OpenWhispr

What is OpenWhispr?

OpenWhispr is open-source dictation software that converts speech to text across any application accepting typed input. It runs locally on macOS, Windows, Linux, and iOS, or optionally uses cloud transcription. The source code is published under the MIT license.

What are the features of OpenWhispr?

• Local speech-to-text

  Runs on-device models so voice data stays on the user's computer without being sent externally.
• Bring your own API keys

  Supports external providers such as OpenAI, NVIDIA, Cohere, and Oruk alongside local processing.
• Voice commands

  Accepts spoken instructions like "clean this up" or "draft an email" to format or rewrite transcribed text.
• Custom vocabulary learning

  Auto-learns names, jargon, and terms from manual corrections added by the user.
• Multilingual auto-detection

  Supports over 100 languages and can switch languages mid-conversation.
• Offline mode

  Transcribes speech without a Wi-Fi connection when using local models.

What are the use cases of OpenWhispr?

• Traveling without internet access

  Local models allow dictation while offline or away from Wi-Fi.
• Interacting with coding assistants

  Voice input replaces typing into tools like Cursor or Claude Code chats.
• Security-regulated environments

  On-device processing combined with stated HIPAA, ISO 27001, SOC 2 Type II, and GDPR compliance supports restricted workflows.

OpenWhispr Pricing Information:

• Free download

  Available at no cost for macOS, Windows, Linux, and iOS; users supply their own API keys for cloud-based models.
• Enterprise Cloud

  Mentioned as an option; requires contacting sales.