Keen

Inspiration

The inspiration behind our application is to help individuals with the challenge of reading, especially those who are visually impaired.

What it does

Keen allows a user to take a picture of a portion of text, then transcribes the text found in the picture, and reads it aloud. The application can be used by numerous individuals, ranging from the visually impaired to simply people trying to learn a new language.

How we built it

The android application works by taking an image captured by a phones camera, and feeding said image into Google Clouds Vision Api to extract and identify text. Once the text has been obtained, the result is used, along with Google Cloud's Text-To-Speech API, to generate an mp3 file of the text being read. Finally the identified text is displayed and the mp3 file is played.

Challenges we ran into

One of the main challenges that we ran into was properly setting our Google credentials in order to access the Cloud APIs.

Accomplishments that we're proud of

We are proud that we were able to deliver a functional application, that works well in conjunction with Google's own TalkBack accessibility setting. Keen is able to recognize and read multiple different languages and provides helpful feedback to users.

What we learned

We learned how to use both the Vision and Text-To-Speech APIs from Google , and gained experience using android studio to create an android application in Java.

What's next for Keen

In the future, we would like to implement a settings page, were users could customize their app experience by allowing them to select their preferred playback speed, as well as the default language of the Google voice model.

Built With

• android-studio
• back-end
• front-end
• google-cloud-apis
• java