Your words. Typed as you speak.
Click a text field, hold your shortcut, and speak. Voice Typing Pro types your words directly into the field while you talk. On-device speech for Apple Silicon Macs, with a 164 MB English model.
English model download · Phonon-2
Smaller download than our Whisper option
Speech recognition on your Mac
Voice typing that stays on your Mac
Live typing example
Spoken words appear directly in the focused text field while you talk.
Built for your Mac
Speak naturally. Keep writing.
A smaller English download
Phonon-2 needs a 164 MB compressed model download. Leave more room for everything else.
Words as you speak
Your words go straight into the focused text field while you talk, with corrections as recognition settles.
Your shortcut
Choose Globe/Fn, Right Option, or a custom key combination in Settings.
Hands-free when you need it
Double-tap your shortcut to start. Double-tap again to finish.
More languages, by choice
Switch to the optional Whisper Turbo model for multilingual transcription.
Local speech processing
Audio is processed on your Mac. Speech recognition works offline after model setup.
Less to download
English doesn’t need a multi-gigabyte model
Model downloads, in decimal MB, rounded. Phonon-2 is a compressed archive; installed model size and total app size are different. Reference model sizes: Fermion Research’s model card. This compares download size, not recognition accuracy or speed.
Use cases
For the words you write every day
Emails & messages
Say the reply while it’s fresh in your mind.
Notes & ideas
Get the thought down before it disappears.
First drafts
Speak the rough version. Edit it into your own.
FAQ
Your questions, answered
Model sizes, local processing, shortcuts, and compatibility.
Words appear directly in the focused text field while you speak. You do not have to finish dictating to see the text. Recognition may refine words as you continue; release your shortcut when you are done.
For Apple Silicon
Your words. On your Mac.
Voice Typing Pro brings voice typing to Apple Silicon Macs. A small English model, local speech recognition, and words that appear directly in your text field as you speak.