Yes, there are open-source apps on GitHub that do the core job of Wispr Flow on a Mac: hold a key, speak, and have the text appear where your cursor is. Handy, VoiceInk, FluidVoice, OpenWhispr, VocaMac, Hold to Talk, Dictly and Voquill all publish their source code, and all of them can run a speech model on your own machine instead of in a cloud service. They differ in licence, in which Macs they support, and in how much they try to do beyond plain dictation.
Publisher note: SuprFlow publishes this guide and is one of the products discussed. Competitor details come from each company's own pages, checked on 1 October 2026. Prices and features change; confirm before you buy.
SuprFlow is proprietary. Its source code is not public, so it is not an open-source alternative and is not in the table. It is mentioned once near the end, for readers who decide they want a packaged, supported app instead.
Which open-source apps do the same job as Wispr Flow?
Licence, platforms and model options are as stated in each repository on 1 October 2026. Release dates are from each project's GitHub releases page, in UTC.
| Project | Licence stated in the repo | Platforms | Local and cloud models | Latest release |
|---|---|---|---|---|
| Handy | MIT | macOS, Windows, Linux | Local only: Whisper (Small, Medium, Turbo, Large) and Parakeet V3. The README says the process is "entirely local". | v0.9.8, 3 October 2026 |
| VoiceInk | GPL v3 | macOS 15.0 or later, per the README | Local models. The VoiceInk site describes an optional Cloud Enhancement that sends transcribed text, not voice. | v2.21, 28 September 2026 |
| FluidVoice | GPL v3 | macOS 15.0 or later; a Windows beta is offered on its download page | Local Parakeet, Nemotron, Cohere Transcribe, Apple Speech and Whisper. Optional AI enhancement through OpenAI, Groq or custom providers. | v1.6.9, 18 August 2026 |
| OpenWhispr | MIT | macOS (Apple Silicon and Intel), Windows, Linux | Local Whisper and Parakeet models, or cloud providers with your own keys. Also meeting transcription and notes. | v1.10.2, 15 September 2026 |
| VocaMac | AGPL-3.0 | macOS 14 or later, Apple Silicon only | Local Whisper, Parakeet and Apple Speech. An optional Custom Endpoint uploads recordings only when you select it. | v1.0.0, 23 September 2026 |
| Hold to Talk | Apache 2.0 | macOS 15 or later, Apple Silicon only | Local Parakeet by default. Optional cloud transcription or cleanup with your own API key. | v2.3.0, 21 August 2026 |
| Dictly | MIT | macOS 15 or later, Apple Silicon only | Local Whisper. The README says "No cloud API. No account. No telemetry." | v1.2.1, 11 July 2026 |
| Voquill | AGPLv3 "unless otherwise noted", per the README | macOS, Windows, Linux | Local Whisper, or a cloud provider you choose. AI text cleanup. | desktop v0.0.652, 26 September 2026 |
All eight had a release within the three months before this page was checked, so none looks abandoned.
What should you know about each one?
Handy is the most permissive and the most portable: MIT licence, three operating systems, and a README that puts "free" and "private" first. It describes itself as a simple transcription tool. Choose it if you also use Linux or Windows. SuprFlow vs Handy
VoiceInk is open source and also sold. The code is GPL v3 and the README invites you to build it yourself. The compiled app on the developer's site costs $25 once for one Mac as of 1 October 2026 and, according to the README, adds automatic updates and priority support. Note that the README lists macOS 15.0 or later while the site lists macOS 14.4 or later on Apple Silicon, so check the current download. Make sure you are on the Beingpax repository, since app names are easy to confuse. SuprFlow vs VoiceInk
FluidVoice is GPL v3 with a wide choice of local speech models. Its README is open about one exception: Fluid Intelligence, its on-device enhancement model, is "a separate, privately maintained local AI runtime" and is not part of the open code. Its download page also offers a newer macOS beta than the release listed above. SuprFlow vs FluidVoice
OpenWhispr is the broadest in scope. Besides dictation it lists an AI agent, meeting transcription, notes and team spaces, and it has a hosted cloud with paid plans alongside the free local mode. That makes it closer to Wispr Flow plus its Notetaker than the other projects are. SuprFlow vs OpenWhispr
VocaMac offers several speech engines and ships a small Whisper model in the app so you can dictate before downloading anything. It uses the AGPL, which matters if you plan to modify and redistribute it. SuprFlow vs VocaMac
Hold to Talk is local by default and documents deployment for managed Macs, including MDM-managed preferences. SuprFlow vs Hold to Talk
Dictly is the smallest: a menu-bar app, one job, signed and notarised by Apple according to its README. SuprFlow vs Dictly
Voquill is cross-platform, with AI cleanup and a personal glossary. Read its licence file before reuse, because the README says the AGPLv3 applies "unless otherwise noted". SuprFlow vs Voquill
What does Wispr Flow have that these projects do not replace?
Wispr Flow is a commercial cloud service, and its own pages list things a GitHub project does not usually come with:
- Phone apps on the same account. Wispr lists iOS and Android alongside Mac and Windows on its pricing page.
- Audit reports and agreements. Wispr publishes a SOC 2 Type II report, an ISO 27001 certificate and a HIPAA Business Associate Agreement, described in its security overview. None of the eight READMEs above mentions an equivalent audit report or BAA.
- Central administration. SSO, SCIM and audit logs are on Wispr's Growth and Enterprise plans.
- Nothing to download or maintain. Wispr's cloud model is updated for everyone at once. With a local app, you pick the model and keep it current.
What the open-source projects have that Wispr does not: you can read the code, most cost nothing, and speech can be recognised on your Mac without an internet connection. Wispr's Enterprise Privacy page states that it does not offer on-premise or offline deployment today. If that is your reason for looking, see Does Wispr Flow work offline?.
How do you check an open-source dictation app before installing it?
- Open the LICENSE file, not just the README badge. A public repository with no licence is not open source.
- Open the Releases page. Note the date and whether the download is a signed, notarised build or something you must build yourself.
- Read the README section on cloud features. Find every option that sends audio or text off the Mac, and whether it is off by default.
- Check the system requirements. Five of the eight projects above need macOS 15 or Apple Silicon or both.
- Download only from the repository or the site it links to. Similar names are common.
- Test with made-up text first. Dictate "Send 1,250 units to Dr. Nakamura on the 9th, not the 19th" into the app you use most and check the name and numbers.
Open source lets you inspect an app. It does not prove that the build you downloaded matches the code or that every optional feature stays local. Does open-source dictation prove privacy? goes through what source access can and cannot show.
If you want a packaged, supported app instead
Building from source and choosing models is not for everyone. If you decide to pay for a ready-made Mac app with local speech recognition, VoiceInk's compiled build is the open-source route at $25. SuprFlow is a proprietary option at $125 once: it runs a WhisperKit model on Apple Silicon Macs with macOS 14 or later and has a personal dictionary and voice snippets. You cannot inspect its source, it has no free tier, and it is Mac only. SuprFlow vs open-source Mac dictation compares that trade directly.
Questions people ask
Is there an open-source version of Wispr Flow?
Not from Wispr. Wispr Flow is a commercial service whose transcription runs in Wispr's cloud, and Wispr says it does not offer an on-premise or offline version. The projects above are independent apps that do a similar job.
Which open-source Wispr Flow alternative works on Windows or Linux?
Handy, OpenWhispr and Voquill list macOS, Windows and Linux. FluidVoice offers a Windows beta.
Is VoiceInk really open source if it costs money?
Yes. The source is published under GPL v3 and you may build it yourself. The price is for the developer's compiled build with updates and support.
Is an open-source dictation app more private than Wispr Flow?
It can be, if you use a local model and leave cloud options off, because audio then stays on your Mac. Check the specific build and settings instead of relying on the licence.
Is SuprFlow open source?
No. SuprFlow is proprietary and its source is not public.
Related guides
- SuprFlow vs open-source Mac dictation
- Does open-source dictation prove privacy?
- Free Wispr Flow alternatives for Mac
- Does Wispr Flow work offline?
- Wispr Flow alternatives for Mac
- Offline voice to text on Mac
Sources
- Handy — GitHub repository
- releases
- VoiceInk — GitHub repository
- releases
- official site
- FluidVoice — GitHub repository
- releases
- download page
- OpenWhispr — GitHub repository
- releases
- pricing
- VocaMac — GitHub repository
- releases
- Hold to Talk — GitHub repository
- releases
- Dictly — GitHub repository
- releases
- Voquill — GitHub repository
- releases
- Wispr Flow — Pricing
- Wispr Flow Help Center — Security and privacy overview
- Wispr Flow — Enterprise Privacy