40+ honest answers to the questions people actually ask.

CaringDictate is voice typing software for Windows. You press a hotkey, speak, and the words appear in whatever you were typing — email, Word, Facebook, web forms, and most other places you type. It is designed for elderly users and adults with medical conditions that make typing hard. $49 one-time, no subscription, free updates included.

About three minutes. After you install the software, a setup wizard asks three questions (which hotkey, whether to use the same key to stop, and whether you want a soft chime when dictation starts and stops). You can change any answer later. The setup is deliberately short.

No. CaringDictate uses general speech-recognition models that are the same for everyone, so there is no enrolment session to sit through before you start. The trade-off is that there is also no way to train it on your voice — tools built around a personal profile, such as Dragon, do let you do that, and if teaching the software your own vocabulary matters to you, that is a real reason to choose one of those instead. You can add individual words to a personal list here.

Usually about ten minutes from download. The trial needs no sign-up and no card: download the installer, run it, answer the three setup questions — which control starts dictation, whether the same one stops it, and whether you want a chime — and you are dictating. Speaking your writing takes a little getting used to, so give yourself a few days of real use before judging it.

The default talk button is the Right Ctrl key — tap it once to start, tap again to stop, or hold it down and let go when you finish, like a walkie-talkie. You can change it during setup, or in the settings later, to any keyboard key or mouse button. If pressing a combination is awkward, a single large key such as Caps Lock, or a mouse button, is usually easier to hit reliably.

Yes. $49 once. Free updates included. No subscription, no upgrade tiers, no monthly fee, no recurring anything. We deliberately rejected the subscription model because most of our users are managing chronic conditions that already cost too much.

We may raise the launch price as the product matures, but anyone who has already purchased will keep their license at the price they paid, with all future updates included.

They are aimed at different buyers. Dragon Professional is a professional tool with a professional feature set — custom voice commands and macros, trainable vocabulary, audio-file transcription — and Nuance do not publish a price on their product page, so check with them or a reseller. CaringDictate is a consumer tool at $49 once: typing by voice, plus recording and transcribing your own meetings. The $49 is simply what we charge, not a discount off anything.

Yes, you can install CaringDictate on up to 3 of your own computers (e.g. desktop, laptop, second laptop). It is one license for one person's use across their devices, not a household or corporate license.

The base price is already set as low as we can sustainably make it. If $49 represents real financial hardship, email hello@caringdictate.com — we have a small fund for users in genuine need and will provide free or reduced-cost licenses.

We are not able to advise on that, and we would rather say so than guess at something that could cost you. CaringDictate is consumer software at $49 — not a medical device, and not on any pre-approved assistive-technology vendor list. Whether a particular plan treats it as an eligible expense is a question for your plan administrator, and worth asking before you buy rather than after.

No. All voice processing happens on your own computer. A speech model sized to your machine is downloaded once during setup and runs locally. While you dictate, audio is processed in real time on your machine and then discarded. Your voice and your dictated text never leave your computer — they are never uploaded to any server, and nobody, including us, can hear or read what you dictate. The one time audio is saved to a file is when you switch on meeting recording, and even then the recording, the transcript and the notes all stay on your computer.

CaringDictate processes speech entirely on your own computer and never transmits your voice or text, so there is no cloud vendor handling your data. But CaringDictate is consumer software: it does not sign Business Associate Agreements, and using it does not by itself make you HIPAA compliant — that depends on your overall practices, not on any single app. What we can say plainly is that because nothing you dictate leaves your machine, there is no third-party upload to worry about.

Only for the initial install (downloading the software and the recognition model). After installation, CaringDictate runs fully offline. You can use it on planes, in areas with poor internet, or on computers that are not connected to the internet.

Two small things, both harmless and both documented in our privacy policy. It checks for software updates periodically. And from version 2.1.0, unless you switch it off in Settings → Privacy, it sends anonymous usage statistics: plain numbers such as the app version, your Windows version and hardware class, which voice model you use and how fast it runs, and whether the app crashed. These help us keep the app working well on every kind of computer. They are keyed to a random ID that identifies nothing and no one, and they never include your audio, your text, your name, or your email.

In your user folder under AppData. CaringDictate looks at your computer and picks the voice model that will run best on it — models range from about 190 MB on modest computers to about 3.1 GB for the largest, optional model on powerful ones. It is downloaded once during setup and never re-downloaded unless you switch to a different model.

Windows 10 (64-bit) and Windows 11. Both Home and Pro editions. Some features may not work on Windows in S Mode — switch to standard Windows mode if you encounter issues.

Windows 10 and 11 only. That is the whole of what exists today, and we would rather not promise anything we have not shipped — if you want a Mac version, there is a list you can join on the system requirements page and we will let you know if one ever arrives. On a Mac today, Apple Dictation is built in and free, and SuperWhisper and VoiceInk are both well-regarded on-device options. On iPhone and Android, the built-in dictation is genuinely good and we would not try to better it.

Typically between about 1 GB and 3 GB total, depending on which voice model CaringDictate picks for your computer — roughly 450 MB for the application plus 190 MB to 3.1 GB for the recognition model. A machine using the largest model, with an update in progress, can briefly need up to about 6 GB. RAM usage is approximately 1-2 GB while actively dictating.

Any microphone that Windows recognizes will work — built-in laptop microphones, USB headsets, Bluetooth earbuds, conference room mics, all of them. We strongly recommend a USB headset microphone for best accuracy, especially in rooms with background noise.

We do not publish an accuracy number, because a single figure without stated test conditions does not tell you anything useful about your own situation. CaringDictate uses open speech-recognition models from the Whisper family, among others, and how well they understand any particular person depends on that voice, the microphone, the room and the model chosen. Specialised vocabulary may need terms adding to your personal dictionary. The free seven-day trial is the only test that answers the question for you.

Yes. You can dictate in 20 languages: English, Chinese, Spanish, French, German, Portuguese, Italian, Dutch, Polish, Swedish, Turkish, Russian, Japanese, Korean, Arabic, Hindi, Thai, Vietnamese, Indonesian, and Czech — or leave it on Automatic and CaringDictate figures out which language you're speaking. The app's menus themselves are currently in English.

Honestly, for a lot of people, nothing that matters — and if the free tools already do what you need, keep using them. The differences are narrow: CaringDictate listens from when you press your key until you press it again, its recognition runs on your own computer, it keeps a personal word list, it works on Windows 10 as well as 11, and you choose the key, mouse button or spoken phrase that starts it. Windows voice typing and Voice Access are free, made by Microsoft and already installed, so try those first.

Yes. Voice activation lets you set a wake phrase — "start dictating," or any phrase you choose — that begins dictation completely hands-free. Turn it on in Settings under Voice activation. It works offline like everything else.

No-questions-asked. Email hello@caringdictate.com within 30 days of purchase with your order number, and we will refund the $49 to your original payment method the same day. There is no form to fill out, no exit survey, no explanation required.

Email hello@caringdictate.com and we will help. The refund window does not start counting against you while we are troubleshooting. Many "I need a refund" emails turn out to be "I need help setting up my microphone" once we look at the specifics.

Not as a standard policy, but if you reach out and explain the situation, we will look at the specifics. For users who hit an unexpected medical issue that prevented them from using the software during the trial window, we generally accommodate.

Yes — arthritis is one of the conditions we built CaringDictate for. The software replaces keyboard typing with voice typing, which eliminates the joint impact of every keystroke. See our dedicated pages on rheumatoid arthritis and osteoarthritis for specifics.

Yes — see our dedicated condition pages for each. CaringDictate is built for the typing gap these conditions create, while speech is still functional. For late-stage ALS or significant speech impairment, more specialized tools (eye-gaze AAC, Voiceitt) may be more appropriate.

CaringDictate uses general-purpose speech-recognition models. How well they understand any particular person depends on that person, their microphone and their room, and we have not tested them against any specific speech difference. A good USB headset microphone helps significantly. The 30-day refund means you can find out whether CaringDictate works for your specific voice without risk.

For stroke survivors with hand weakness but functional speech, yes — voice typing is one of the most commonly recommended assistive technologies for this case. For stroke-related speech impairment (dysarthria, aphasia), the answer depends on severity; mild cases work well with CaringDictate, severe cases may need specialized AAC.

For adults with CP and mild-to-moderate speech variation, CaringDictate often works well, and it is a one-time $49 price rather than a subscription. For severe dysarthria, Voiceitt is purpose-built for impaired speech and is the better tool.

Yes. CaringDictate types into Word, Word Online, OpenOffice, LibreOffice, WordPad, and Google Docs in your browser. It also works in every other text-input area on Windows.

Yes — both. Outlook desktop, Outlook Web App, Gmail in Chrome/Edge/Firefox, Yahoo Mail, AOL, Apple Mail via web. CaringDictate types into the focused text field regardless of which email client you use.

Yes — for the chat fields. CaringDictate types into Zoom chat, Teams chat, Slack messages, Discord, and any other messaging app. It can also record and transcribe the calls themselves on your own computer; see the meeting recording questions below.

Yes, in every browser (Chrome, Edge, Firefox, Brave, Opera, Vivaldi). CaringDictate types into web search bars, form fields, social media post boxes, banking sites, and any other web text field. No browser extension is required.

Yes, CaringDictate is compatible with major screen readers. The text it inserts is announced by the screen reader the same as text you typed by hand.

Yes. Once you switch it on, it notices when a call starts in any program — Zoom, Teams, WhatsApp, WeChat, Discord, a call in your browser — records it, and transcribes the recording afterwards on your own computer. You can also press Record now and capture anything manually. It is included in the $49; there is no per-minute charge because there is no server.

No. It is off when you install CaringDictate, and nothing is recorded until you have read and accepted a one-time note confirming that having consent to record is your responsibility. A visible indicator shows on your screen while a recording runs, and there is no hidden mode.

Yes, once two small speaker-detection files (about 34 MB) have downloaded. Transcripts then read as Speaker 1, Speaker 2 and so on, with a breakdown of how long each person talked. The labels are positional — the app does not know who anyone is and keeps no voice profile between recordings.

Yes, if you turn the notes switch on. A small language model downloads once and then writes a summary and action items into each recording after it is transcribed. It runs on your own computer, so it takes a few minutes on a slower machine.

Yes — drag a WAV, MP3 or M4A file onto the Meetings page. You get the same transcript, the same optional speaker labels, and the same export choices: Word, PDF, plain text, Markdown, or .srt / .vtt subtitles. Files up to five hours are supported.

Each recording gets its own folder, in your CaringDictate data folder by default or anywhere you choose, holding the compressed audio, the transcript and the notes. Keep them forever, or have recordings older than 30, 90 or 365 days removed automatically, or set a disk budget of 5 to 50 GB. Anything you pin is never deleted.

Your $49 buys CaringDictate outright. The version you buy is yours to keep on up to 3 of your PCs, with every 3.x feature update included and at least 24 months of updates guaranteed — plus security fixes beyond that. If we ever release a paid major upgrade, it is optional and discounted for existing customers.

CaringDictate checks for updates periodically and notifies you when one is available. You can install the update with a click, or postpone it. We never auto-install updates without your permission.

We test against new Windows releases and ship compatibility updates as needed. If you ever hit a Windows compatibility issue, email hello@caringdictate.com and we will help.

Email help@caringdictate.com. A real human reads every email and usually replies within 24 hours (often much faster). We do not have phone support (it would compromise our ability to help everyone fairly) but the email turnaround is fast.

Not yet — we are a small operation and a forum is more work to maintain well than to do badly. If you have specific questions, email us; if you have feature requests, also email us. We track everything by hand.

Absolutely. We assume many of our users have family members involved in setup and troubleshooting. There is no special process — just email and let us know who you are helping.