KnowTap

Inspiration

Teachers have not been able to one-one focus on every student and also students have been very shy about asking questions about the concepts that they do not know.

What it does

To help middle schools, high school and even college students to learn by taking pictures of real world and asking questions about it. Its very interactive and amazingly engaging.

How we built it

First, We use highly accurate image recognition (classifier based on supervised learning) to identify image. Second, we encourage the user to ask any type of question about that topic and help them find a solution.

Challenges we ran into

Image recognition, text to speech

Accomplishments that we're proud of

We have successfully built an application that is capable of recognizing images when a photo has been taken through a handheld device and we are able to give relevant information over a website or through Amazon Alexa ( for Lazy people ). We also provide a visualization trend of the percentage of people who have discussed about the image in the last 60 minutes using social media such as Twitter.

What we learned

Using Amazon Echo as a question answering system, difficulty in implementing text to speech and vice versa in cordova.

What's next for KnowTap

Since it is a prototype, the image recognition has to be improved with more training examples. Implement speech to text feature.

Note: It is an android app and it has not yet been uploaded to the app store

Built With

• apache
• bluemix
• ibm-watson
• google
• question-answering
• amazon-alexa
• twitter
• flask
• bootstrap
• mongodb
• c3js
• amazon-web-services