Guides

A free Spokenly alternative, open source with both models inside

By · Published · Updated

YapToText is a free Spokenly alternative for the Mac: it is open source under the GPL, its speech and cleanup models come inside the App Store download, and it collects no analytics. Spokenly is free too, with unlimited local Whisper and Parakeet models, and it does things I do not: it runs on Windows, Linux and iPhone, offers cloud models when you want them, and transcribes files with speaker labels.

I built YapToText, so read this with that in mind. Every Spokenly fact here comes from spokenly.app, read on October 3, 2026.

My hands make typing difficult, so I dictate every day with YapToText, and I wanted the app I depend on to be one anyone can read. If you came for Spokenly vs Wispr Flow, the three-way comparison with Wispr Flow is further down.

Download YapToText on the Mac App StoreDownload for free. No account, nothing locked, macOS 14 or later on Apple silicon. Source on GitHub.

On this page: 7 sections
  1. Side by side
  2. What you keep, and what changes
  3. Spokenly vs Wispr Flow, and where YapToText sits
  4. Move over in about ten minutes
    1. Step 1: Install YapToText
    2. Step 2: Allow the microphone and Accessibility
    3. Step 3: Quit Spokenly, then dictate once
    4. Step 4: Bring your words over
  5. Where Spokenly is better
  6. The short version
  7. Questions

Side by side

Checked October 3, 2026. Prices change, so check before you switch.

YapToTextSpokenly
PriceFree, no paid tierFree with local models or your own API keys; Pro $9.99 a month or $99.99 a year
Where speech is processedOn the MacCloud models by default; on the device once you download a local model
DownloadAbout 2.9 GB on the App Store, both models inside; about 12 MB from Homebrew, without them20 MB for the Mac; local models download separately
AnalyticsNoneAnonymous usage analytics, blocked by Local Only Mode
AccountNoneNone for local models
Open sourceYes, GPL-3.0No; its site says its code is proprietary
PlatformsMac onlyMac, Windows, Linux and iPhone; Android has a waitlist
Mac requirementmacOS 14 or later, Apple siliconmacOS 13.3 or later; Intel Macs with some limits
Default keyRight Command, tap to start and stopRight Command, held while you talk
Mac App Store versionTypes at your cursor once Accessibility is onPastes when you press Command V, and is no longer updated; the website build pastes for you and updates itself
AI cleanup on the free versionPhi-3.5 Mini Instruct, built in, on the MacApple Intelligence, or your own model key
Edits selected text by voiceYes, Quick EditNot stated
Transcribes filesYes, as plain textYes, with speaker labels, timestamps and subtitle export
Languages29 plus auto-detectMore than 100, with automatic detection

The download row is a trade I made on purpose. The two models are most of those 2.9 GB, so nothing is left to fetch after the install. Spokenly starts small and lets you choose a model, and YapToText starts big so there is nothing to choose.

What you keep, and what changes

  • Dictating into any app. Spokenly's Mac page says Right Command is its default key, held while you talk. YapToText also starts on Right Command, tapped once to start and again to stop, or held if you pick Hold to talk. Both fix spacing and capitalization mid-sentence: Spokenly calls it Smart Spacing, and YapToText calls it Intelligent Insert.
  • The App Store version. Spokenly's docs say its Mac App Store version cannot paste on its own, so you press Command V to stop and paste, and its changelog says that version is no longer updated. YapToText's App Store version types at your cursor once you switch on Accessibility.
  • Where your voice goes. Spokenly's own summary says it uses cloud models by default, and your audio stays on the device once you choose and download a local model. The app sends anonymous usage analytics, never audio or transcripts, unless Local Only Mode is on. YapToText runs both models on your Mac, collects no analytics, and goes online only for a model download you start.
  • Cleanup and formatting. Spokenly's AI Instructions strip filler words, fix grammar and format text for the app you are in, using Apple Intelligence or your own model key on its free plans. Its privacy policy adds that a mode with AI Instructions may send your transcript to an AI service even when transcription runs locally. YapToText ships Phi-3.5 Mini Instruct, which runs every AI mode on the Mac with no key.
  • Editing by voice. Spokenly's site does not describe rewriting a selection by voice. In YapToText you select text anywhere, tap Right Option and say the change, and the Quick Edit help page covers the rest.
  • Your names and words. Spokenly has Word Replacements, plus a dictionary for supported cloud models and a prompt for local Whisper. YapToText's Dictionaries shape what the speech model hears as well as what it types.
  • Price. Spokenly's local models and your own API keys are free, and Pro is $9.99 a month or $99.99 a year on its website. YapToText is free, and tips add no features.
App Store poster headed Nothing leaves your Mac, over YapToText's AI Models page with both models marked Included.
Figure 1. Both models are marked Included, so there is no model to choose or download before your first dictation.

You give up other platforms and cloud models, and you get both models inside, no analytics and code you can read.

Spokenly vs Wispr Flow, and where YapToText sits

Spokenly vs Wispr Flow comes down to where your voice is processed and what the free plan allows. Wispr Flow's privacy page says transcription always happens in the cloud, and its free plan allows 2,000 words a week on desktop and 1,000 on mobile. Spokenly's local models are free with no usage limits, and its cloud models come through Pro or your own keys.

Checked October 3, 2026, on spokenly.app and wisprflow.ai.

YapToTextSpokenlyWispr Flow
PriceFreeFree with local models or your own keys; Pro $9.99 a month or $99.99 a yearFree plan; Pro $15 per user a month, or $12 billed yearly
Free version limitNo word limitNo usage limits on local models2,000 words a week on desktop, 1,000 on mobile
Where speech is processedOn the MacIn the cloud by default, or on the device with a local modelAlways in the cloud
PlatformsMacMac, Windows, Linux and iPhoneMac, Windows, iOS and Android
Edits text by voiceYes, Quick EditNot statedYes, Command Mode
Open sourceYes, GPL-3.0No; its site says its code is proprietaryNot stated
App Store poster headed Edit anything by voice: the Quick Edit card listening, applying the edit, and done.
Figure 2. Quick Edit is the YapToText answer to Command Mode: select text, say the change, and the card says Done when it is replaced.

Where YapToText sits: free with no word limit, like Spokenly's local models, but with its cleanup model inside, no cloud option at all, and only on the Mac. The free Wispr Flow alternative page compares Wispr Flow with YapToText in detail, and the Wispr Flow pricing page lists every plan.

Wispr Flow if you want the cloud on every device, Spokenly if you want the choice, and YapToText if you want the Mac to do all of it.

Move over in about ten minutes

Four steps. Spokenly and YapToText both start on Right Command out of the box. Great minds, same key. Quit Spokenly before your first test, so only one app answers that key.

Step 1: Install YapToText

  • You: Get YapToText from the Mac App Store and open it.
  • YapToText: Walks you through a short setup, including whether you want to tap or hold the key. Both models are already inside, so there is nothing else to download.
  • Check: The capybara is sitting in your menu bar.

Step 2: Allow the microphone and Accessibility

  • You: On the Home page, in the Get set up card, press Continue next to Microphone, then Grant next to Accessibility.
  • macOS: Asks once for the microphone, then opens System Settings, Privacy & Security, Accessibility, where you switch YapToText on yourself. An app cannot do that part for you.
  • Check: The switch next to YapToText is on. If it will not stay on, press Show in Finder in the same card and drag the app into the list. Without Accessibility, every dictation goes to the clipboard instead of your cursor.

Step 3: Quit Spokenly, then dictate once

  • You: Quit Spokenly. Click into any text field, tap Right Command, say a sentence and tap it again.
  • YapToText: Shows your voice as a wave and your words as they form, then types the finished text at your cursor.
  • Check: The words land where your cursor was. If the wave stays flat, the app is not hearing your microphone yet.
YapToText's Dictation settings page on a Mac: the Keys, Recording, Microphone and Delivery cards.
Figure 3. The Keys card at the top is where you swap tapping for holding, or move YapToText off Right Command.

To keep both apps, give one of them another key. The Keys card on the Dictation page takes any single key, and the keys and shortcuts help page lists the choices.

Step 4: Bring your words over

Say: Select a name anywhere on screen, tap Right Option, say this, and tap Right Option again.

add this to my dictionary

What happens: YapToText adds the selected spelling to your dictionary, and from then on it spells that word your way.

  • You: Add the rest of your Spokenly Word Replacements on the Dictionaries page, each as what it gets heard as and the spelling you want.
  • YapToText: Uses those words to prime the speech model, so it hears them as well as types them.
  • Check: Dictate a sentence with one of them. It comes out spelled your way.
App Store poster headed Teach it your words, showing the Dictionaries page and the Commands page side by side.
Figure 4. Dictionaries do the job Word Replacements did in Spokenly, and they also teach the speech model the words.

The Dictionaries help page covers folders, case and whole-word matching.

Where Spokenly is better

  • More places to dictate. Spokenly runs on the Mac, Windows, Linux and iPhone, and keeps a waitlist for Android. YapToText runs only on the Mac.
  • Cloud models when you want them. Pro adds managed cloud transcription with models such as GPT-4o Transcribe, Deepgram Nova 3 and Soniox, or you can bring your own API keys and Spokenly adds no fee. YapToText has no cloud option at all.
  • More languages. More than 100, with automatic detection, against YapToText's 29 plus auto-detect.
  • Files and meetings. Spokenly transcribes files with speaker labels, timestamps and subtitle export, and on a Mac it records your microphone and the Mac's system audio without a meeting bot. YapToText transcribes files as plain text, with no speaker labels or timestamps, and has no meeting recorder.
  • Older Macs. Spokenly needs macOS 13.3 and runs on Intel Macs with some limits. YapToText needs macOS 14 and Apple silicon.
  • Automation. Agentic Actions turn your voice into web searches, app launches, Shortcuts and shell commands, and there is an MCP server for coding agents and a command line tool. YapToText has nothing like that.

If you need any of those, Spokenly is the better choice, and it is good software. If your dictation has to follow you off the Mac, Spokenly goes where YapToText cannot.

The short version

  • Who should switch. Anyone whose dictation lives on one Apple silicon Mac and who wants both models inside, no analytics and source code to read.
  • Who should stay. Anyone who needs Windows, Linux or iPhone, cloud models, meeting recording or more than 29 languages.
  • What it costs. YapToText is free. Spokenly is free with local models, and Pro is $9.99 a month or $99.99 a year, checked October 3, 2026.
  • How long it takes. About ten minutes: install, Accessibility, one test dictation, then your words.

Both are free on a Mac, so the real question is whether you want to pick models or just talk. If something here no longer matches Spokenly's site, tell me on GitHub. Every issue gets read.

Download YapToText on the Mac App StoreDownload for free. No account, nothing locked, macOS 14 or later on Apple silicon. Source on GitHub.

About the author. Ryleigh Newman is a developer whose hands make typing difficult and who dictates every day with YapToText, the app this site describes. Facts about other apps come from their own sites, with the date they were read. More about the project.

Sources

Questions

Is there a free alternative to Spokenly?

Yes. YapToText is free and open source, with its speech and cleanup models inside the App Store download and no account. Spokenly's own local models are free as well. YapToText differs by shipping its own cleanup model, editing a selection by voice and collecting no analytics.

Is Spokenly free?

Yes, with local models or your own API keys. Spokenly's site says its on-device Whisper and Parakeet models are free forever with no word caps and no account. Pro, which adds managed cloud models and hosted AI cleanup, is $9.99 a month or $99.99 a year on its website, checked October 3, 2026.

Is Spokenly open source?

No. Spokenly's own open source page says it is not open source and its code is proprietary, though it is free to use with local models. YapToText is open source under GPL-3.0, with its source on GitHub.

Does Spokenly work offline?

Yes, with a local model. Spokenly's site says its Whisper and Parakeet models run on the device, and Local Only Mode blocks network access except local model downloads. Its own summary says it uses cloud models by default, so download a local model first.

Spokenly vs Wispr Flow: which is better?

Spokenly, for privacy and price: its local models are free with no usage limits, while Wispr Flow's privacy page says transcription always happens in the cloud and its free plan allows 2,000 words a week on desktop. Wispr Flow is the one with an Android app, and its Pro plan is $15 a month or $12 billed yearly. Checked October 3, 2026.

VoiceInk vs Spokenly: what is the difference?

How you pay and where it runs. VoiceInk is open source under GPL v3 and sells its ready-built app for a one-time $25 for one Mac after a 7-day trial, on Apple silicon Macs only, per its pricing page. Spokenly is free with local models, charges $9.99 a month for Pro, and also runs on Windows, Linux and iPhone.