How we built it
Challenges we ran into
Embedding the original tweets was a major challenge since the API required was obscure and had little documentation. Another challenge was the SQL database for the chatroom since we were not very experienced with databases. The JS for the text-to-speech bot was not documented particularly well, which lead to trouble adding that to the website and deploying it with Flask. The Twitter API required API keys, which took a while to get, and it was not documented well either.
Accomplishments that we're proud of
We believe that we did a good job styling Twarble's pages to look like Twitter. We also think that we integrated the voice functionality well, which was difficult. Finally, we believe that Twarble's MS Paint logo perfectly captures the essence of the project.
What we learned
We learned a lot over the course of this project. We learned how to use the Web Speech API for text-to-speech, how to use a database for the chatroom, how to use the Datamuse API and the TweePy API, how to bodge together a lot of things in a short period of time, and how to work as a team without yelling at each other.
What's next for Twarble
Currently, Twarble queries tweets and translates them every time the site is loaded, which can make loading the webpage take several seconds. In the future, we wish to incorporate a cache that is updated every few minutes so loading times are drastically decreased.