Beaver Notes

Inspiration

We were inspired by our struggles in previous classes were some professors go over material too fast to be able to accurately take notes or don't explain the topic very well. One of the main challenges in these classes is that its very hard to multitask listening to the professor and writing notes. Therefore we wanted to make an app that would listen or take in lectures and display them to a webpage were the user would have access to all their previous notes in a detailed fashion to aid their studying and allow the user to focus on listening.

What it does

Our app specifically works by giving the user the ability to record audio in their surroundings, typically a lecture, and the app transcribes it to text and then uses that text with an AI to understand it an return to the user a summary and detailed notes of the audio provided. How we built it

How we built it

We built this app by using many different technologies. The app is a Vite + React app built mainly with JSX, JS, and CSS as the front end and Rest API and Express as the back end. We also used Google Gemini API for the summarizing and generation of detailed notes of the provided speech, and used a Speech to text API to be able to transform the speech into text and give it to Gemini. Then so the user can always access all their notes, we connected a MongoDB database using Mongoose to store all the lecture notes for the user.

Challenges we ran into

We ran into two major challenges throughout our hackathon: troubles with the API calls and issues merging our group work when working on similar files. These challenges boil down to inexperience. We were very ambitions and used technologies that we have never used. This was our first times using basically all the technologies listed above except JS, as we have coded in JS before. Therefore we ran into a lot of issues setting up the back end and connecting it to the front end, and then struggled connecting the Gemini API in he backend and using it from the frontend. It also didn't help that due to our inexperience in working in teams, we messed up a critical file that controlled a lot of the front end by messing up a merger and this caused us to have to revert back to and older version since we continued to work for about 3 hours before realizing that we had messed up the file resulting in about 6 hours total of work wasted on fixing the issue.

Accomplishments that we're proud of

Overall, as a team we are proud of ourselves and our product. This was the first project we have taken up from start to finish like this and we used the whole 24 hours given to try to deliver our best work. We ran into a lot of issues and at one point while fixing that terrible merger problem, we were really down and on the verge of giving up, but we went to an old version and re did it all again and persevered. Now, looking back we are proud of ourselves for not giving up continuing the challenge despite all the difficulties. Also, considering that we have never used any of these technologies, we think that we did a pretty good job of stepping out of our comfort zone and learning some new things to make something that we believe in.

What we learned

We learned a lot of important lessons during this hackathon, and are very glad we did this. We learned a lot more about the importance of communication and teamwork in software development. We ran into a lot of issues working as a team and realized that this is a skill we need to improve. We also learned how to use new technologies like API's, React a little bit, and things lie MongoDB. Overall it was a very fun learning experience as we got to try out a lot of things we never used before and learned a lot of our weakness as software developers and what we need to improve moving on. What's next for Beaver Notes

What's next for Beaver Notes

Beaver Notes unfortunately has not launched yet, but we will in the future as we implement more of our base features such as introducing a raw text file upload and raw audio file uploads (.mp3/.wav) that can be transcribed and summarized. Another goal will be to add a database of transcribed recordings from various courses at OSU and use the Gemini API to generate study flashcards for students to refine their understanding. With this as the backbone of our app, we can then dive into even more personalized learning by allowing our users to have the option of an AI making tests, flashcards, and study guides to continue learning material in different ways based on the lecture recording provided by students. Finally, we believe tha learning is better with friends, so we would like to add a social aspect to Beaver Notes were people can study together through completive minigames to encourage students to study.

Built With

• css
• geminiapi
• javascript
• jsx
• mongodb
• react
• restapi
• stt
• vite