Deploying Whisper App locally involves completing the following four key steps:
- Getting the source code: via the Git command
git clone https://github.com/Nutlope/whisper.git
Clone the project repository and use thecd whisper
Go to the project directory - environmental preparation: You need to pre-install Node.js (LTS version is recommended), run the
npm install
Install Next.js, Vercel AI SDK, and other dependencies - Service Configuration::
- Register the Clerk service to get the authentication key, configured in the .env.local file
CLERK_SECRET_KEY
cap (a poem)NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY
- Create a Convex project and set up
CLERK_ISSUER_URL
- Register the Clerk service to get the authentication key, configured in the .env.local file
- launch an application: Implementation
npm run dev
Starts the development server, by default, via thehttp://localhost:3000
interviews
Note: Windows systems need to check microphone permissions (Settings→Privacy→Microphone), and it is recommended to prepare at least 2GB of free memory for the first time use. If you encounter API connection problems, you need to check the service status of Together.ai.
This answer comes from the articleWhisper App: free speech-to-text & AI note organizer toolThe