Page 214 - Designing Sociable Robots

breazeal-79017 book March 18, 2002 14:16

Expressive Vocalization System 195

final value is the sum of the effects of the controlling parameters. The total of the absolute
values of these percentages must be 100%. See table 11.3 for the allowable bounds of
synthesizer settings. The computational mapping occurs in three stages.
In the first stage, the percentage of each of the vocal affect parameters (VAP i ) to its total
range is computed, (PP i ). This is given by the equation:
+ VAP offset
VAP value i
PP i =
VAP max − VAP min
VAP i is the current VAP under consideration, VAP value is its value specified by the current
emotion, VAP offset = 10 adjusts these values to be positive, VAP max = 10, and VAP min =−10.
In the second stage, a weighted contribution (WC j,i ) of those VAP i that control each
of DECtalk’s synthesizer settings (SS j ) is computed. The far right column of table 11.4
specifies each of the corresponding scale factors (SF j,i ). Each scale factor represents a
percentage of control that each VAP i applies to its synthesizer setting SS j .

For each synthesizer setting, SS j :
For each corresponding scale factor, SF j,i of VAP i :
If SF j,i > 0
WC j,i = PP i × SF j,i
If SF j,i ≤ 0
WC j,i = (1 − PP i ) × (−SF j,i )

SS j =
i WC j,i
At this point, each synthesizer value has a value 0 ≤ SS j ≤ 1. In the final stage, each
.
synthesizer setting SS j is scaled about 0.5. This produces the final synthesizer value, SS j final
The final value is sent to the speech synthesizer. The maximum, minimum, and default
values of the synthesizer settings are shown in table 11.3.
:
For each final synthesizer setting, SS j final
= SS j − norm
Compute SS j offset
> 0
If SS j offset
))
SS j final = SS j default + (2 × SS j offset × (SS j max − SS j min
≤ 0
If SS j offset
))
SS j final = SS j default + (2 × SS j offset × (SS j default − SS j min
Generating the Utterance
To engage in proto-dialogues with its human caregiver and to partake in vocal play, Kismet
must be able to generate its own utterances. The algorithm outlined below produces a style