ZapBase

Overview

ZapBase enables users to initiate Ethereum transactions using voice commands. The system integrates a smart contract that maps basenames (e.g., alice.base.eth) to wallet addresses, enabling secure and seamless voice-triggered asset transfers aligned with the theme of Next Generation of Autonomous Businesses. Our focus is on leveraging AI and blockchain technologies to create impactful and scalable solutions. This application enables users to:

• 🎤 Use voice commands like “Send 0.01 ETH to jane.base.eth”
• 💸 Trigger on-chain ETH transactions
• 🔗 Interact with a smart contract for name-to-address resolution

---

Getting Started

Access the Telegram bot here

To check the app is running, you can ping: you will get the response Bot is running.

https://t-mini-app.onrender.com/health

Basename Resolution Smart Contract

A Solidity smart contract that stores and resolves basenames to Ethereum addresses.

mapping(string => address) public nameToAddress;

function getAddress(string memory name) public view returns (address);

Voice-Driven Transaction Flow

1. User speaks:
   “Send 0.02 ETH to jane”

2. Frontend captures and sends:

   {
     "basename": "jane.base.eth",
     "amount": "0.02"
   }
   

3. Backend (/send endpoint):

   • Uses web.js to interact with SC to resolve jane.base.eth → 0xReceiver
   • Calls sendAsset("0xReceiver", "0.02")
   • Responds with transaction hash

Voice recognition is handled externally (e.g., browser or mobile), with text commands sent to this backend for blockchain execution.

Project Goals

• Develop an AI-powered autonomous agent as the core of our solution.
• Ensure deployment on Base with on-chain execution where applicable.

Steps of Action

1. Research & Ideation: Explore the problem space and finalize our project idea.
2. Development: Build the AI agent and integrate it with blockchain components.
3. Testing & Deployment: Ensure functionality and deploy on Base.
4. Submission: Prepare the required materials (demo, repository, pitch deck).

---

Repository Structure

• /backend: Source code for website.
• /frontend: Documentation and pitch deck.
• /frontend/src/api: Source code for AI agent.
• ens-asset-sender : Source code for Smart contracts.

---

Front-end

How to run

Start the server

cd frontend
npm run dev

Speech-to-Text Backend API

Overview

This backend service provides audio transcription capabilities using the Groq AI platform. It accepts audio file uploads and returns text transcriptions along with timestamps for segments and individual words.

Prerequisites

• Node.js (v14 or higher)
• npm
• A Groq API key (sign up at https://console.groq.com/)

Setup

1. Install dependencies

   npm install

2. Configure environment variables

      GROQ_API_KEY=your_groq_api_key_here
   

3. Start the API server

   cd frontend
   npm run start:api
   

Using the API Endpoint:

• Transcribe Audio

  POST /api/aiagents/transcribe

Example using cURL:

curl -X POST http://localhost:3000/api/aiagents/transcribe   -H "Content-Type: multipart/form-data"   
-F "audio=@/path/to/your/audio/file.mp3"

Telegram Bot Backend API

Overview

This backend service provide telegram bot. It

Prerequisites

• Go 1.23.4
• A Gemini API key (sign up at here)

Setup

1. Install dependencies

   go tidy

2. Configure environment variables

   GEMINI_API_KEY=your_gemini_api_key_here
   TELEGRAM_BOT_TOKEN=your_bot_token_api_key_here
   PUBLIC_URL=hosted_website_url
   

3. Start the API server

   cd backend
   go run main.go
   

To check bot is running well, run:

curl -X GET http://localhost:9000/health

Team

• Blessing Chika - Product Manager
• Allan Robinson K G - Frontend Dev
• Kenneth Njoroge - AI Dev
• Fred Gitonga - Backend Dev
• Marreyharm - Designer