Whispering

Whispering is an open-source desktop application that turns spoken audio into text. It gives you a choice: you can use fast, cheap cloud AI models from providers like Groq or OpenAI, or you can run a local model called Speaches for completely private, offline transcription. The core idea is transparency—you see where your audio goes, you use your own API keys, and you pay the providers directly without any markup from the app itself. It's a native app for macOS, Windows, and Linux, and there's also a web version you can try in your browser.

What makes Whispering stand out is its commitment to user ownership and cost control. Unlike subscription-based services, it doesn't lock you in or add fees. You set up shortcuts to record audio from anywhere on your computer, and the transcribed text appears ready for you to use. The app includes AI-powered text formatting and grammar correction to polish the output. Everything is designed to keep your data private; recordings stay on your device, and when using cloud services, the audio goes straight to the provider.

This tool is a great fit for anyone who regularly needs accurate transcripts but is wary of subscription costs or privacy policies. Journalists transcribing interviews, students recording lectures, and content creators working with audio can all benefit. It works well for quick, on-demand transcriptions, and the ability to choose between free local processing or low-cost cloud options means you can tailor it to your specific needs for speed, accuracy, or budget.