About OpenASR Desktop
OpenASR Desktop is a Windows application for private, offline speech recognition and dictation. It provides system-wide dictation, allowing spoken words to be entered into applications, and can transcribe audio or meetings. Recognition runs entirely on the device using the open-source OpenASR engine, which is based on ggml and supports CPU and GPU inference. The software offers eleven model families. Models are downloaded once from a signed catalog and can then be used offline. The product description states that nothing is uploaded, there is no cloud fallback or telemetry, and no account is required. This makes it suitable for users who want speech-to-text without sending audio to a cloud service. OpenASR Desktop combines dictation and transcription in a local workflow. The evidence does not enumerate languages, accuracy levels, or hardware requirements. Performance depends on the chosen model and available CPU or GPU resources.
Categories
Platforms
Website & Downloads
Trusted package or store listing
Related software
- 5ire — 5ire is an AI assistant and MCP client designed to connect language models with tools and data sources.
- AI Browser Toolkit — AI Browser Toolkit is a JSON-over-HTTP server that lets an AI agent control a real Chrome or Edge browser window.
- AI Context Builder — AI Context Builder converts C# and .NET solutions into compact Markdown context for coding agents.
- AI Limits — AI Limits is a small Windows 11 overlay that displays usage limits for supported AI providers.
- AI Usage Dashboard — AI Usage Dashboard is a local Windows application for viewing usage windows and balances across supported AI services.
- AI Viewer — AI Viewer opens and previews Adobe Illustrator (.ai) graphic files and can convert them to other image formats such as JPEG, bitmap, GIF, and PNG.
