Our group was inspired to launch Projectile as a means to provide all students with equal access to learning from lecture. More often than not, our hearing-impaired peers are left in the dark and require school-appointed translators to guide them through the lecture. Since most professors commonly project their lecture slides and other supplementary materials on screens, it only make sense to provide their words as well.

What it does

The professor talks into the mic. As he talks, the students can glance at their phones to see a live transcript of what the professor is saying.

How we built it

The professor talks into the mic, which then relays data to the arduino. The arduino makes a .wav file out of the data and sends it to the server. The server leverages IBM Watson's Speech-to-text API to convert the wav into a string. In this prototype phase, we just append the string to a file on the server. The server then serves the file contents to the ios device.

Challenges I ran into

Bluemix NA went down, which meant we had to deploy to Europe. We also had to learn how to deploy using bluemix. We also had to pick up the Watson Speech-to-text API, which was not too hard. Ankur ran into some problems on iOS involving Transport Layer Security due to the new TLS rules on iOS 9, but he overcame that. We ran into a lot of problems on the hardware end. First, we had a huge amount of problems connecting the onboard wifi chip to the wifi due to how spotty the wifi was. We had to set up a hotspot and connect the arduino to the hotspot instead of the hackathon's wifi. We ran into a lot of mic problems, so we had to replace our mic. We then had a bunch of problems loading the software we needed onto the arduino board.

Accomplishments that I'm proud of

On the software end, we were able to pull together a bunch of technologies and leverage them well. On the hardware end, we were met with a lot of problems and challenges that took a lot of time and effort to figure out, but in the end we solved a lot of them!

What we learned

We learned a lot about leveraging technologies like IBM Watson, as well as learned a bunch about different hardware problems and how to deal with them.

Share this project: