Dictate notes, project updates, task briefs, and page drafts into the Notion field that has focus. StarWhisper uses a global Windows hotkey and local Whisper transcription, so you can speak first and edit the text before sharing it.
Four steps turn a spoken thought into an editable page or task description.
Download StarWhisper, install it on Windows, and allow microphone access. Start with a local Whisper model. The free tier includes 2,000 words per week.
Right-click the StarWhisper circle, open Settings, choose the Basic tab, and use the Hotkey section. Pick a combination that is easy to hold while Notion is active and does not collide with another shortcut.
Open the page or task entry and click exactly where the words should appear. StarWhisper types into the active field, so check the cursor before recording. This works for a page draft, update, or note.
Hold the hotkey, speak in complete sentences, and release it when finished. Read the inserted text, fix names or dates, and format it before you move the note or share the page.
Capture the thinking quickly, then give the page its final shape.
Speak an idea, decision, or research note while it is fresh. Leave the wording natural, then edit it into a clear page after the transcript appears.
After a meeting, dictate your own summary, decisions, owners, and next steps. Review names and dates before you share the recap with anyone else.
Dictate the goal, context, acceptance checks, and questions for a task. The complete brief gives the next person more than a short title.
Use voice for the first draft of a plan, guide, or project update. Pause between sections and let the editable text become the page outline.
Notion pages often begin as unfinished thoughts. You may know what happened, what needs to change, or what the team should decide, but typing the first version can slow the capture. Dictation lets you record the context in ordinary language while the details are still easy to recall.
Once the transcript is in the focused Notion field, you can shorten it, add headings, move it to the right page, and make it ready for other people. StarWhisper handles the input; you still decide what becomes a shared record.
StarWhisper uses the OpenAI Whisper model family through local whisper.cpp processing. In Local Mode, your speech is transcribed on your computer. The audio does not need to leave the device before text appears in the active Notion field.
Pro users can choose an optional cloud Whisper fallback. For a local-first setup, read the offline speech-to-text guide before you decide which mode fits your notes.
StarWhisper supports models from tiny through large-v3 and 100+ languages. A quick task note and a long multilingual page can use different model choices. Choose a language in the app or use its available language options.
CUDA, Vulkan, and Metal acceleration paths are available according to hardware and operating system. CPU-only processing is also supported. If a transcript takes longer than you want, adjust the model or use an available acceleration path before changing your note-taking workflow.
Use one global hotkey across notes, prompts, and coding fields.
Dictate longer prompts and notes before you send them.
Draft channel updates and replies, then review them.
See the product overview, platform details, and features.
Review the paid plan for unlimited dictation and larger models.
Learn about local Whisper processing on Windows.