People
often come to museums to expand their horizons.
Being immersed in a quiet room filled with thematically connected
exhibits can foster a level of engagement that helps learning. However, many gallery visitors today simply
glide past the exhibits, barely viewing each for a few seconds. When they are attracted to an exhibit, they
find very little material that explains the exhibit's significance and
details. Each visitor would have
different questions about the exhibits, depending on their level of knowledge
and interest, and the limited description may not provide the answers. A
trained tour guide could help by answering questions and tuning their responses
to each visitor's level; different responses are needed for a child vs. an
adult, or a knowledgeable person vs. a novice.
Unfortunately, it is not practical to assign a human guide for each
visitor. A guide for a group is nice,
but assuming a diversity of interest and knowledge levels in the group, it
would be difficult for the guide to give answers suitable for each visitor.
Seba
is a personalized AI based guide that addresses this need. It is knowledgeable about all the exhibits in
the gallery and can verbally converse with each visitor to explain and answer
questions. It does not get impatient or
rude and it will not ask for tips! Seba runs on the visitor's phone and
communicates with a server in the cloud.
For training, it is given details of each exhibit, similar to the text
that may be published at the venue's website.
This material is used to generate an introduction to the exhibit, to
suggest follow up topics and to answer specific questions. The process starts with Seba identifying the
exhibit. Currently, this is done by
asking the visitor to aim their phone camera at the exhibit's descriptive
text. Seba reads the text description
and identifies the exhibit. In its next
version, the camera can also be aimed at the exhibit itself.
An
initial prototype has been built. It is
set up with a few exhibits at the Asian Art Collection at the Princeton
University Art Museum. More exhibits can
be included by simply adding their descriptions. Other museum venues can be set up just as
easily.
Seba Visitor’s Guide — How to use this App.
Gallery Director’s Guide — How to add exhibits or set up a new Gallery.
Prototype — A web app implements a personal guide to the Asian Art Collection at
the Princeton University Art Museum.
To try out the app offsite, you will need to simulate the exhibits at
the museum using a PC (or another phone).
Select one of exhibits below by clicking on it; it will expand so that
you can scan it. Currently, there are
only a few exhibits set up.
Next, on your mobile phone, scan or click on
the QRC to the left or on “Prototype”
above. An app will appear. Press and
hold the yellow button labeled “Press to Scan”, frame the text description
under the expanded exhibit image while keeping the button pressed. Release the button when the text is
framed. This tells the app that you wish
to discuss that exhibit. The app will
audibly describe the exhibit. On iOS,
you may need to press the “Press to hear” button below the description.
You can ask questions verbally by pressing and holding the “Press to
Talk” button. Release the button when
you are done speaking. For more
instructions, see the Visitor’s Guide linked above.
Introduction — Read
more about what it is.
Sample Conversation — A sample transcript of a visitor’s session.
Sample Exhibit Description — A sample of training data.
Project Plan — What
next…
Architecture — How it works, for techies.
Text recognition – Scan image for text.
Audio — Record and play back speech.
Text to Speech — Multilanguage, but needs translation API.
Speech to Text — Multilanguage.
POC2 — Exhibits – Same software. Different exhibit collection.
POC3 — Manage – Exhibits – Image recognition version
Identify Image – Register Image – Image match tests.