01
Reading is fast. Writing it out isn't.
A screen reader gets you through a page quickly. Composing a reply runs the other direction—every word entered, then confirmed by ear. For a long email or message, that adds up.
All use cases
Composing it word by word is slow. Speak it instead.
Your screen reader handles output. Entering a long message is still the slow part—word by word, confirming each one by ear. Voicetypr lets you speak the text into any field, then review and edit it with VoiceOver, NVDA, or JAWS the way you already do.
works with your screen readerlocal transcriptionpastes into any field
Last updated
The friction is real, and specific
02
Most dictation ships your audio to a server.
You dictate personal messages, sometimes with people nearby. Cloud tools stream that audio off your machine. Voicetypr transcribes on-device by default, so your voice stays with you.
03
Voice input only helps if you can check it yourself.
Speech recognition still misses things—names, homophones, the odd dropped word. That's fine when the text lands in a normal field your screen reader can read back and let you fix. It's not fine when a tool traps the text in its own window.
The shape of the day, different
spoken
Text entry at speaking speed
Hold the hotkey, say the sentence, release. A paragraph goes in at the speed you talk instead of word by word—then you review it with your reader.
Outcome · input
local
Audio stays on your machine
Transcription runs on-device by default. Private messages and personal details aren't streamed to a transcription server.
Outcome · privacy
any
Pastes into any text field
Email, chat, search boxes, docs, forms. The text arrives as if typed, so whatever screen reader you use reads it straight back.
Outcome · reach
Three workflows where it shows up
- 01
Compose a long message, then review it your way
Navigate to the field with your screen reader as usual, hold the hotkey, say the message, release. It pastes in. Arrow back through it with VoiceOver or NVDA, fix anything the recognizer missed, send.
- 02
Forms, search bars, and address fields
Short strings are tedious to enter and confirm one character at a time. Speak the query or the field value instead; your screen reader announces the result so you know it landed.
- 03
Notes, chat, and docs where the words pile up
Slack replies, quick notes, a paragraph in a doc—say them instead of entering them by hand. Push-to-talk sits on Option/Alt+Space, one key you can hit without hunting for a button.
The honest Blind & low vision FAQ
Pulled from real conversations with people who use Voicetypr for this exact reason.
Does Voicetypr replace my screen reader?+
No. It only handles input—getting words into a field. Your screen reader still does navigation and reads everything back, including the text you just dictated. They run side by side.
Can I review and correct what it transcribed?+
Yes. The text pastes into the field as if you typed it, so your screen reader reads it back and you edit it the usual way. Speech recognition isn't perfect; reviewing with your reader is part of the workflow.
Does it work with VoiceOver, NVDA, and JAWS?+
Yes. Voicetypr pastes into any standard text field, so whatever reads that field reads the dictated text—VoiceOver on macOS, NVDA or JAWS on Windows 10+.
Is my voice sent to the cloud?+
No. Transcription runs on your machine with a local model by default, so your audio stays on your device. Optional AI formatting sends text only when you turn it on—never the audio.
Can I start it without finding a button on screen?+
Yes. It's a global hotkey—⌘+Shift+Space on macOS, Ctrl+Shift+Space on Windows—plus push-to-talk on Option/Alt+Space. You trigger it from the keyboard, anywhere, without locating an on-screen control.
Not answered here? Email support
Voicetypr is productivity software, not medical software. It can help reduce typing load, but it does not diagnose, treat, or prevent any condition. Follow medical and ergonomic guidance for your own situation.
Speak the words. Review them your way.
Speak text into any field, then check and fix it with the screen reader you already use—audio stays on your machine by default. Try Voicetypr free for 3 days.