENT

6
,
1
,
1
>
2
,
3
,
1
>
4
,
2
,
1
>
6
,
1
,
2
>
2
,
3
,
2
>
4
,
2
,
2
>
6
,
1
,
3
>
2
,
3
,
3
>
4
,
2
,
3
>
6
,
1
,
4
>
2
,
3
,
4
>
4
,
2
,
4
>
3
,
5
,
1
>
5
,
5
,
1
>
3
,
5
,
2
>
5
,
5
,
2
>
3
,
5
,
3
>
5
,
5
,
3
>
3
,
5
,
4
>
5
,
5
,
4
>
4
,
1
2
,
1
>
Figure 3. Chords may be considered as the third dimen-
6
,
1
2
,
1
>
4
,
1
2
,
2
>
6
,
1
2
,
2
>
sion of the graph: the first layer represents alternative fin-
4
,
1
2
,
3
>
6
,
1
2
,
3
>
gerings for the same chord shape, whereas in the second
4
,
1
2
,
4
>
6
,
1
2
,
4
>
5
,
1
5
,
1
>
and third layer, single fingered positions (for the individ-
5
,
1
5
,
2
>
ual notes D and G) are displayed.
5
,
1
5
,
3
>
5
,
1
5
,
4
>
Figure 2. Graph generated for a three notes (F2-D4-E3)
culties are considered and how they are evaluated. Weights
fragment. Each vertex represents a fingered position (e.g.,
estimating difficulties are based on the work in motor be-
<6,1,1>). Weights on the edges are omitted.
havior by Heijink & Meulenbroek [3], on the ergonomic
approach to keyboard fingering by Parncutt et al. [5], and
on the intuitive introspection of the authors. The elements
struction procedure guarantees that the graph has a layered
designed for the model have been quantitatively tuned on
structure, in that the vertices can be grouped in layers,
an implemented prototype.
and all the edges connect vertices of adjacent layers. The
Underpinned by [3], we assume that moving hands hor-
number of layers is equal to the number of events, be them
izontally along the fingerboard and vertically across
chords or one-note events, and any path from the leftmost
the fingerboard
constitute two different sources of dif-
layer (the first note) to the rightmost layer (the last note)
ficulty. Horizontal movements can span over a large dis-
represents a fingering for the score. Thus, the problem of
tance, and are named hand repositionings, while vertical
finding a suitable fingering for a piece corresponds to the
movements are considered less complex and are named
problem of finding a path in the graph.
finger displacements. The weight estimation proceeds by
Chords are represented as groups of vertices (collapsed
arranging the transitions between two positions into two
into one vertex), distributed along a third dimension (Fig-
classes, along the neck (henceforth ALONG) and across
ure 3); in facts, while layers correspond to the succession
the neck (ACROSS) respectively. The weight (WEIGHT)
of fingered notes, these groups model the simultaneity of
between two fingered positions p =
<string , fret ,
positions in chords. Such grouped representations result
and q =
is the
finger >
<string , fret , finger >
from a modeling of chord fingering as a constraint satis-
linear sum of the two difficulties:
faction problem (CSP) [2].
W EIGHT
= ALON G
+ ACROSS
(1)
Since performers pursue an overall effort-saving be-
havior, we assign a difficulty score (a cost) to the tran-
In turn, ALONG and ACROSS are computed by means of
sitions between vertices, and seek for a path that ensures
the following expressions:
the lowest overall difficulty score. An effective traversal
of the graph is then obtained by means of the dynamic pro-
ALON G
= f ret stretch
+ locality
(2)
gramming technique, that can find the shortest path with
ACROSS
= vertical stretch
(3)
linear, O(m), time, where m = Edges . The piece is
split into phrases, each one being fingered separately. The
We assume the following value ranges for the factors men-
motivations and the salience of phrasing and musical ar-
tioned here: ALONG varies between [0, 13.25], with fret
ticulation in the fingering process have been investigated
stretch varying from [0.5, 5] and locality varying from [0,8.5]
by Radicioni et al. [6]. Since evidence was provided that
(some combinations are not possible); ACROSS as well as
hand replacements are most likely to occur at the phrase
vertical stretch varies between [0.25, 0.5]. The rationale
boundaries than within the phrases, we adopt a local opti-
for these assignments will be clear after the description of
mization approach, where the model might choose a glob-
the individual factors.
ally harder fingering, but simplifying the execution of the
ALONG difficulties. Two factors contribute to ALONG
individual phrases.
difficulties: Fret Stretch, which accounts for the geomet-
ric distance between two positions on the fingerboard, and
Locality, which accounts for transitions around some area
3. ESTIMATING BIO-MECHANICAL
of the fingerboard.
DIFFICULTY
Fret Stretch. Given two fingered positions p and q, Fret
Here we show how weights are estimated to account for
Stretch is a measure of the difficulty due to moving hori-
performers’ bio-mechanical comfort, which kind of diffi-
zontally. We introduce a directed distance, deltaf ret =

ADVERTISEMENT

00 votes

Related Articles

Related forms

Related Categories

Parent category: Life
Go
Page of 4