VSR

Visual speech recognition aims to recognize the content based on lip movements, without relying on audio stream

Simplified: Takes the movment of your lips to interpret what you are saying and then generates text according to that, doesn’t use background audio at all so that doesn’t come in play

MakeUofT project §

Use Visual Speech Recognition (VSR) to interpret text and do something with it. This is going to be an android app so be considerate of that.

We can use VSR and then use arduino with all of it’s sensors to do something

Structure §

• Phone
  • Android App
    • Camera
      • VSR

        • Text from VSR §

Idea:

During a zoom meeting when you want to communicate ideas rather than interrupting someone while they are talking we can use VSR to read people’s lips and display the text on the screen