A tool that allows users to better understand their surroundings using image processing and speech recognition. Utilizing both Google Cloud Vision API and Houndify's Speech-To-Text API, VisAids lets users ask it questions about their environment and will respond with an answer. It does this by leveraging OpenCV to capture images and by using Text-To-Speech to respond to user queries.

Make sure you install the following packages before running:

Authenticate with Google Cloud API prior to running code:

Use this line: export GOOGLE_APPLICATION_CREDENTIALS=“/home/user/Downloads/[FILE_NAME].json”

Run your program with this line of code:

rec -p | sox - -c 1 -r 16000 -t s16 -L - | ./ CLIENT ID CLIENT KEY

-Replace CLIENT ID and CLIENT KEY with your unique keys on your Houndify Dashboard

Built With

Share this project: