TL;DR
Get travel and outdoor gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
BetterWispr has introduced a free, Apache 2.0-licensed dictation app for macOS 14 and later. It uses speech models installed on the Mac, offers optional text cleanup and per-app styles, and says it has no account requirement or cloud transcription fallback.
BetterWispr has been introduced as a free, open-source dictation app for Mac, letting users speak into a text field and insert a cleaned-up transcript with a keyboard shortcut. The developer says speech is processed on the Mac using installed models, without an account or cloud transcription fallback; the app requires macOS 14 or later, with Apple Silicon recommended.
The app’s default shortcut is Option + Space: users hold the keys while speaking and release them to finish recording. BetterWispr then transcribes and inserts the text into the app that was active. The developer says transcription starts after recording ends, rather than appearing word by word as the user speaks. Users can change the shortcut or set it to toggle recording.
BetterWispr offers nine built-in speech models from Apple speech, Parakeet and Whisper, according to its website. Language availability depends on the selected model. The developer says Parakeet options cover English, Japanese and 25 European languages, while Whisper supports a broader range. Users can also configure certain external providers or compatible APIs; that differs from the app’s local-model workflow and may involve sending dictation to a selected provider.
Text handling is configurable. The default Light cleanup removes English filler words, repetitions and false starts, while Medium also edits for clarity using a chosen language model. None preserves the spoken wording. Users can set writing styles for different apps, add names and terms, and correct transcripts through History. The app also includes a meeting notetaker, which the developer describes as separating the user’s microphone from audio captured from other apps.
BetterWispr is released under the Apache 2.0 license, and its source code is available on GitHub, according to the project. The developer says updates are delivered through GitHub Releases and Sparkle, with installation dependent on a matching EdDSA signature. The site describes a simulated typing-versus-speaking illustration as illustrative, not a benchmark; it does not provide independent performance measurements.
Local Dictation Without a Subscription
BetterWispr offers Mac users another option for turning speech into text without routing standard dictation through a transcription service. That may appeal to people concerned about sending voice recordings to a cloud provider, as well as users who want offline use after installing a compatible model. Those privacy and offline benefits are tied to the local-model setup described by the developer, not necessarily to optional external integrations.
The project is also notable for combining transcription with cleanup and app-specific writing preferences in a free, open-source package. Those features could reduce editing after dictation, but the developer’s materials do not establish how accurately they work across accents, hardware, languages or apps. The app’s stated requirements and reliance on local model inference mean results and responsiveness may vary between Macs.
As an affiliate, we earn on qualifying purchases.
How BetterWispr Handles Speech
Unlike a live captioning tool, BetterWispr’s described workflow records while the shortcut is held and transcribes after the user releases it. The website says installed local models can work offline after their initial download. It also says model files occupy storage on the Mac, and availability depends on hardware, language and required assets.
Using the app requires microphone permission to record speech and Accessibility permission for automatic text insertion; Apple speech may also require Speech Recognition permission. The developer says that if an app blocks safe insertion, the transcript can be retrieved from the dashboard and copied manually. Secure fields and some applications may prevent automatic pasting.
“Speech stays on your Mac.”
— BetterWispr website
As an affiliate, we earn on qualifying purchases.
Performance and Adoption Still Unknown
The project materials do not provide independent tests of transcription accuracy, latency, battery use or performance across supported Mac hardware. They also do not report downloads, active users or feedback beyond references to launch-thread replies. The model list and language coverage are described by the developer, and users may need to test their chosen model with their own speech and setup.
Some details depend on configuration. The site says external services can be selected for certain workflows, but does not establish that every optional provider handles data in the same way. The developer’s local-processing claim applies to speech processed with installed on-device models; the handling of data by an external provider is subject to that provider’s practices. The release date was not specified in the supplied materials.
As an affiliate, we earn on qualifying purchases.
Users Can Test the Mac Release
The project directs users to its website and GitHub repository for installation instructions, source code and issue reporting. Mac users who meet the macOS requirement can install a model, grant the permissions needed for their workflow and test dictation in their usual apps. They can also choose whether to keep transcript history, which the developer says can be turned off.
Further evidence about reliability, supported configurations and user uptake may emerge through project updates and user reports. Until independent testing or broader usage data is available, claims about speed, accuracy and the practical privacy effects of optional integrations remain matters for users to verify on their own systems.
As an affiliate, we earn on qualifying purchases.
Key Questions
Is BetterWispr free?
Yes. The developer describes it as free and open source under the Apache 2.0 license, with no dictation subscription, account or API key required for the basic local workflow. Model files use storage on the Mac.
Which Macs can run BetterWispr?
The stated requirement is macOS 14 or later. The developer recommends Apple Silicon for local model inference. Performance and model availability depend on the Mac and selected model.
Does dictation work offline?
The developer says dictation can work offline after a compatible local model and its required assets have been installed. Model downloads require an internet connection, and Apple speech availability can depend on language support and installed operating-system assets.
Does BetterWispr send speech to the cloud?
The developer says built-in local models transcribe speech on the Mac and there is no cloud transcription fallback. The app also lists optional external providers and compatible APIs; choosing one may send dictation to that provider, so users should check the selected service’s data practices.
Does it show words while I speak?
No. The project says users hold the shortcut while speaking and release it to finish; transcription begins after recording stops, rather than appearing word by word during speech.
Source: hn
Columbus Day / Indigenous Peoples' Day Picks
long weekend sales
As an affiliate, we earn on qualifying purchases.
