NoteWriter - 71

Front page of our website

Inspiration

As a group of high schoolers, our lectures often force us to choose between listening and note-taking. So we decided to try to take on this problem by creating notetaking software that is accessible to everyone.

What it does

Using OpenAI Whisper, we are able to translate recordings into very accurate transcriptions. Then once the transcription has been created, we utilize OpenAI’s GPT-4 8K API to generate relevant notes or summaries. Finally our application has a modern and minimalistic design, ensuring ease of use.

How we built it

Together we first connected to OpenAI Whisper and created our web design. We then added a way for the text from Whisper to give it to ChatGPT to then summarize/take notes. Then we added finishing touches and bug-fixed it.

Challenges we ran into

Throughout the process, we ran into a number of challenges to seemingly random bugs with PyAudio and spacing on the website. We had problems getting the audio file from the front end to the back end.

Accomplishments that we're proud of

We are really proud of our final product and how everything came together. We're proud of the effects we added and how polished everything feels.

What we learned

We learned that communicating and working together are key. We had a lot of challenges, but together we problem-solved, and even though we had some low moments we always made sure to have fun and cheer each other up.

What's next for NoteWriter

We want to make it an official website that anyone can access and make some UI changes. We would also like to potentially try to make our own OpenAI Whisper. In addition, we would like to make it easier to get the summaries of youtube videos.

Built With

css
gpt4
html
javascript
python
whisper

Submitted to

Los Altos Hacks VII

Created by

I worked on the front-end for most of it, which made me learn html and css better. I also worked on the backend at the beginning to get it started, which was the first time I worked with JavaScript and Flask.

Grant Kruttschnitt
I worked on the live audio transcription, getting the audio from the browser to the python server, and connecting the frontend to the backend, which I had never done with these languages before

Beat Zoid
I worked on the front end for the first time learning HTML and CSS. I also worked on the backend with Flask also for the first time.

Christopher M
I worked on the API calls and GPT-4 integration. I used my limited beta GPT-4 API access to facilitate our application. I also helped to create the main UI and worked on the audio processing.

Private user

Updates

Grant Kruttschnitt started this project — Apr 09, 2023 09:49 AM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.