AskOpenSource by The Arthi AI Collective
openwhispr / openwhispr

openwhispr

OpenWhispr is a free, open-source, privacy-first alternative to WisprFlow and Granola: press a hotkey, speak, and your words appear at your cursor, with fully offline transcription or your choice of cloud AI models, plus meeting transcription with speaker diarization.

View openwhispr/openwhispr on GitHub
GitHub stars
11,505
Forks
—
Language
JavaScript

Project details

Repository
openwhispr/openwhispr
Licence
MIT
Primary language
JavaScript
Catalog status
Available
Catalog check
2026-11-13

Repository description

Voice-to-text dictation app with local (Nvidia Parakeet/Whisper) and cloud models (BYOK). Privacy-first and available cross-platform.

Documented setup

  1. Check the desktop release and local-model requirements

    Read the linked README and platform guide for your OS and computer. Confirm the local transcription mode and hardware requirements; this catalog does not establish a RAM or CPU-only guarantee.

  2. Follow the packaged desktop setup

    Use the project's documented desktop release and setup. Download the speech model while online, then keep the documented local-processing mode. Cloud models and sync are outside this offline route.

  3. Transcribe one non-sensitive sample

    Use the documented transcription workflow on a short sample. Compare the words with the recording and confirm local processing before trying private material. Check the documentation separately if caption-file export is needed.

Requirements and limits

  • Scope: local voice transcription and dictation using a downloaded local speech model. Select local processing; optional cloud models and cloud sync are outside this offline path.
  • Free use means no mandatory software or service fees for this documented local mode. It does not establish zero total cost: compatible hardware, storage, electricity, internet access, and model downloads depend on your setup.
  • Install the packaged desktop release and download the model while online before using local transcription offline. Internet, storage, and existing hardware are still required for setup.
  • No required paid AI API for local transcription. Hardware fit, RAM, transcription accuracy, caption export formats, and performance have not been checked here.

Sources

Explore by topic

More tools in Voice & audio