AI Meme Generator

• 

  User interface to a meme generator

• 

  Camera view that contains different filters

• 

  Captions are applied to the image

• 

  An image is uploaded and Captions are generated using AI

Inspiration

Memes are the language of the internet. But creating them often takes time thinking of captions, hashtags, and formats. I wanted to build a tool that automates creativity using AI so anyone can turn an image into a fun, shareable meme in seconds.

What it does

->Upload an image or take a live photo using the camera(include filters) ->AI generates funny captions, hashtags, and descriptions ->Apply filters like grayscale, sepia, blur, contrast ->Finalize a meme with your chosen caption ->BONUS: Use Text-to-Speech so the meme can “talk” in different voices

How we built it

->Frontend: React (deployed on Vercel) ->Backend: Flask (deployed on Railway) ->AI Models: OpenAI API for caption generation, hashtags, and descriptions ->Image Processing: Pillow for filters and meme generation ->TTS (Text-to-Speech): Web Speech API for voice playback

Challenges we ran into

->Setting up backend deployment on free-tier platforms (Railway, Render, etc.) ->Handling API rate limits ->Ensuring smooth integration between frontend and backend ->Making the UI responsive and beginner-friendly

Accomplishments that we're proud of

->Fully working AI meme generator that runs online ->Successful integration of AI captioning + hashtags + filters + TTS ->Deployed on free-tier platforms (no paid infra needed!) ->Clean, intuitive interface that anyone can use

What we learned

->Deploying ML/AI apps on serverless/free platforms ->Handling API rate limits and optimizing API usage ->How to integrate AI into creative apps ->UI/UX for fun, playful apps

What's next for AI Meme Generator

->Add multi-language support (Spanish, Hindi, French memes!) ->Improve AI humor with fine-tuned meme datasets ->Add GIF support and short meme videos ->Enable social sharing directly to Instagram/Twitter/TikTok ->Add voice cloning so memes can talk in celebrity voices

Built With

• api
• axios
• css
• descriptions)-image-processing:-pillow-(pil)-(filters-&-meme-rendering)-speech:-web-speech-api-(text-to-speech-playback)-version-control:-git-+-github-other:-axios-(http-requests)
• flask
• form-data-api
• hashtags
• html
• javascript
• openai-api
• pillow
• python
• python-ai-/-ml:-openai-api-(captions
• railway(backend-deployment)
• react
• tts-api.com
• vercel(frontend-deployment)