VoiceInk and OpenCues are kindred projects: both open source, both local-capable, both priced to be owned rather than rented. VoiceInk is local-first dictation for macOS; OpenCues is inline AI for typed text across terminals, coding agents and Chrome.
Where they overlap
The philosophy. VoiceInk runs Whisper-class models on your machine with optional cloud enhancement; OpenCues runs any of ten providers including fully local Ollama. Neither sits between you and a subscription meter.
Where VoiceInk struggles
Like every dictation tool, its writes are pastes: one-shot insertions with no in-place transform of existing text and no revert beyond app undo. It requires Apple Silicon, and the optional cloud enhancement reintroduces the provider question its local-first pitch otherwise avoids.
How OpenCues differs
Modality and scope. VoiceInk turns speech into text on a Mac; OpenCues works on typed text everywhere it integrates: fill-in-the-blank generation, in-place transforms, cued alternatives and live data. VoiceInk is GPL-3.0, OpenCues is Apache-2.0. Use VoiceInk to speak, OpenCues to compute on what lands.
VoiceInk drops text into any Mac app, which OpenCues integrations do not reach yet: they cover terminals, coding agents and Chrome. OpenCues answers with per-word revertable edits and PII tokenisation on cloud calls.
OpenCues is also extensible by design: cues and blanks are markdown files you write yourself, and blanks can bind shell scripts or runtime classes, all under an open standard anyone can implement. See write your own cue and write your own blank.
Related: Full comparison · Ollama vs cloud · VoiceInk