Skip to content

AlexisLasselin/Algosup-Moonshot-Project

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ALGOSUP Moonshot Project

Context and Problematic

There are over 70 million deaf people in the world. Although there are 121 sign languages, these are rarely learned by the people concerned, and even less by people without this disability. When we ask the people concerned why, it's for two main reasons:

  • A lack of education about it
  • A lack of usefulness, because few people can understand them.

My solution

Although I can't help people learn sign language, I want to help people who can speak it by enabling them to communicate with as many people as possible.

How can I do this? My idea is to design an application that translates sign language into a natural language, like Google Translate. I know that there's a company called SignAll, that has partially implemented my project. But this company works more in B2B and requires a certain amount of hardware, whereas my idea would be to make it in the form of a mobile application to democratize this.

My aim for the very first version would be to translate LSF, the French Sign Language,then go the ASL, the American Sign Language as it is the most widely signed in the world with almost half a million signers, into English.


Action plan

Specific short-term goals (1-3 weeks)

  • Send requests to all French deaf Universities to get info about the various LSF signs
  • Contact Elix for the same reason
  • Find a dataset of LSF signs
  • Create an AI model

Concrete actions to take starting tomorrow

  • Prepare my emails for LSF organizations
  • Send a DM to Franck Jeannin, asking him if I might have the contact of AI "experts" from the Jury of the Bac+4 oral presentation.

Possible obstacles and anticipated solutions

  • I might not find a dataset of LSF signs
    • I will create my own dataset by asking the community and the teachers from my LSF classes to help me.
  • AI model might not exist for my goal.
    • I will create it myself, with help from Franck's contacts.
  • I might be giving up too quickly.
    • By giving small goals, I will be able to see the progress I make and keep my motivation.
  • I might not have enough time.
    • I will work on it every day for an hour, and try to make a big session on Saturday, keeping my Sundays off.
  • I might not have enough money.
    • I will try to find free resources, and if I can't, I might reconsider the importance of this ressource in my project.
  • I might not have enough knowledge.
    • I will learn as I go, and ask for help when I need it.
  • I might not have enough motivation.
    • I will send my progress to a classmate every week, allowing them to ask me about it, and I will also keep a journal of my progress.

How I will measure my progress

  • I will have sent all my emails by the end of the week.
  • I will have received an answer from at least one of the organizations before the holidays.
  • I will explain my project to the experts by the end of the week.

For later, I will have other progress indicators, but for now, I will focus on these.


About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published