This project is my Day 1 submission for the Murf AI Voice Agents Challenge.
I built a simple but fully functional voice agent with a custom backend and a premium UI.
The agent converts text into speech using the Murf TTS API and plays the generated audio in the browser.
- Custom frontend (HTML + CSS + JavaScript)
- Custom backend (Node.js + Express)
- Integrated Murf TTS API (Gen2 voice model)
- Audio playback in browser
- Typing animation for better UX
- Dark mode premium UI
- Ready for future challenge tasks
Frontend:
- HTML
- CSS
- JavaScript
Backend:
- Node.js
- Express.js
- node-fetch
TTS Engine:
- Murf AI (Gen2 voice model)
├── index.html # Frontend UI ├── server.js # Backend server (Express) ├── package.json # Dependencies ├── package-lock.json └── README.md
Run this inside the project folder:
The backend will run on:
Double-click index.html to open it in any browser.
- User enters text
- Frontend sends text to backend
- Backend calls Murf TTS API
- Murf returns a temporary MP3 URL
- The browser plays the audio
- Typing animation and UI make the experience smooth
This project is built as part of the:
Murf AI Voice Agents Challenge – Day 1
Task: Build a working voice agent and post the demo on LinkedIn.
- Add microphone input
- Add chat history
- Improve response speed
- Add personas
- Add streaming TTS
- Integrate LLM (Gemini/OpenAI)
Aditya (aditya-ai00)
Participant of Murf AI Voice Agents Challenge – 10 Days of AI Voice Agents