Projects
Virtual Worlds
The Liberated learning Consortium's "Virtual Worlds" project actively researched
and identified potential accessibility gaps within interactive three dimensional
learning environments.
Virtual Worlds, CVEs, MUVEs, or MMOGs, MMORLGs are poised to significantly
impact social, recreational, and economic activities. Virtual technologies and
platforms introduce new opportunities to address longstanding issues of
inclusion, but can alternatively contribute to a growing digital divide.
A key issue is the introduction of multimedia including in-world speech, voice
based communication, video, and other audio sources. Speech technologies can
potentially enhance accessibility through targeted captioning, transcription,
indexing, and synchronization tasks. Other speech enabled applications include
machine translation, navigation, and command and control.
Given virtual technology is in its relative infancy, there is an opportunity to identify
accessibility issues and construct techniques, technologies, and standards that
ensure Virtual technologies and environments evolve in an accessible fashion.
The Liberated Learning Consortium team profiled and subsequently
investigated emerging accessibility issues related to multimedia in some of the
most prominent Virtual Worlds, particularly the inline virtual world Second Life.
The team identified specific areas where speech technologies can be
introduced to ameliorate information accessibility. Examples include:
- Using voice as a dictation technique for inputting text in various on-line chat
situations;
- Using voice to navigate menus, command and control interfaces;
- Captioning existing uncaptioned video, audio;
- Transcribing real time audio;
- Converting existing text to synthesized speech;
- Synchronizing disparate media sources.
The team performed testing of the existing SR applications IBM ViaScribe,
and Dragon Naturally Speaking, in the Second Life online virtual
environment. For technological gaps that were discovered,
the team articulated functional requirements for extending existing SR
applications and developed early prototypes for these applications. A proof of
concept "bridge" was developed using ViaScribe that allowed real time speech to
be captioned in Second Life. A macro set was also created using Dragon NatuarallySpeaking
that facilitated avatar movement and interface navigation.
Video Demonstrations of this technology in action are displayed below:
Liberated Learning and Virtual Accessibility
The increase in growth of Virtual Worlds have generated vast opportunities for people around the
world to share knowledge and ideas, participate in engaging social environments and become part
of a collaborative society capable of achieving remarkable accomplishments. Unfortunately the
technologies that enable people to connect in this setting are limited, with many groups and individuals
not able to contribute to this social network. Liberated Learning and Virtual Accessibility proposes
to bring together virtual societies to develop speech recognition technologies that will create new ways
to interact and communicate within virtual worlds. This will allow many disadvantaged users to become
part of this learning environment while providing existing participants with new and exciting
communication mediums.
The HASTAC (Humanities, arts, Science and Technology Advanced Collaboration) initiative are
potential sponcers for the second phase of the investigations into virtual worlds. They promote
the exciting concept of "Participatory Lerning"; innovative technological ways to contribute
towards shared learning environments.
Phase 2 of the investigations into virtual worlds proposes to incorperate this new concept
into the future development and refination of SR technologies for Virtual World interaction.
Within the 12 month grant term, the Liberated Learning and Virtual Accessibility project will
establish a presence within the mainstream virtual world Second Life, developing a virtual island.
The speech captioning system ViaScribe from IBM will be integrated into the virtual world allowing
verbal communication to be captioned and displayed as text in real time. This technology will enable
participation for people who prefer to speak rather than write or those who prefer to read rather than
listen. Users with disabilities who have difficulty typing or who are hard of hearing would greatly
benefit from this additional mode of communicating, enabling them to interact with others. The multiple
language base of ViaScribe provides a means for international consortium members from China, Germany,
and Spain to set up and maintain international pavilions upon the virtual island, encouraging
multicultural interaction. Machine translation technologies will also be investigated as a means of
bridging linguistic barriers. Streamed video and audio information with SR enabled captions/transcripts
will be provided at interactive workstations allowing new entrants a means to quickly learn and contribute
towards the project. Speech Recognition technologies will be developed for these systems to provide
additional ways to interact with these world objects. This will provide the perfect environment to
learn about various aspects of speech recognition technologies, virtual world interaction and
accessibility/disability awareness, while developing and testing new technologies. In world disability
groups and other project enthusiasts will be given the freedom to use this virtual space to set up social
events, conferencing meets and participatory leaning classes. Events such as these will also be organized
and encouraged by the project team to provide further opportunities for participatory learning.
|
