VoiceInk
A Mac dictation app with local models, free to build from source and paid if you want the ready-made download.
Instead of Superwhisper, Wispr Flow

Screenshot from github.com/beingpax/VoiceInk, taken 2026-10-30, shown to help you recognise the tool. It belongs to its maker. Ask us to remove it.
VoiceInk is the one in this group where the money question is out in the open. The source is GPL-3.0 and anyone can build it, and the author sells the ready-made download, with a free trial first, saying purchases pay for full-time work on the project. It runs on macOS 14.4 or later and says its local models process everything offline.
VoiceInk scores a 4 on lock-in, self-direction and delivery and a 3 on footprint, power and longevity.
It is Mac only, and a paid download for a program you could compile yourself is a fair arrangement that still puts a price on convenience. Its page does not say what the Context Aware and AI Assistant features send over the network, so we have not checked them. One developer built it over five months before opening the source.
- Mac dictation with local models, and a GPL-3.0 source anyone can build
- The ready-made download is paid, after a free trial
- Mac only, from one developer
What stands out
Do you make VoiceInk? Get the badge for your site
This badge reports a dated review. It is not a certification and it costs nothing. Link it to this page and leave the numbers alone: the image always shows the current score.
For dark backgrounds, use voiceink-dark.svg in place of voiceink-light.svg.
For comparison, we also reviewed Otter.ai (14/30) as a benchmark.