OpenLess

OpenLess is an open-source desktop app that converts spoken language into polished text. It functions as a system-wide voice input tool for macOS and Windows, allowing users to dictate text directly into any active text field, such as email drafts, chat boxes, or AI interfaces.

The application operates via a global hotkey. When the key is held, OpenLess records audio, transcribes it, and uses an AI model to refine the text based on a selected style before inserting it at the cursor. If the target application blocks direct insertion, the software copies the resulting text to the clipboard to prevent data loss. This workflow allows users to capture thoughts quickly and have them formatted automatically without manual editing.

Key features

• AI-prompt mode that structures loose speech into detailed prompts for LLMs
• Streaming insertion that writes polished text character by character to the cursor
• Style Pack Marketplace for installing and sharing community-created system prompts
• Custom style pack creation to define specific tones for emails or reports
• Bring-your-own-cloud credentials for Volcengine ASR and DeepSeek-compatible completions
• Support for both toggle and push-to-talk recording modes
• Integration with MediaPlayPause triggers for wired earbud controls

Built with a Rust backend using Tauri and a React frontend, OpenLess is designed for users who want to compose complex prompts or long messages without typing. It differs from standard dictation tools by focusing on text cleanup and restructuring rather than word-for-word transcription. For example, a rambling spoken request can be transformed into a structured list with specific constraints. The software does not maintain conversation context or answer questions, as its sole purpose is to transform speech into usable written text.

OpenLess is a local-first, transparent tool, ensuring that user data and dictionaries stay on the local machine.