Welcome to OpenVoiceLink Documentation

Documentation Overview

OpenVoiceLink is a powerful AI voice assistant that combines local natural language processing with the flexibility to access hundreds of the latest AI models through Openrouter. This documentation will help you master all aspects of the application.

Getting Started

OpenVoiceLink is a desktop AI voice assistant that lets you have natural voice conversations with AI models.

System Requirements

• Operating System: Windows 10 or later (64-bit)
• Processor: 64-bit processor
• Memory: 8 GB RAM minimum (16 GB recommended for local models)
• Storage: 2 GB available space
• Microphone: Built-in or external microphone
• Internet: Required for cloud AI services

Installation

1. Download the latest version for Windows
2. Run the installer and follow the setup wizard
3. Launch OpenVoiceLink - it will start in Expanded mode by default

Initial Configuration

1. Configure your AI service (OpenRouter recommended)
2. Select your microphone and speaker
3. Set up voice activation (Push-to-Talk or Continuous Listening)

Core Features

Window Modes

OpenVoiceLink offers two main window modes:

• Expanded Mode: Default mode with settings panel and resize grip
• Fullscreen Mode: Immersive experience, press F11 to toggle

Voice Features

Configure voice input/output for the best experience:

• Voice Input: Speech-to-text using Whisper
• Voice Output: Text-to-speech with multiple providers
• Voice Activation: Push-to-Talk or Continuous Listening

AI Services

OpenRouter (Recommended)

Unified API for hundreds of AI models:

1. Get an API key from OpenRouter.ai
2. Enter your API key in Settings > OpenRouter tab
3. Choose your preferred model from the dropdown

Local Models

Run AI models locally for privacy and offline use:

1. Install Ollama from ollama.ai
2. Pull a compatible model (e.g., ollama pull llama3.2)
3. Configure Ollama URL in Settings

Settings & Troubleshooting

Key Settings

• General: Audio input/output, voice activation, theme
• OpenRouter: API key, model selection, parameters
• Local Model: Ollama configuration, model selection
• TTS: Text-to-speech provider and voice selection
• Hotkeys: Customize keyboard shortcuts

Common Issues

• Windows blocked the app: Right-click the .exe, choose "Properties", check "Unblock", then click "Apply"
• Microphone not detected: Check Windows privacy settings
• API key invalid: Verify key and connection
• Slow responses: Try smaller models or local models
• No audio output: Check speaker selection and test output

Next Steps