Inspiration

Our inspiration for Savor came from our own experiences as home cooks. We often found ourselves staring blankly at the contents of our fridges, wondering what to make for dinner. We also recognized the issue of food waste, hunger, and how much perfectly good food was being thrown away because people didn't know what to do with it. We wanted to create an app to help solve both problems.

Savor was born out of a desire to reduce food waste and help people make the most out of the ingredients they already have. As a team of food enthusiasts and tech lovers, we recognized that there was a need for a more accessible and user-friendly approach to cooking and meal planning. We wanted to create an app that could take the stress out of meal planning and make cooking easy for everyone. With a simple tap of a button, Savor generates dynamic, delicious recipes based on the ingredients you have on hand, taking into account nutritional value and dietary preferences. Whether you're a beginner cook or a seasoned pro, Savor has something for everyone, and its user-friendly interface makes cooking easy and fun.

What it does

Savor is an innovative solution to the way people approach cooking and food preparation. Its AI-generated recipes allow users to make the most out of the ingredients they already have in their fridges, reducing food waste and making healthy, nutritious meals more accessible and affordable. Frequently, people order takeout from outside or spend a lot of money on excess groceries because they are unsure what to cook with the ingredients they already have at home. Savor solves this problem by providing users with a variety of recipes that are tailored to their specific dietary preferences and nutritional needs, while only using the ingredients they already have lying in their fridge. By using Savor, you'll never be at a loss for what to make for dinner again, and you'll be making a positive impact on the world by reducing food waste and helping address the issue of world hunger.

How we built it

Savor prioritizes a simplistic interface. The front-end is powered by Streamlit and is beautiful, responsive, and clean. Users have the option to list the ingredients they have on hand and their dietary restrictions in plain text, or simply upload a photo of their fridge. Savor uses a combination of computer vision through YOLOv5 and natural language processing through OpenAI's powerful gpt-3.5-turbo algorithm to identify the ingredients present in the photo and generate recipes based on the user's preferences. It also uses Google's Images API to find an appropriate image of each recipe generated for the user.

Challenges we ran into

We initially planned to use Stable Diffusion to generate images of each recipe our model generated. Unfortunately, most cloud Stable Diffusion APIs were too expensive for us to implement. Instead, we used Google's Images API to get a thumbnail for each recipe. Another issue we faced during development was maintaining a beautiful yet clean interface for the user. While Streamlit is an excellent way to make a front-end, it can be quite limiting at times. However, we prioritized an accessible front-end in our project and worked hard until this goal was achieved.

Accomplishments that we're proud of

We're very proud of our project's computer vision capabilities since this truly streamlines the process for the user and helps realize our main goal of making this project as easy to use for everyone as possible. We're also proud of our implementation of the Google API, since even though Stable Diffusion was not a viable solution for us we managed to persevere and make our program pleasing to the eye anyway. Finally, we're quite proud of our project's ability to take into consideration dietary restrictions on top of just the ingredients available, once again improving the accessibility of the project to everyone.

What we learned

During the development of this project, we learned about the sheer amount of food that goes to waste every single day. We have also learned about the importance of saving food and the immense nutritional value of a home-cooked meal. On the technical side of things, we have gained a lot of experience working with multiple APIs and stringing them all together with a beautiful front end.

What's next for Savor

We hope Savor will help many people around the world save food while simultaneously enjoying delicious, cheap, home-cooked meals. In the future, we hope this project helps aid in the fight against growing malnutrition and the huge world hunger problem that unfortunately persists today.

Built With

Share this project:

Updates