help center
How can we
help you?
Everything you need to set up Dimmy, dictate your first sentence, switch styles, and troubleshoot. Search below or browse by category.
Press/to focus
Browse by category
🚀Getting Started
Install, grant permissions, dictate your first sentence.
Getting Started
Install Dimmy on Mac, Windows, or Linux
Download the installer for your operating system, run it, launch Dimmy. Takes around a minute.
Getting Started
Your first dictation in 60 seconds
Open any app with a text field, press the hotkey, say a sentence, release. The text lands where the cursor was.
Getting Started
Permissions Dimmy needs
Microphone everywhere. Accessibility on macOS to type into other apps. Nothing else, ever.
💊The Pill
The small overlay that turns into a live waveform when you talk.
The Pill
What the pill is
The small always-on-top overlay. It shows whether Dimmy is idle, recording, transcribing, or done.
The Pill
All pill states explained
Idle, recording, transcribing, LLM, done, error. Each one has a unique shape, colour, and animation.
The Pill
Move, resize, hide the pill
Drag the pill anywhere. Pin to a corner. Change opacity when idle. Hide it entirely until the hotkey is pressed.
The Pill
Tray and menu-bar icon
Besides the pill, Dimmy puts a tiny icon in your system tray (Windows, Linux) or menu bar (macOS). Quick access to Settings, History, Quit.
⌨️Hotkey
How to start and stop recording. Defaults, custom combos, gestures.
Hotkey
Default hotkey per platform
Cmd + Option on macOS, Win + Alt on Windows and Linux. Two-key combos chosen to never clash with system shortcuts.
Hotkey
Change your hotkey
Settings → Shortcut. Press the new combo. Dimmy validates it against system shortcuts and saves.
Hotkey
Toggle vs hold-to-record
Hold-to-record: hold the hotkey while talking. Toggle: press once to start, again to stop. Both work, pick what feels right.
Hotkey
Dictionary and Command Mode shortcuts
Two optional shortcuts beside the dictation hotkey: add selected text to your dictionary (Ctrl + Shift + D) and run a one-shot AI command.
Hotkey
Switch style or tone with the scroll wheel
Scroll on the pill to cycle AI styles. Ctrl + scroll to cycle tones. No need to open Settings.
✨AI Styles & Tones
Polish, summarise, translate, or rewrite the transcription on the fly.
AI Styles & Tones
The 13 AI styles
After transcription, an optional LLM pass rewrites your text. From Correct to Imbruttito, with 11 styles in between.
AI Styles & Tones
Command Mode: speak instructions, not text
Select text and say what to do with it. Or, with nothing selected, ask Dimmy to write something at the cursor. The AI does the rest.
AI Styles & Tones
The 5 tones
Neutral, friendly, formal, casual, enthusiastic. The tone slider sits on top of the style and changes register only.
AI Styles & Tones
Write your own system prompt
Pick the Custom style and paste any system prompt. Dimmy uses it for every dictation until you change it.
AI Styles & Tones
Strip filler words automatically
Dimmy can drop "um", "uh", "basically", "cioè" before paste. Works in 6 languages, independent of the LLM pass.
🤖Providers & Models
Local Whisper or cloud STT, your own API key, optional local LLM.
Providers & Models
Local mode: offline by default
Dimmy ships with Whisper running on your device. No API key, no internet, no per-minute cost. Works on a plane.
Providers & Models
Cloud providers compared
Groq is fastest, OpenAI is the gold standard, Deepgram leads on streaming. All optional. Bring your own key.
Providers & Models
Add your API key
Open Settings → Providers & keys, find the provider card, paste the key. It is encrypted on your machine. You can update keys at any time.
Providers & Models
Choose a Whisper model
Smaller is faster but less accurate. base-q8_0 is the default. large-v3-turbo is the top tier on-device.
Providers & Models
Optional on-device LLM
Run the LLM enhancement pass locally via llama.cpp. No cloud, no key, no per-token cost. Pick a model to match your GPU memory.
🛡️History & Privacy
Search past transcriptions. Understand and control what is sent.
History & Privacy
Browse and search your history
Every transcription is saved locally and searchable with full-text. Stats show count, time saved, words dictated.
History & Privacy
Telemetry: what's sent, how to disable
Anonymous, opt-out. Event names and numbers only, never audio, never text, never API keys. Two independent toggles.
🛠️Settings Reference
Every Settings tab explained: General, Models, Shortcut, Output, Overlay, History, Privacy, Audio, Stats, About.
Settings Reference
Settings overview: every tab, in one map
Right-click the pill → Settings. Ten tabs grouped by purpose. Use this as the index to find anything you can configure.
Settings Reference
Audio input: pick a mic and tune preprocessing
Settings → Audio. Select the input device, monitor the level meter, toggle highpass, VAD, and AGC. Defaults are usually right.
Settings Reference
Pick the dictation language
Auto-detect or pin a specific language. 99 languages via Whisper. Affects transcription and filler removal.
Settings Reference
Stats: count, time saved, words dictated
Settings → Stats. Aggregate numbers computed from your local history. Time saved is the difference between speaking and typing the same text.
Settings Reference
About: version, updates, links
Settings → About shows the current version, runs an update check, and links to the repo, license, and acknowledgements.
Settings Reference
Pause Dimmy temporarily
Disable the hotkey without quitting the app. Useful during screen recordings, presentations, or focus blocks.
🎯Use Cases
AI prompts, meeting recaps, email drafting, code comments, multilingual content, accessibility.
Use Cases
Write prompts for ChatGPT, Claude, Cursor, Claude Code
Speak a paragraph of context in 10 seconds, paste into your AI tool. Use the Prompt style to get a clean, ready-to-send prompt.
Use Cases
Record a meeting, get an AI recap
Dimmy Meeting records the call, transcribes it live, and writes a structured recap with action items. All inside Dimmy, on Windows and Mac.
Use Cases
Draft emails in Gmail, Outlook, Apple Mail
Pick a tone, dictate the body in 30 seconds, paste, send. The Professional style handles greetings and sign-offs.
Use Cases
Code comments and PR descriptions
Talk through what you changed and why. Use a Custom prompt that formats the result as a code comment or a PR body.
Use Cases
Multilingual workflows
Switch languages mid-dictation. Translate as you speak. Use auto-detect when you really alternate, pinned when you mostly stay in one.
Use Cases
Accessibility, RSI, hands-busy contexts
If typing hurts or isn't an option, dictation becomes your primary input. Hold-vs-toggle hotkey, hands-free workflow, screen-reader compatible.
🔌Integrations
Plug Dimmy into Claude Desktop, Claude CLI, Notion. MCP setup and recipes.
Integrations
Connect Claude Desktop to Dimmy (MCP bridge)
One-click wizard installs a small MCP server. Ask Claude Desktop about your meetings, dictations, and let it write recaps back into Dimmy.
Integrations
Use your Claude subscription via Claude CLI (no API key)
If you already pay for Claude Pro, Team, or Max, route Dimmy's LLM rewrites and recaps through the local claude CLI. No API key, no per-token cost.
Integrations
Use Dimmy's MCP server from Claude CLI (advanced)
Dimmy's MCP bridge has a one-click setup for Claude Desktop. There's no built-in button for the terminal, but you can point the Claude CLI at the same binary by hand. Here's how.
Integrations
Send meeting recaps to Notion
Dimmy has a built-in Notion integration. Connect once with your own integration token, pick a destination page or database, then send any recap with one button (or auto-send every meeting).
Integrations
Use your ChatGPT subscription via Codex (no API key)
If you already pay for ChatGPT Plus, Pro, Team, Business, or Enterprise, route Dimmy's LLM rewrites and recaps through the local codex CLI. No API key, no per-token cost.
🔧Troubleshooting
Microphone, hotkey, install warnings, transcription quality.
Troubleshooting
Mic or hotkey not working
Check OS permissions first. Then check hotkey conflicts. The pill tells you which one of the two is failing.
Troubleshooting
Empty, wrong, or slow transcription
Three different problems, three checks: silence in the audio, wrong language detected, model or provider mismatch.
Troubleshooting
Install warnings: macOS Gatekeeper, Windows SmartScreen, Linux AppImage
The OS may warn the first time you run Dimmy. These warnings are expected for indie apps. Here's how to bypass them safely.
Still stuck?
Open an issue on GitHub or write to us. Real human, fast reply.