What if I want to take an audio tour of a national park or a University campus on my own time? What if I want to take an audio tour of a place that doesn't even offer audio tours?
With Toor, we are able to harness people's passions for the places they love to serve the curiosity of our users.
What it does
We enable users to submit their own audio tours of the places they love, and we allow them to listen to other user submissions as well. Users can also elect to receive a text alert if a new audio tour has been updated for a specific location.
How we built it
We built the front-end using React, and back-end with multiple REST API endpoints using Flask. Flask then uses SQLAlchemy, an ORM to submit records to the SQLite3 database and query data to and from. The audio files are stored in Google Cloud Firebase database. The front end is also hosted on Firebase.
Challenges we ran into
Enabling users to listen to audio without having to repeatedly download the files was our first major obstacle. With some research we found that either an AWS S3 bucket or a Google Firebase database would solve our problems. After issues with permission with the AWS S3 bucket, we decided that Google Firebase would be a more apt solution to our issue.
Accomplishments that we're proud of
Enabling audio streaming was a big win for us. We are also proud of the our team synergy and how we got things done quickly. We also are proud of the fact that we applied a lot of the things we learned from our internships this summer.
What we learned
- Audio streaming, audio file upload
- Upload audio player on react
- Thinking about minimal viable product
- Soft skills such as interpersonal communication with fellow hackers
What's next for Toor
Adding the ability to comment on an audio tour, expanding the scope outside of just college campus, using Google Cloud Platform to implement Speech-To-Text and NLP to filter out "bad" comments and words in audio files.