Inspiration
What inspired me to start this project is the desire to solve real problems in my local community. Most people around me are farmers, and they don’t have quick or reliable access to agricultural knowledge or AI-powered solutions. I saw this as a great opportunity to build a solution that empowers them — especially those living in local areas with limited internet or tech access.
What it does
AgriVoice AI Companion is an AI-powered offline farming assistant that helps farmers solve agricultural problems through voice or text. It supports local languages, provides farming insights, and offers solutions on crop management, irrigation, and soil care — all while being accessible even in areas with poor internet connection.
How we built it
I started by analyzing my surroundings and identifying the most common farming challenges. Then I decided to use my web development and AI skills to build a solution around that. I designed the user interface first to make it simple and beautiful, then integrated APIs for AI chat and translation. I added local language support to make it usable by a wider audience. Technologies used include HTML, CSS, JavaScript, Node.js, Express.js, and AI APIs.
Challenges we ran into
I faced challenges especially when integrating the APIs. The prompt API was quite complex to understand and slow to respond — sometimes taking up to 5–10 seconds. The translator API also lacked support for local languages like Hausa, and required detecting both source and target languages manually. The language detector API returned array-based responses, which made it tricky to handle. Finally, I wanted the AI to process crop images and provide solutions, but I was limited by the multimodal API which didn’t work as expected and needed fine-tuning.
Accomplishments that we're proud of
I’m proud that I was able to build this project entirely by myself despite all the limitations. I successfully created a working AI-powered offline web app that can actually make a difference for farmers. I also managed to integrate multiple APIs, design a full PWA, and bring my idea to life using just my knowledge and determination. Completing this project at a young age with limited resources is a huge accomplishment for me.
What we learned
I’ve learned that building AI-powered applications that run offline can create a big impact on accessibility. I also learned how to work with new APIs, handle complex API structures, and optimize for slow responses. This project helped me understand the power of PWA technology and how AI can be integrated into web apps to solve real-world problems.
What's next for AgriVoice AI
In the future, I plan to make AgriVoice AI more powerful and versatile — supporting more local languages and improving the accuracy of its AI responses. I also want to add image recognition features so farmers can upload pictures of crops and get instant solutions. My vision is to make AgriVoice AI a platform that other websites can integrate using iframes, allowing it to reach even more users and create a real impact in agriculture.
Log in or sign up for Devpost to join the conversation.