OpenWhispr

OpenWhispr is an open‑source dictation application that converts spoken language into typed text across any program that accepts keyboard input. Running on macOS, Windows, Linux and iOS, it can operate entirely offline or route audio to a cloud transcription service if you choose that option.

When you activate the hotkey, the program captures your microphone stream and feeds it either to a locally installed model, such as Whisper Base, or to an API you provide. The recognized words are displayed in the active text field in real time, with plosive filler words and false starts removed. You can then give voice commands to refine the text: saying “clean this up” or “draft an email to Mike” triggers the built‑in AI cleanup routine, which rewrites the captured sentence, inserts punctuation, and formats it for the target context. Corrections you make are fed back into the local learning system so the model learns domain‑specific terms and proper names automatically.

The tool supports over 100 languages and auto‑detects which one you're speaking. Switching languages in the middle of a sentence is smooth; as long as the input channel remains open, OpenWhispr will continue to transcribe in the new language. Because all processing can stay on‑device, it works even without Wi‑Fi, making it suitable for confidential settings where external connectivity is limited. For those who need higher throughput or domain‑specific accuracy, the cloud mode can be enabled; recordings are deleted immediately after transcription, and data is never retained for training.

OpenWhispr is tailored for professionals who take meeting notes, code comments, or chat messages and want to use speech instead of typing. It is especially valuable in collaborative environments like Slack or Google Docs, where typed output must appear instantly. However, it may not satisfy users who rely on real‑time translation into complex, context‑heavy documents without a local device, or who require integration with proprietary transcription engines not listed on the site. The listing does not provide pricing information; all features are offered for free, but advanced server‑side options may incur costs depending on the chosen provider.

Key features

• Offline on‑device speech‑to‑text in 100+ languages
• Voice‑command driven editing and formatting
• Auto‑learning of new terminology from user corrections
• Zero data retention by default, with optional cloud mode that deletes recordings after use