How to set up push-to-talk hotkeys for dictation in Amical
A practical setup tutorial for picking a non-conflicting hotkey, choosing hold versus toggle, configuring per-app modes, and fixing common permission issues.
The hotkey is the single most important setting in any dictation app, and most people pick it wrong the first time. They choose something that conflicts with a Slack shortcut, or something that requires three fingers, or a toggle that leaves the mic open all afternoon. This guide walks through the exact decisions to make when configuring Amical: which key combination to bind, when to use hold mode versus toggle mode, how to set different behaviors per app, and how to fix the two permission errors almost everyone hits within the first hour. By the end you will have a hotkey that disappears into your muscle memory within a day.
Step by step
- 1
Pick a hotkey that no app you actually use has already claimed
Open Amical Preferences and look at the default hotkey. Before you accept it, mentally walk through your top five apps: Slack, your browser, your editor, your calendar, your terminal. Does the default collide with any of them? On macOS, ⌥ space, ⌃ space, and ⇧ ⌘ space are usually safe. Spotlight already owns ⌘ space, so avoid it. On Windows, Alt+Space opens the window menu in many apps, so prefer Ctrl+Shift+Space or a Caps Lock remap. Bind it, then test in Slack and your editor immediately. If anything misbehaves, change the binding before you build muscle memory.
- 2
Default to hold mode, switch to toggle only for long-form writing
Hold mode means the microphone listens only while you hold the key down, and stops the instant you release it. This is what you want ninety percent of the time because it is impossible to accidentally leave it on. Toggle mode is press-once-to-start, press-again-to-stop, which is useful when dictating a long blog post or article where holding a key for two minutes is uncomfortable. Set hold as your global default in Amical, then enable toggle only for specific apps like Notion, Bear, or your blogging tool. Mixing the two modes intentionally beats picking one for everything.
- 3
Configure per-app modes so terminal and code editors behave differently
In Amical's app settings, add per-app rules for any context where you want different behavior. Common setup: disable AI rewrites in terminal apps so commands transcribe literally without 'cleanup'; enable a code-friendly custom vocabulary in Cursor and VS Code; switch to toggle mode in Notion and Obsidian. The per-app rule overrides your global defaults the moment that app becomes frontmost. Test each rule by tabbing between two apps with different settings and watching the menu bar indicator change. If it does not change, the bundle ID match is wrong — re-add the app from the picker.
- 4
Fix macOS Accessibility and Microphone permissions in the right order
On macOS, Amical needs two permissions: Microphone and Accessibility. Microphone lets it record audio; Accessibility lets it type the transcribed text into other apps. Grant both in System Settings under Privacy and Security. If dictation captures audio but does not type anything, Accessibility is missing or stale. Toggle Amical off and on in the Accessibility list, then fully quit and relaunch Amical — macOS frequently caches the old permission state until restart. Microphone permission is rarely the issue once granted; Accessibility is the one to recheck after every macOS update.
- 5
Work around Windows AltGr and keyboard-layout quirks
On Windows, the AltGr key on European layouts acts as Ctrl+Alt, which means a hotkey like Ctrl+Alt+Space will fire every time you type AltGr+Space — disastrous if you write in German, French, or Spanish. Avoid Ctrl+Alt combinations entirely on non-US layouts. Safer choices: Right Shift+Space, a Caps Lock remap via PowerToys, or a single function key like F19 mapped through a keyboard utility. Also check that no gaming overlay (Steam, Discord, NVIDIA) has grabbed the same shortcut globally. Disable overlays one at a time if your hotkey suddenly stops working after a driver update.
Experience the future of dictation
Local and Cloud Models
Choose from powerful cloud-based AI models or run everything locally for maximum privacy and control. Switch seamlessly between providers to find the perfect balance of speed, accuracy, and security.
Custom Vocabulary
Customize the AI to recognize your specific terminology, jargon, and proper nouns for industry-specific accuracy and personalized transcription results.
Custom Shortcuts
Create personalized voice commands and shortcuts to streamline your workflow and boost productivity with hands-free operation.
Multi Language Support
Transcribe and dictate in 100+ languages with native-level accuracy. Switch between languages seamlessly or use mixed-language dictation.