← All comparisons

VoiceInk alternative

An opinionated local dictation app or a configurable voice toolkit?

VoiceInk is an open-source macOS dictation toolkit with local and optional cloud models, per-app modes, context sources, assistant features, and custom commands. IraVoice is the opinionated alternative: one shortcut, fewer moving parts, Mac, Windows, and Linux, and code names spelled from your project in developer tools.

Source-reviewed

Side by side

IraVoice and VoiceInk

A factual comparison of the source-reviewed product boundaries. Follow the source links below for the latest competitor details.

FactorIraVoiceVoiceInk
Product shapeOne dictation shortcut (hold or tap), custom words, replacements and snippets, and code names from your projectConfigurable modes, local and cloud models, context sources, assistant output, and custom commands
ProcessingOn-device NVIDIA Parakeet by default; Cloud recognition and AI polish planned for IraVoice Cloud, coming soon, with no user-configured providersLocal models by default; optional user-configured cloud transcription and enhancement providers
Source and licensingCommercial app built on the MIT-licensed Handy; IraVoice’s own code is proprietary; one-time license, with a Cloud plan coming soonGPLv3 source can be built manually; the precompiled app uses a one-time commercial license
Platform supportmacOS 13+ on Apple silicon, Windows 10 or 11 (64-bit), and Linux (beta)VoiceInk’s website lists macOS 14.4 or later; its current docs say macOS 15 or later
Developer automationSpells code names from the Git project in terminals, editors, and coding agents; types at the cursor and runs no commandsModes can auto-send text or pass the final transcript to a local custom command

Decision guide

Choose for your actual constraint.

VoiceInk is the better fit when…

  • You want to inspect, build, or modify a GPLv3 codebase.
  • You want to choose among local models, cloud providers, prompts, and context sources.
  • Per-app modes or custom shell-command automation are central to your workflow.

IraVoice is the better fit when…

  • You want one opinionated shortcut and a sensible default model instead of configuring a model catalog.
  • You also work on Windows or Linux and want the same dictation habit there.
  • You dictate into terminals and coding agents and want code names spelled the way your project spells them.

Primary sources

Check the claims at the source.

Competitor details were reviewed July 26, 2026. Pricing, platform support, and policies can change.

  1. VoiceInk source repository ↗

    Current macOS requirement, GPLv3 source, commercial build model, and high-level feature list.

  2. VoiceInk documentation ↗

    Current local and cloud model choices, modes, context, shortcuts, assistant behavior, and output options.

  3. VoiceInk terms ↗

    Dual licensing, one-time commercial license, local defaults, and optional cloud-provider boundaries.

  4. VoiceInk mode settings ↗

    Current per-app triggers, context, auto-send, output, and custom-command behavior.

Questions

About IraVoice and VoiceInk

Is VoiceInk free and open source?

VoiceInk publishes its source under GPLv3, so experienced users can build it. Its ready-to-use compiled app is offered under a separate one-time commercial license with updates and support.

Is VoiceInk always offline?

No. VoiceInk supports local models, but it also supports optional cloud transcription and enhancement providers that users configure with their own keys. The active mode determines the data path.

Is IraVoice more private than VoiceInk?

This page does not claim that. Both can use local speech processing. VoiceInk’s privacy boundary depends on its selected model, provider, context, and mode; IraVoice Local keeps audio and text on the computer, with its model download and license check-ins disclosed, and IraVoice Cloud’s connected path (coming soon) is named separately.

macOS 13+ on Apple silicon, Windows 10 or 11 (64-bit), and Linux (beta)

Prefer fewer settings and a narrower promise?

Try IraVoice’s opinionated dictation workflow for 3 days with no email, card, or account.