About Verba

Verba is a revolutionary language learning application that allows users to practice speaking with an AI companion. This AI is designed to be endlessly patient, non-judgmental, and can operate entirely offline, ensuring privacy and accessibility. The application facilitates full conversations, provides inline corrections, generates graded reading materials, automatically captures new vocabulary, and offers a natural voice interaction on both ends of the conversation. Verba runs entirely on the user's machine, making it a truly local-first and offline-first solution.

Key Features:

• Offline Functionality: Integrates with local AI providers like Ollama, allowing for complete functionality without an internet connection.
• Provider Flexibility: Supports various AI providers including Ollama, LM Studio, OpenAI, Anthropic, Gemini, and OpenRouter, with easy switching in settings.
• Interactive Voice: Features built-in voice and dictation capabilities using bundled models (Piper for voice, Whisper for dictation) that run in-process, requiring no external keys or servers.
• Spaced Repetition System (SRS): Automatically tracks stumbled words to create a spaced-repetition deck that informs future conversation sessions.
• Local Data Storage: All data is stored in a local SQLite file, with options to export or sync across devices via services like iCloud Drive or Google Drive.
• Multi-language Support: Currently supports eight languages, with the ability to add more by simply adding a folder.
• Cross-Platform Compatibility: Available for macOS, Windows, and Linux.

Download and Installation:

Verba is available for download on macOS, Windows, and Linux. Due to its unsigned nature, users may need to follow specific instructions to open the application on macOS and Windows. Building from source is also an option for users with Node.js and Rust installed.

Build from Source:

To build Verba from source, you will need Node.js and Rust installed. The build process involves running npm install, followed by either 'npm run tauri dev' for development or 'npm run tauri build' to package the application.