A full-stack application that allows users to upload audio files, transcribe speech using AI (Whisper), and receive the transcription securely via email.
This project demonstrates real-world system design, not just a demo:
- Frontend UI
- Backend API
- AI model inference
- Email automation
- Cloud-ready architecture
Users enter their email and upload an audio file (MP3 / WAV / M4A).
The file is sent to the backend, processed using an AI speech-to-text model, and queued for delivery.
Once completed, the transcription is emailed to the user.
Recent uploads are tracked in the UI.
- The frontend handles file uploads and UX
- The backend processes audio using Whisper
- Transcriptions are emailed automatically
- No audio is stored permanently
- Drag & drop audio upload
- Email-based delivery of transcription
- AI-powered speech recognition (Whisper)
- Upload history tracking
- Secure handling of files
- Responsive, clean UI
- React
- Vite
- CSS (custom, no UI frameworks)
- FastAPI
- OpenAI Whisper
- Uvicorn
- Python
- SMTP (email automation)
- REST APIs
- Environment-based configuration