The short answer. The local-first group is the interesting one: OpenQuack, TypeWhisper, EnviousWispr and Tambourine transcribe on the machine with Whisper-family models, and the promise is that the audio never leaves it. FnKey and OpenTypeless are push-to-talk and paste at the cursor, which is the interaction people actually want — hold a key, speak, keep typing. Spokenly adds a hundred languages and an offline mode; Voxt adds per-app and per-URL rules plus translation; Ottex formats the text for the app you are in; Willow Voice edits and matches style; Whispering and VoxFlow expose multiple providers; VoiceInk is the real-time one. The dividing question is whether the model runs locally, because that single choice decides latency, privacy and what happens when the network is down.

This guide covers 13 voice input apps for macOS. Every fact below is drawn from the same verified records that back each app’s own page. The recommendations are our editorial judgment, not payment.

Quick comparison

App What it is Pricing Licence Version
EnviousWispr productivity app Free not published —
FnKey productivity app Open source GPL-3.0 0.5.0
OpenQuack productivity app Open source MIT 2.0.0-alpha.22
OpenTypeless productivity app Open source MIT 1.1.61
Ottex productivity app not published not published —
Spokenly productivity app not published not published 2.30.6
Tambourine productivity app Open source AGPL-3.0 —
TypeWhisper productivity app Open source GPL-3.0 1.7.0
VoiceInk productivity app Open source not published 2.21
VoxFlow productivity app Open source GPL-3.0 1.18.1
Voxt productivity app Open source not published 1.15.0
Whispering productivity app Open source not published 7.11.0
Willow Voice productivity app not published not published 2.5.1

What actually decides this

  • Local model or cloud service. A local model is slower to install, faster at the margin, and private by construction. A cloud service is better at hard audio and sends your microphone to someone else. Every other feature is downstream of this choice.
  • What triggers the recording. Push-to-talk, a toggle and always-listening are three different products. Push-to-talk is the only one that cannot accidentally record a meeting, which is why most of these implement it.
  • What happens to the text after transcription. Raw transcription and formatted text are different outputs. Filler removal, punctuation and per-app tone are the features that make dictation usable in an email and wrong in a code editor.
  • Which languages it handles. A model that is excellent in English can be unusable in another language, and the record usually states the coverage. If you dictate in more than one language, this is the first field to check.
  • How it inserts the text. Pasting at the cursor works everywhere and breaks in some secure fields. Typing the characters is slower and more compatible. The insertion method explains most of the “it works in Notes but not in Slack” reports.
  • What it keeps. A dictation tool with a history is a transcript of everything you have said into it. A tool that discards after insertion is a different privacy posture, and the setting is usually buried.

Our picks

  • OpenQuack — Its record describes privacy-first dictation that transcribes locally and inserts text at the cursor with a hotkey, which is the local push-to-talk case. Choose it when you want local transcription with a hotkey and no account.
  • TypeWhisper — Its record describes local Whisper-based speech-to-text with a global hotkey, which is the simplest local case. Choose it when you want a local Whisper model behind a global hotkey.
  • FnKey — Its record describes push-to-talk voice input that pastes transcribed text instantly, which is the latency-first case. Choose it when you want hold-to-talk that pastes instantly.
  • Spokenly — Its record describes voice-to-text with more than a hundred languages, an offline mode and AI formatting, which is the multilingual case. Choose it when you dictate in more than one language.
  • Voxt — Its record describes hold-to-talk input with transcription rules per app and URL plus built-in translation, which is the per-context case. Choose it when the text should be formatted differently per app.
  • VoxFlow — Its record describes a voice input workspace with local and cloud ASR, OCR, history and coding-agent workflows, which is the workspace case. Choose it when you want a workspace around dictation rather than a single shortcut.

The apps

EnviousWispr

EnviousWispr is a productivity app for macOS. Sub-second dictation running Whisper/Parakeet locally with privacy-first AI text polish. It is free.

What it does. Free, private AI dictation for macOS. Speak naturally and get polished text in the apps you use, with on-device transcription and offline AI polish.

Full profile and verified facts for EnviousWispr →

FnKey

FnKey is a productivity app for macOS. Push-to-talk voice input tool that pastes transcribed text instantly. It is released under GPL-3.0.

What it does. Real-time streaming - Audio streams to Deepgram as you speak (no waiting). Deepgram Nova-3 - Latest model with smart formatting and punctuation. Groq fallback - Whisper large-v3 batch mode if Deepgram unavailable. Audio enhancement - DC offset removal, high-pass filter, peak normalization (Groq mode).

FnKey is listed with current version 0.5.0.

Full profile and verified facts for FnKey →

OpenQuack

OpenQuack is a productivity app for macOS. Privacy-first voice dictation tool that transcribes locally with WhisperKit and inserts text at the cursor with a hotkey. It is released under MIT.

OpenQuack is listed with current version 2.0.0-alpha.22.

Full profile and verified facts for OpenQuack →

OpenTypeless

OpenTypeless is a productivity app for macOS. Open-source AI voice input that types polished text into any app. It is released under MIT.

What it does. Stripe checkout for new purchases with existing Creem subscriptions kept valid. More reliable cloud sign-in and billing with secure session storage, clearer service-status messages, and stronger account-state recovery. Modern model support for GPT-5/o-series requests and the native Anthropic Claude Messages API, while Claude through OpenRouter remains supported. More STT choices and safer recording limits including regional Qwen3 ASR Realtime, provider-aware duration controls, and preserved audio while providers connect.

OpenTypeless is listed with current version 1.1.61.

Full profile and verified facts for OpenTypeless →

Ottex

Ottex is a productivity app for macOS. AI dictation tool that formats text for the app or site you are using. Its pricing is not published.

What it does. Turn rough speech and meeting conversations into the emails, tickets, CRM updates, and docs that move work forward. Free to start — local models or your own.

Full profile and verified facts for Ottex →

Spokenly

Spokenly is a productivity app for macOS. Voice-to-text with 100+ languages, offline mode, and AI-powered formatting. Its pricing is not published.

What it does. Free dictation app for Mac, iPhone, Windows, and Linux. Speak in any app and get accurate text from local Whisper or modern cloud models. 100+ languages, offline.

Spokenly is listed with current version 2.30.6.

Full profile and verified facts for Spokenly →

Tambourine

Tambourine is a productivity app for macOS. Open-source, customizable AI voice dictation that works in any app. It is released under AGPL-3.0.

What it does. An open-source platform for AI voice dictation that you can customize and personalize to your exact needs.

Full profile and verified facts for Tambourine →

TypeWhisper

TypeWhisper is a productivity app for macOS. Local Whisper-based speech-to-text with a global hotkey. It is released under GPL-3.0.

What it does. Private speech-to-text for macOS, Windows, and iOS: system-wide dictation, file transcription, AI workflows, and local engines. Free and open source.

TypeWhisper is listed with current version 1.7.0.

Full profile and verified facts for TypeWhisper →

VoiceInk

VoiceInk is a productivity app for macOS. Real-time speech-to-text app. It is open source.

What it does. Dictate anywhere on your Mac with fast, private voice-to-text. VoiceInk runs locally, works across apps, and is available with a one-time purchase—no subscription.

VoiceInk is listed with current version 2.21.

Full profile and verified facts for VoiceInk →

VoxFlow

VoxFlow is a productivity app for macOS. Open-source voice input workspace with local and cloud ASR, OCR, history, and coding-agent workflows. It is released under GPL-3.0.

What it does. Global dictation: Hold to speak and release to insert text in any editable field. Local asset workbench: Dictation, screenshots, screen recordings, clipboard items, and notes become searchable assets. Multiple ASR providers: Apple Speech works out of the box; local and cloud providers are available when configured. Personal corrections: Local deterministic rules and optional conservative LLM correction help stabilize names, terms, and technical words.

VoxFlow is listed with current version 1.18.1.

Full profile and verified facts for VoxFlow →

Voxt

Voxt is a productivity app for macOS. Hold-to-talk voice input tool with AI transcription rules per app and URL, and built-in translation. It is open source.

What it does. We’ve moved to SGLang Omni Voxt: https://github.com/sgl-project/sglang-omni/tree/main/Voxt.

Voxt is listed with current version 1.15.0.

Full profile and verified facts for Voxt →

Whispering

Whispering is a productivity app for macOS. Multi-provider speech-to-text with AI transformations and keyboard shortcuts. It is open source.

What it does. Truly open source transcription you control. Local or cloud, always transparent. No middleman markup.

Whispering is listed with current version 7.11.0.

Full profile and verified facts for Whispering →

Willow Voice

Willow Voice is a productivity app for macOS. AI dictation with automatic editing, style-matching, and noise optimization. Its pricing is not published.

Willow Voice is listed with current version 2.5.1.

Full profile and verified facts for Willow Voice →

How to choose

Open source: FnKey, OpenQuack, OpenTypeless, Tambourine, TypeWhisper, VoiceInk, VoxFlow, Voxt and Whispering (9 of 13).

Common mistakes

  • Judging a local model on its first run, before the model has finished downloading.
  • Dictating into a password field and wondering why the text did not appear.
  • Leaving a history enabled on a machine other people use.
  • Choosing a cloud service for privacy-sensitive work because its accuracy demo was better.

How this guide was put together

Each app’s facts come from its own record: the developer’s own pages and repository, the App Store listing where one exists, and the documentation, each with a source grade and an access date. Where a fact rests on the developer’s own statement rather than on something we verified independently, the sentence says so. Where our data has no value for something, the guide says “not published” instead of guessing.

A price appears here only where the record carries one from a stated source; the guide never estimates one, and it records the pricing model alongside so the two can be read together. Prices change and vary by region, so each app links to its own page for the current detail.

The facts and the recommendations are kept apart on purpose. Everything in the app entries, the comparison table and the answers above is composed from the records. The short answer, the criteria and the picks are our editorial judgment: they reflect what we would choose and why, and they are not paid for. Where we make a recommendation we name the condition it applies under, because we do not call anything “the best” without explaining the criteria.

Frequently Asked Questions

How do I choose between these voice input apps?

There is no single answer — it depends on what you need. This guide covers 13 voice input apps: EnviousWispr, FnKey, OpenQuack, OpenTypeless, Ottex, Spokenly, Tambourine, TypeWhisper, VoiceInk, VoxFlow, Voxt, Whispering and Willow Voice. They differ mainly in pricing model, licence and the trade-offs each one carries. Pick by the trade-off you care about rather than by rank.

Are these voice input apps free?

Pricing across this group: free (EnviousWispr), open source (FnKey, OpenQuack, OpenTypeless, Tambourine, TypeWhisper, VoiceInk, VoxFlow, Voxt and Whispering) and not published (Ottex, Spokenly and Willow Voice). Where we list pricing as not published, the developer does not state a model we could verify — check their own site before assuming anything about cost. We quote a price only where the record carries one from a stated source, and we never estimate one. Prices change and vary by region and by store, so always check the developer’s own page for the current figure.

Which of these voice input apps are open source?

FnKey (GPL-3.0), OpenQuack (MIT), OpenTypeless (MIT), Tambourine (AGPL-3.0), TypeWhisper (GPL-3.0), VoiceInk (not published), VoxFlow (GPL-3.0), Voxt (not published) and Whispering (not published) are open source.

Do these voice input apps support Apple Silicon?

Our data does not carry an explicit Apple Silicon statement for any app in this group, so we do not claim one. The absence of a statement is not the same as the absence of support.

How do I choose between them?

Start with how you pay: the guide groups these apps by pricing model, because the budget is the constraint most people actually have. Then check the trade-offs listed for each app: the limitations are the part that decides most people’s choice, because they describe what the app will not do for you.