This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend
What I Built
One of my college friends has amnesia, which makes remembering everyday things a real challenge.
That got me thinking: what if I could build something that helps her keep track of important information, save her thoughts, and find things whenever she needs them?
That's how EchoMemo started.
EchoMemo is a voice-first AI assistant that turns simple recordings or typed thoughts into organized notes, summaries, and tasks.
Some of its key features:
- 🎤 Voice recording with automatic transcription.
- 🧠 AI-powered summaries and task extraction.
- 📥 Smart Inbox to organize captured thoughts.
- ✅ Task Manager with priorities and deadlines.
- 💬 Ask Echo, a chatbot that answers questions using personal notes through RAG.
- 🔊 Voice-based questions and spoken AI responses.
The idea is simple: make it easier to store and retrieve information without having to remember everything.
Demo Video
Deployed Link
Code
GitHub:
🎙️ EchoMemo
A voice-first AI second brain for people who can't keep up with life's noise.
Built for a friend who forgets everything — lectures, meetings, ideas, deadlines.
💡 The Story
My college friend can never remember things — she records lectures but never reviews them, jots ideas on napkins that vanish, and misses deadlines because tasks live in 12 different places.
EchoMemo is the assistant I built for her: speak or type a thought, and AI turns it into organized notes, extracts every action item, and lets you search everything with natural language.
✨ Features
| Feature | What it does |
|---|---|
| 🎤 Voice Capture | Record voice memos → auto-transcribed via ElevenLabs Speech-to-Text |
| 📝 Smart Notes | Rich text notes with tags, full-text search, and markdown editing |
| 🧠 AI Processing | Gemini summarises notes, classifies them, and extracts actionable tasks |
| 📥 Smart Inbox | AI-processed notes sorted by category (task, idea, meeting, journal…) |
| ✅ Task |
How I Built It
I built EchoMemo using React, TypeScript, FastAPI, and MongoDB Atlas, and deployed it on Render.
For the AI features, I integrated:
- Gemini for summaries, task extraction, embeddings, and RAG.
- MongoDB Atlas for storing notes and retrieving relevant information using Vector Search.
- ElevenLabs for speech-to-text and text-to-speech.
One of the most interesting parts was connecting all these services into a single workflow, especially making the chatbot retrieve relevant information from a user's own notes instead of giving generic answers.
Why Does Open Innovation Matter?
I wanted to keep EchoMemo flexible instead of depending entirely on one technology.
Using open-source frameworks and a modular architecture makes it easier to experiment with different models, tools, and databases in the future.
It also gives me more control over how EchoMemo evolves and how it can be adapted to different needs.
Prize Categories
MongoDB Atlas: Powers EchoMemo's database and vector search, helping users retrieve relevant notes through semantic search.
ElevenLabs: Enables voice-first interaction with speech-to-text transcription and realistic text-to-speech.
Render: Hosts and deploys EchoMemo, making the application accessible online.
Built with ❤️ for a friend who inspired the idea.
EchoMemo — Capture it now. Find it when it matters.
Top comments (0)