Last summer I tried gardening for the very first time. Armed with recycled coffee cups and cheap soil, I germinated store-bought seeds, tracked frost dates and planted the surviving seedlings in my small yard.

I quickly realized that plants showed different personalities based on their care and needs. Tomatoes and marigolds almost craved a bit of neglect, while basil and roses would show me the middle leaf and dramatically wilt when the pH level wasn’t just right.

If basil could have spoken, she would surely have been a cranky one, always complaining about something! So when I came across the voice AI hackathon opportunity, I thought: why not use AI to make a sassy plant talk to me?

Meet the one and only, Lady Monstera Deliciosa! She is native to the tropical forests of southern Mexico, and is also known as the “Swiss cheese plant” due to the holes or “eyes,” that develop in her leaves, resembling the holes in Emmental cheese.

These holes and splits are called fenestrations, and they are an evolutionary adaptation that help the plant maximize sunlight and let rain pass through to its roots in dense jungle environments.

With her dramatic and "cheesy" green leaves, it was easy to pick my Monstera plant as the perfect sassy talking plant!


In order to understand the plant's state at any point in time, I connected a phototransistor and a soil moisture sensor to an Arduino. Here is the schema:

The Arduino collects the sensors data and sends it over to the back-end (a Python FastAPI application) via asynchronous serial communication. The data is then processed and displayed in React on the front-end. I used a combination of WebSockets and Http for communication across the stack. I am fairly new to WebSockets, so it was really cool implementing a two-way street communication channel between back-end and front-end.

I used GPT-4o by OpenAI to handle text-to-speech (TTS) communication and give Lady Monstera her voice.


Seeing the plant’s internal state read in real time was absolutely thrilling! I tracked water and light changes over 5-second intervals, and let the plant comment on significant state changes. Here is how Lady Monstera recently handled an abrupt dimming of the lights:





As you can see, she is also capable of detecting changes in light mid-speech but doesn’t send a talking notification while she is already talking.

And here is how she reacted to being watered!

Notice how she records and updates the last time she was watered. Anything to remind you how sassy she is, and how she won’t forgive you for forgetting to water her!

• I opted for text-to-speech instead of realtime speech. Realtime was very sensitive to ambient noise, difficult to control and it quickly became too expensive to experiment with.
• It would be fun to gamify the experience, with multiple plants talking to each other, each with distinct personalities based on their needs. Maybe even connect hobbyist gardeners and let their plants gossip about their owners 🤭
• Ultimately, I hope this project helps us get closer to our house plants and the nature around us, reduce house plant neglect and make interacting with our green friends a little more entertaining and curiosity driven.

Waterloo Voice AI hackathon was my first ever real-world hackathon and I loved every minute of it. Coding next to other hackers, exchanging ideas and simply absorbing the energy and enthusiasm in the room has left an indelible mark on me. It has given me the confidence to keep building, no matter how silly or small the initial idea might seem.

A big thank you to the organizers and judges, in particular Ian Pillon, Mike Bird, Ti Guo and Prabal Gupta for your dedication and support.

Keep hacking and see you all at the next one!

note
As a rule of green thumb (yes, I am proud of this one 🙈🌱), don't leaf your plants unattended for too long. Also, singing to them improves their greenery and promotes luscious, beat-loving flowers!