The honest Willow Voice alternative

What this comparison is, and who built it

The boring truth is that Willow is a clean tool. It ships on three platforms, including a phone app, and its whole pitch is press a hotkey, speak, and get formatted text. That auto-formatting is genuinely nice. If you are already on it and happy, you can probably close this tab. I mean that. There is a whole section near the end that tells you exactly when to stay put.

For everyone still reading, the gap is the free tier and where your audio goes. That is the argument, and I will spend the rest of the article showing it instead of asserting it.

No fake review counts, no invented user numbers, no logos of teams that supposedly love us. Just two feature lists and a table you can check against both homepages.

What Willow Voice actually does

Willow runs on Mac, iPhone, and Windows, and it is free to download on all three. So if you read somewhere that it is a Mac-only app, that is out of date. It ships on Windows too. The core loop is a hotkey. You press it, you speak naturally, and formatted text appears at your cursor in whatever app you are in. It works across Slack, Gmail, and the rest. The auto-formatting is the part people like most, and it earns the praise.

It claims to work in any language, though it does not put a number on it, so I will not invent one. Its security page mentions SOC 2, HIPAA, zero data retention, and a privacy mode. One thing Willow does not publish anywhere I could find is whether transcription runs on your machine or on a server, and it does not name the engine it uses. I am not going to guess. I will only say that "where does my audio go" is a fair question to ask before you pick any dictation tool.

The pricing is where the comparison sharpens. Willow's free plan gives you 2,000 words per week, with no credit card required. Paid plans are an Individual tier billed annually with unlimited words, a Team tier with a minimum of three seats, and a custom Enterprise tier. The free tier asking for no card is a fair, friendly move. The catch is the weekly word cap.

Two thousand words is roughly a long email plus a couple of Slack threads, and then you wait for the week to roll over. I am keeping our own dollar figures out of the body on purpose — read our pricing page and do the math against your own week.

What you also get with Whisper by Remskill

Here is the part where I describe the thing I built, then let you judge the table. Whisper by Remskill is two products on one hotkey. The free tier is the whole local pipeline, and it has no weekly word cap. You get the 8 Whisper transcription models, the Parakeet engine, fully-offline AI cleanup through Ollama, transcription history, presets, hotwords, hardware acceleration, model downloads, and a custom hotkey. No payment method at signup. You make an account, download the app, press the hotkey, and talk for as long as you like.

The part I can say plainly, because it is verifiable, is that all local transcription runs on your machine. You can dictate on a plane in airplane mode and nothing leaves the laptop. You pick your local engine based on what you need, not what we push. Whisper gives you 99 languages, translate-to-English, custom vocabulary, beam-size control, and hotword biasing, at the cost of speed. Parakeet is the NVIDIA TDT engine, about 600 MB, and it runs 5 to 10 times faster than Whisper on a CPU, covering English plus 24 other European languages. Pick Parakeet for speed and English. Pick Whisper for languages, translation, or fine control. Neither is the default. That is your call.

Local accuracy typically lands between 95% and 99%, and it all runs on your CPU with no GPU required. The app itself is about 25 MB on disk, which is smaller than most of the screenshots I have taken of it. The Pro tier adds the Cloud surface, and it is bring-your-own-key. You paste your own OpenAI key, you pick the transcription model, and the AI enhancement runs on gpt-5-mini by default. There is also web search at your cursor through OpenAI's Responses API, which pastes a synthesised answer in a few seconds. We take no cut on top of OpenAI's rate. Your key, your bill.

The default hotkey is Ctrl+Space on Windows, remappable, with a push-to-talk chord on Mac, and one account covers up to 3 devices. If you want the full setup walkthrough, the Windows voice-to-text guide covers it step by step.

The platform story is the honest part. Windows and macOS on Apple Silicon both ship today. There is no iPhone app, no iPad app, no Android. If you dictate from your phone, this is where Willow is plainly ahead.

Willow Voice vs Whisper by Remskill, side by side

The table nobody else seems to fill with real rows. No dollar figures in it — check both pricing pages for those.

A few honest reads of this table. Willow ships a real iPhone app and clean auto-formatting, and its free tier asks for no card at all. Those are real wins. Everywhere the row is about a word cap, about whether your audio stays on your machine, and about bring-your-own-key cloud, the gap lands on the other side.

Free tier vs free tier, where the real difference lives

This is what most people came here to compare, so let me be concrete. Willow's free tier asks for no credit card and gives you 2,000 words a week. That is a genuinely generous on-ramp. You can try the whole thing without handing over a card, which I wish more apps did. The catch is the meter. Two thousand words is a long email plus a couple of Slack threads. If dictation actually replaces typing for you, you will brush that ceiling by Wednesday and then you are waiting for the week to tick over or reaching for the paid tier.

Whisper by Remskill has no such meter. The entire local pipeline is the free tier, with no expiry and no word cap. Whisper transcription, Parakeet transcription, offline Ollama AI rewrites, history, presets, hotwords, model downloads, all of it, for as many words as your microphone can produce. You can dictate for eight hours on a flight in airplane mode and the only ceiling is your laptop battery. The one thing the free tier leaves out is the Cloud surface, and that is the part you upgrade for.

Here is the one opinion I will spend in this article. Your dictation audio should stay on your machine when it can. The email to your kid's school, the legal note, the half-formed idea you blurt at your cursor, none of that needs a server in the loop to become a paragraph. Whisper's local pipeline keeps every word on your laptop, and you can verify that by pulling the plug and watching it still work. Willow does not publish whether it processes on your device or in the cloud, so I cannot tell you either way. I can only tell you that "where does my audio go" is a fair question to ask before you commit, and that ours has a checkable answer. For the upgrade math on our side, read the pricing page and do the subtraction yourself.

Most "AI dictation" apps are a transcription model, a tidy UI, and a recurring invoice. We give the local layer away and charge only for the optional Cloud layer, because your laptop already has a microphone and a CPU and does not need a meter bolted onto it to type one paragraph.

When to stay on Willow Voice

This section earns the rest of the article. There are real reasons to stay, and I am not going to pretend otherwise.

Everyone else should start with our free tier and see whether they ever hit a wall. On the local pipeline, there isn't one.

If you only remember one thing

The thing voice typing fixes is the gap between having something to say and getting it into the document. Willow closes that gap on three platforms with clean formatting, and meters the free tier at 2,000 words a week. We close it on two platforms, give the whole local pipeline away with no cap, and keep your audio verifiably on your machine. Pick the one whose free tier matches how you actually work. If you are not sure, the uncapped one costs nothing to find out.

If you are weighing other dictation tools too, the honest superwhisper alternative walks through the same comparison from a different angle.