Practical overview
When it fits
Use it for hands-free text entry in macOS apps, or to transcribe meetings and produce summaries and action items. Choose a built-in model when you want speech processed on the Mac; choose an external endpoint only when you intend to send audio to that service.
Who it is for
- Mac users who want to dictate text directly into other applications.
- People who need meeting transcripts, summaries, and action items.
- Users who want to personalize recognition of names or specialized terms.
Documented uses
- Speak text into the active app using a keyboard shortcut.
- Transcribe meetings and create titles, summaries, and action items.
- Improve recognition of names and jargon with vocabulary hints, phrase replacements, and corrections.
- Choose between on-device models and a configured external transcription endpoint.
Requirements & limits
Requires macOS 14 or later; Apple Silicon is recommended. Setup requires Microphone and Accessibility permissions. The project is open source under Apache-2.0.
- Apple on-device recognition needs no model download; Parakeet and Whisper download once before offline use.
- Apple recognition may not support every language or system speech-asset configuration.
- A configured external endpoint receives audio, and its pricing and retention policies apply; the app does not automatically fall back to cloud transcription.
- History can be turned off, but retained settings, vocabulary, and history are stored in a workspace file the project identifies as unencrypted.
- Recordings stop after two minutes, and large models may load slowly the first time.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up local voice dictation
Install Betterwispr, choose a built-in speech model, grant permissions, and dictate into an app.
Before you start
- macOS 14 or later
- Apple Silicon recommended
- Microphone and Accessibility permissions
Steps
Download the DMG from the project’s GitHub Releases and drag Betterwispr to Applications.
Open the app and follow its welcome guide.
Choose a model in Models. Apple on-device speech needs no download; Parakeet and Whisper download once and can then work offline.
Grant Microphone and Accessibility access.
Click a text field, hold ⌥ Space, speak, and release.
Expected outcome
The recognized text is pasted into the active app; if safe pasting is unavailable, it is copied to the clipboard.
Guide limitations
- Recordings stop automatically after two minutes.
- Apple on-device speech may be unavailable for some languages or system speech assets.
- Large models may take time to load on the first transcription.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 13 October 2026
- Original website title
- opennookorg/betterwispr · Official homepage title · retained in its source language