![](bg9.png)
7 DISCUSSION
Focus on the communication between the system and
the user – multimodal output – is the main novelty of
our architecture and prototype. In general, systems
in this area of application incorporate various modali-
ties to communicate with the user (such as voice, text
or images) but they are completely devoid of any au-
tonomy, i.e. simply output the messages they receive
using default definitions. In the proposed model,
AdaptO, it is intended that these modalities have some
independence and self-adaptability to user and con-
text of use (ex: environment).
Another distinctive aspect of our proposal is the
delegation to the autonomous output agents of the re-
sponsibility for registering and, thereafter, of receiv-
ing the relevant user and context dynamic attributes
necessary for maximizing its adaptability to its desti-
nation users.
This design choice has several advantages: It sim-
plifies the fission coordination service, which is no
longer required to know everything about all the avail-
able output devices. It also easies the introduction of
new output agents, making it a more modular and ex-
tensible architecture.
However, there are some parts of the output prob-
lem that we do not yet address. First, nothing has
been done regarding the semantic fission. Second,
our user model service is yet a very simplified proof-
of-concept engine, composed only of a few user re-
lated attributes. Future versions will require not only
a strong database infrastructure but also the ability to
gather and to learn user related information, such as
to register its expressed preferences and to learn from
its historical usage patterns. Such historic informa-
tion is to be systematically registered by an history
agent service which is also missing from our current
implementation.
A separation between styling and rendering is also
missing in out output agents making them less adapt-
able to different types of output devices (like when we
move from a large LCD device to a small smartphone
one).
In short, in the WWHT model we have essen-
tially addressed the middle WT parts (“Which” and
“How”).
8 CONCLUSIONS
The basis for an intelligent adaptation of output - that
we called AdaptO, a Portuguese word meaning (I)
adapt - was proposed and discussed. First versions of
the required services – context and user services – and
of a few output agents for text and speech synthesiz-
ing are already implemented and are being used for
supporting a prototype remote telerehabilitation ap-
plication.
Ongoing and future work includes: first tests
with elderly users (the target for our work); re-
fining adaptation heuristics; use of more advanced
user models, possibly making use of ontologies such
as GUMO (Heckmann et al., 2005); and creating
new output agents such as 3D dynamic graphics and
avatars.
ACKNOWLEDGEMENTS
This work is part of the Living Usability Lab
for Next Generation Networks (www.livinglab.pt)
project, a QREN project, co-funded by COMPETE
and FEDER.
REFERENCES
Coetzee, L., Viviers, I., and Barnard, E. (2009). Model
based estimation for multi-modal user interface com-
ponent selection. In 20th Annual Symposium of
the Pattern Recognition Association of South Africa
(PRASA 2009), pages 1–6.
Dumas, B., Ingold, R., and Lalanne, D. (2009a). Bench-
marking fusion engines of multimodal interactive sys-
tems. In ICMI-MLMI ’09: Proceedings of the 2009
international conference on Multimodal interfaces,
pages 169–176, New York, NY, USA. ACM.
Dumas, B., Lalanne, D., Guinard, D., Ingold, R., and
Koenig, R. (2008). Strengths and weaknesses of soft-
ware architectures for the rapid creation of tangible
and multimodal interfaces. In Proceedings of 2nd in-
ternational conference on Tangible and Embedded In-
teraction (TEI 2008), pages 47–54.
Dumas, B., Lalanne, D., and Oviatt, S. (2009b). Multimodal
interfaces: A survey of principles, models and frame-
works. In Lalanne, D. and Kohlas, J., editors, Human
Machine Interaction, volume 5440 of Lecture Notes
in Computer Science, pages 3–26. Springer Berlin /
Heidelberg.
FIPA (2010 (accessed 7 November 2010)). The foun-
dation for intelligent physical agents. Available:
http://www.fipa.org.
Gordon-Salant, S. (2005). Hearing loss and aging: new
research findings and clinical implications. Journal
of Rehabilitation Research and Development, 42(4
Suppl 2):9–24.
Heckmann, D., Schwartz, T., Brandherm, B., Schmitz, M.,
and von Wilamowitz-Moellendorff, M. (2005). Gumo
- The General User Model Ontology. User Modeling
2005, pages 428–432.
AdaptO - Adaptive Multimodal Output
99