39 lines (27 loc) · 1.23 KB

AI Audio Transcriber (Gemini 2.5)

A Next.js application that transcribes audio files using the Google Gemini 2.5 Flash model, featuring automatic speaker detection.

🚀 Deployment (Vercel)

Push to GitHub: Create a new repository on GitHub and push your code.
Import to Vercel: Go to Vercel and import your repository.
Environment Variables: Add the following variable in the Vercel project settings:
- GEMINI_API_KEY: Your Google AI Studio API key.

⚠️ Important Note on File Sizes

Vercel's Serverless Functions (Hobby plan) have a 4.5MB payload limit.

Local Development: Works for large files (up to 100MB as configured).
Vercel: Uploads larger than 4.5MB may fail on the standard plan. For production with large files, consider using Vercel Blob or direct client-side uploads to a storage bucket (S3/GCS).

🛠️ Local Development

Install dependencies:
```
npm install
```
Setup environment: Create a .env.local file and add:
```
GEMINI_API_KEY=your_actual_api_key_here
```
Run the app:
```
npm run dev
```

📄 License

MIT