Frequently Asked Questions

Libre Voice Note App is a privacy-focused voice recording and transcription solution that runs on your own server. It includes:

• An Android mobile app for recording voice notes
• A Windows server for processing recordings
• AI-powered transcription using Whisper
• Additional AI features for organizing and summarizing your notes

Unlike cloud-based alternatives, Voice Note App keeps your data completely private by processing everything on your personal server.

Libre Voice Note App stands out with these key differences:

• Self-hosted: Your data never leaves your personal network
• Smart synchronization: Record anywhere, sync automatically when you return home
• AI-powered organization: Generate summaries, mindmaps, and action items
• Full control: You decide which AI models to use and how your data is processed
• No subscription fees: One-time purchase with the option to upgrade to Pro

Yes, the mobile app works offline for recording voice notes. You can record anywhere, even without an internet connection. Notes are stored securely on your device and will automatically sync with your server when you reconnect to your home network (or manually when you choose).

However, transcription and AI features require your server to be running, so those processes will only complete after synchronization.

For optimal performance, we recommend:

• Windows 10/11 PC with NVIDIA GPU
• At least 8GB of RAM (16GB recommended)
• 10GB of free disk space for installation and storage
• Processor: Intel i5/AMD Ryzen 5 or better

NVIDIA GPUs significantly accelerate the transcription process, especially for larger models, resulting in faster processing times and lower CPU usage.

Yes, you can run Voice Note Server on computers without a dedicated GPU. In this case, you'll need to configure the server to use WhisperX which is optimized for CPU processing.

To enable CPU optimization:

1. Open the Server Manager after installation
2. Navigate to Settings > Transcription
3. Select "WhisperX" as the transcription engine
4. Choose a smaller model size (tiny or base) for better performance

Yes, the server can run on laptops that meet the minimum requirements. Keep in mind:

• For optimal performance, connect to power when processing recordings
• Configure the laptop to prevent sleep/hibernation when the lid is closed if you plan to keep it running
• Consider using a smaller Whisper model (tiny or base) to reduce resource usage

Laptops with NVIDIA GPUs will provide better performance than those with integrated graphics only.

Currently, the official Voice Note Server package is only available for Windows. We're working on versions for other platforms in future updates.

Advanced users with technical knowledge can run the server components on Linux or Mac by directly using the open-source code available in our GitHub repository, but this is not officially supported and requires manual configuration.

Plugins are AI-powered tools that enhance your voice notes by automatically generating useful artifacts from your transcribed content. The server can call a Large Language Model (LLM) to process your notes in various ways:

• Summary Plugin: Creates concise summaries of longer recordings
• Mindmap Plugin: Generates visual mindmaps showing relationships between concepts
• Items Plugin: Extracts action items, to-dos, and key points as a structured list
• Correction Plugin: Fixes transcription errors and improves formatting

Plugins can run automatically after transcription or on-demand when you request them.

To enable plugins, you need to set up LLM API access:

1. Obtain an API key from an LLM provider (OpenAI, Anthropic, etc.) or set up a local LLM server
2. In the Server Manager, go to Settings > AI Plugins
3. Enter your API key or local server connection details
4. Select which plugins you want to enable and whether they should run automatically
5. Save your settings

Yes, Voice Note Server supports connecting to locally-hosted LLM servers. You can use applications like:

• LM Studio: A desktop application for running local LLMs
• Ollama: A lightweight LLM runner with various model options
• Other compatible servers: Any server that exposes a compatible API

To configure a local LLM:

1. Install and set up your preferred local LLM application
2. In Voice Note Server Manager, go to Settings > AI Plugins
3. Select "Local LLM" as the API provider
4. Enter the server URL (typically http://localhost:port)
5. Test the connection and save your settings

Voice Note Server supports the following LLM providers:

Provider	Supported Models	Notes
OpenAI	GPT-3.5-Turbo, GPT-4	Requires API key from OpenAI
Anthropic	Claude models	Requires Anthropic API access
Local (LM Studio)	Various open-source models	Runs locally, no API key needed
Ollama	Llama, Mistral, etc.	Runs locally, no API key needed
Custom	Any OpenAI-compatible API	For advanced users with other providers

Setting up Voice Note Server is straightforward:

1. Download the server package from our downloads page
2. Run the installer and follow the prompts
3. Launch the Server Manager from your Start Menu
4. Configure basic settings (ports, storage location)
5. Create an administrator account
6. Download a Whisper model for transcription
7. Generate an SSL certificate for secure connections

For detailed setup instructions, see our Server Setup Guide.

To connect your mobile app to your server:

1. Install the Voice Note App from Google Play Store
2. Open the app and go through the welcome screens
3. When prompted, enter your server's IP address or hostname
4. Enter the HTTPS port (default is 443)
5. Enter your username and password
6. Tap "Test Connection" to verify connectivity

Yes, there are several ways to access your notes from outside your home network:

• VPN Connection: Set up a VPN server on your network and connect through it
• Port Forwarding: Configure your router to forward the server's port (advanced)
• Reverse Proxy: Use a secure reverse proxy service (advanced)

For security reasons, we recommend using a VPN for remote access rather than exposing your server directly to the internet.

If you're having connection issues:

1. Verify the server is running (check Server Manager)
2. Confirm your mobile device and server are on the same network
3. Check that you've entered the correct IP address and port
4. Ensure your firewall isn't blocking the connections
5. Verify the SSL certificate was generated correctly
6. Try restarting both the server and the mobile app

The server logs (in Server Manager > Logs) can provide additional details about connection attempts.

If you're experiencing slow or failed transcription:

• Try using a smaller Whisper model (tiny or base)
• If running on CPU only, make sure you've selected WhisperX engine
• Check that your server meets the minimum hardware requirements
• Ensure no other resource-intensive programs are running simultaneously
• Verify the audio file uploaded correctly (check server logs)

For NVIDIA GPU users, make sure you have the latest CUDA drivers installed.

If your AI plugins aren't functioning:

1. Verify your API key is valid and hasn't expired
2. Check that you have sufficient credits/quota with your API provider
3. Ensure your server can access the internet (for cloud APIs)
4. For local LLMs, confirm the model is loaded and the server is running
5. Check the server logs for specific error messages

Also verify that you haven't reached the plugin limit for your account type (free accounts can only use one plugin).

Voice Note App is designed with privacy and security as core principles:

• All processing happens on your personal server, not in the cloud
• Data is encrypted during transfer between your mobile device and server
• SSL certificates secure the connection
• Authentication is required for all access
• No data is shared with third parties without your explicit configuration

When using cloud-based LLM APIs (like OpenAI), note content may be sent to those services for processing. If complete privacy is required, use local LLM options instead.

No, Voice Note App does not send any of your voice recordings, transcriptions, or note content to our servers. The app is entirely self-contained and processes everything locally on your own server.

The only data sent to our servers are:

• Anonymous usage statistics (if you opt in during setup)
• Error reports (if you choose to send them when an error occurs)
• License validation for Pro users

These are all optional and can be disabled in the settings.