17 pages
Documentation
What each setting does, and what it costs you. Written against what the app actually does.
Getting started
Dictation
Press a key, talk, and the text lands at your cursor — plus everything that happens in between.
Permissions
The three permissions Zerm asks for, what each one actually unlocks, and how to fix the one that keeps going red.
Keyboard shortcuts
Every shortcut Zerm offers — global triggers, recorder controls, and the ones that are unset until you choose them.
Audio input
Choosing a microphone, and what to do when the one you want keeps disconnecting.
Dictation
Output modes
Instant, Instant + Refine, and Enhanced — what each one does to the text, and when the difference actually shows.
AI enhancement
Providers, prompts, context, and the timeout settings that decide what happens when a model is slow.
Enhancement shortcuts
Toggle enhancement and switch prompts mid-recording, without leaving the app you are in.
Dictionary
Teach Zerm the names, products and jargon that transcription models reliably get wrong.
Power Mode
Change dictation settings automatically for the app you are in — or the website you are on — without touching your global settings.
Models and speech
Models
Which transcription model to use, what each one costs you in speed, size and memory, and how local and cloud differ.
Custom local models
Import your own fine-tuned Whisper model and use it exactly like a bundled one.
Transcribe File
Drop in an audio or video file and get a full transcript, with each speaker identified, renameable, and ready to export.
Read aloud
Select any text on your Mac and hear it back — with the markup, code and symbols cleaned up first.