Building an AI Audio Dubbing Web App Using Claude as a Coding Agent
Recently I built a small AI project that automatically generates dubbed audio from uploaded speech.
The idea behind the project is simple:
Upload an audio file → transcribe it → translate it → generate dubbed speech in another language.
What made this project particularly interesting was that most of the development was done with the help of Claude as a coding agent.
It feels like we are entering a world where ideas and imagination can quickly become real applications with the help of AI coding agents.
What the Project Does
The application allows users to:
- Upload an audio file
- Automatically transcribe the speech
- Translate the text into another language
- Generate dubbed audio using AI voice synthesis
- Play and download the generated dubbed audio
The goal of this project was to explore how quickly an idea can be turned into a working application using AI-assisted development.
AI Pipeline
The application follows this processing pipeline:
Audio Upload
↓
Speech-to-Text (OpenAI Whisper)
↓
Translation (OpenAI API)
↓
Text-to-Speech (ElevenLabs)
↓
Dubbed Audio Output
Tech Stack
Frontend:
- Next.js
- TypeScript
- Tailwind CSS
AI Services:
- OpenAI API (Translation)
- ElevenLabs API
Deployment:
- Vercel
Using Claude as a Coding Agent
A large portion of the development process was assisted by Claude.
Instead of writing every part of the code manually, I used Claude iteratively to:
- Generate UI components
- Implement audio upload and processing logic
- Integrate AI APIs
- Fix UI and runtime errors
- Debug deployment issues
The workflow looked something like this:
- Describe the feature I wanted
- Let Claude generate the initial implementation
- Test the code
- Ask Claude to refine or debug the result
This process significantly accelerated development compared to writing everything from scratch.
MVP Status
This project is currently an MVP (Minimum Viable Product), so there are still many areas that could be improved.
Still, it was fascinating to see how quickly an idea could turn into a working application using AI coding agents.
Live Demo
https://perso-ai-dubbing.vercel.app
GitHub Repository
https://github.com/YOUR_USERNAME/perso-ai-dubbing
Final Thoughts
Projects like this make me feel that we are entering a new era of software development.
With AI coding agents, developers can focus more on ideas and problem solving rather than writing every line of code manually.
It will be interesting to see how tools like Claude, ChatGPT, and other coding agents continue to change the way we build software.
Top comments (0)