Amanda Dash, Nora Huang, Dany Cabrera,
Tristan Partridge and Maria Ferman
SignTalker
Slide 2
Slide 2 text
Persona: Laura
◎ Type: Co-Editor-in-Chief of y.a.m.
magazine
◎ Role: user
◎ Problem:
She can lip-read and uses sign
language. She loves her work, but
finds it very challenging to
communicate with people who don't
understand sign language.
2
Slide 3
Slide 3 text
Project Proposal
◎ People who belong to the deaf community find it
very difficult to communicate with non-deaf
people
◎ A common solution is to use text-to-speech tools,
but this does not allow for dynamic conversation
◎ Our Solution:
+ =
3
Slide 4
Slide 4 text
Deaf
user
Scenario
Non-deaf
friend
X ?
Do you
finish the
report?
1 2
3
4
Slide 5
Slide 5 text
Overall Architecture
5
Slide 6
Slide 6 text
Existing solutions
● Translator from words or sentence to ASL
○ ASL Translator
● Translator from ASL to words or sentence
6
Slide 7
Slide 7 text
Milestone 1
14/10/2016
Milestones
Milestone 2
4/11/2016
Milestone 3
2/12/2016
● Creating relationships with
the ASL community (users)
● Companion Website
● Machine Learning
framework
● Text to audio
● Finger point mapping at
60% accuracy
● Android App with
Calibration tool
● Implemented all features
● Increase accuracy to 80%
● Test framework
● Training the text prediction
● Fixing bugs and responding
users feedback
● Stretch goals (Tutorial)
● Improve UI visual elements
● Increase accuracy to 90%
7