Every time I've had an idea for a video or piece of content, I struggled with how to start or what to say. I soon realized many of my friends had the same challenge: while they could use AI to generate images, voices, or even videos, they didn’t know what to actually say or write.

This common issue inspired me to create Vocal Script AI — a tool that helps content creators transform raw ideas into structured scripts with natural voiceovers.

Vocal Script AI is an intelligent Adobe Express Add-on that allows users to:

• Speak or type their idea
• Generate a structured script
• Instantly hear it as an AI voiceover

All in one seamless experience, powered by modern AI models.

1. 🎤 User records or types their idea
2. ✍️ The tool turns it into a structured, readable script
3. 🔊 A voiceover is generated using AI models
4. 🎞️ The user can export or embed the script in their video content

vocal-script-ai/

├── assets/

├── dist/

├── public/

├── src/

├── .gitignore

├── package-lock.json

├── package.json

├── tsconfig.json

└── README.md

Watch the demo video →

• Migration to paid AI models for higher quality
• Clean architecture refactor (modular & SOLID)
• Better UI to match Adobe Express design
• Persistent user data & storage
• Launch on Adobe Express Add-on Marketplace

• Frontend: HTML, CSS, Vanilla JS
• Backend: Node.js, Express.js, TypeScript
• AI Models: HuggingFace (text gen & TTS), OpenAI via HuggingFace
• Audio: MediaRecorder API, kokoro-js (TTS), onnxruntime-node
• Dev Tools: dotenv, tsx, nodemon