Confluence voice typing on Mac

Confluence has speech-to-text for Rovo creation. Ordinary editor text is a different path.

Confluence now lists speech-to-text inside Create with Rovo, where a connected AI prompt becomes a draft or canvas for review. That is not generic voice typing across every page, live doc, comment, or title field. On Mac, system Dictation can supply ordinary text to a supported focused field, Rovo Desktop can dictate a prompt, and Loom can preserve a recording with a transcript. Each path creates a different artifact and carries a different publication boundary.

Source-reviewed

1. Choose the Confluence artifact before the microphone

Create with Rovo can begin a page, live doc, blog post, whiteboard, database, Smart Link, or slide from a prompt and connected context. Confluence currently includes speech-to-text among the helpers for that creation workflow. The result is an AI-assisted draft or canvas, not text inserted at an arbitrary cursor everywhere in Confluence.

A focused Confluence editor field is the direct path for ordinary prose. Rovo Desktop dictation creates or edits a prompt inside the desktop assistant. Loom begins a recording that can later expose a transcript. Choose among draft generation, field text, an AI prompt, and a retained recording before granting microphone access.

  • Use Create with Rovo for a generated Confluence starting point.
  • Use focused-field Dictation for text in a page, live doc, title, or comment.
  • Use Loom when video or audio should remain the primary artifact.

2. Use Create with Rovo speech-to-text for AI-assisted creation

Select Create, choose Create with Rovo, then describe the intended content by speech or text. Add relevant Confluence links, attached files, or a template when they should constrain the result. Rovo produces a preview you can continue editing before choosing Add to Confluence.

Because Rovo can use connected organizational sources and create several Confluence content types, inspect the selected destination, links, files, generated claims, and content type. Speech recognition is only the first step; Add to Confluence remains the visible commitment.

3. Dictate ordinary text into the focused Confluence editor

Open the intended Confluence space and content item, enter edit mode when required, then click the exact title, body, table cell, or comment field. Start the macOS Dictation shortcut shown in System Settings → Keyboard, speak, stop, and confirm the insertion point did not move to search, navigation, or a Rovo prompt.

Confluence supports collaborative editing and automatically saves work in progress. That protects against some local loss but does not verify names, dates, links, mentions, formatting, confidential details, or the audience. Use the slash command or keyboard shortcuts after dictation when structured content is needed.

4. Respect the difference between pages and live docs

Pages keep drafts, publish an initial version, and can later expose unpublished changes before an update. Confluence version history can compare or restore previous published versions. Live docs have no draft state: edits become visible to permitted collaborators as they happen.

That difference changes the review boundary for speech. In a page, review the full draft before Command + Enter publishes or updates it; Command + Shift + Enter can publish with a version note. In a live doc, assume recognized text is already visible and correct mistakes immediately.

5. Keep Rovo Desktop dictation inside the prompt box

Rovo Desktop can request microphone permission and use dictation to create or edit prompts. It can also use optional screen context. Those controls belong to the desktop assistant conversation; they are not documentation of a generic microphone that types into every Confluence editor field.

Rovo Desktop requires an Atlassian account with Rovo access. Before speaking, check whether screen context is active, which app content is visible, and whether the prompt may use connected organizational knowledge. Review the response before copying, applying, or publishing anything.

6. Treat Loom recording and transcription as a separate workflow

From a Confluence page, live doc, blog, whiteboard, or comment, Loom can record video and audio directly into the collaboration flow. The recording is stored in the user’s Loom library, and Loom can provide a transcript after processing.

A Loom transcript does not become ordinary editor text automatically and the recording has its own sharing model. Visibility can depend on both Confluence content restrictions and Loom permissions, so verify both systems before recording or sharing sensitive material.

7. Keep permissions, AI availability, and publication explicit

Confluence editing and viewing permissions still govern who can change or read content. Atlassian currently says AI is automatically enabled for Standard, Premium, and Enterprise plans, while organization administrators can control Rovo access. Rovo is not available in Atlassian Government organizations or Confluence Cloud sandbox sites.

System Dictation supplying text does not bypass Confluence permissions or choose the correct space. Create with Rovo does not make generated claims true. Loom transcription does not unify recording and page permissions. Confirm the organization, space, content restrictions, audience, and final Publish, Update, Add, or Comment action separately.

8. Troubleshoot the feature that actually failed

If Create with Rovo or its speech helper is unavailable, verify the site plan, organization access, administrator controls, account, supported environment, and the current Create menu. If Rovo Desktop cannot hear you, inspect its microphone permission rather than the browser’s editor-field behavior.

If ordinary Mac Dictation produces no text, test one sentence in TextEdit, confirm the Dictation shortcut, language, microphone, Voice Control setting, connection requirements, and the focused Confluence field. If Loom recording or transcript creation fails, check Loom permission, recording status, processing, and library access separately.

9. Add IraVoice only for a distinct local text path

Use Create with Rovo when connected AI should help create a Confluence artifact, Rovo Desktop for an assistant prompt, Loom for a retained recording, and Confluence’s normal controls for publication and permissions.

IraVoice 0.7.1 is a separate option for macOS 26+ Apple-silicon Macs when recognition and supported formatting should run on the Mac across supported focused text fields, or when speech should first become reviewable Markdown. IraVoice does not sign in to Confluence, read spaces, choose pages, use Rovo, access connected sources, record Loom video, set restrictions, insert mentions, publish, update, or post comments.

Decision guide

Which Confluence voice path fits?

Choose by destination, connected context, saved artifact, visibility, and final publication action.

If you needChooseWhy
Generate a new Confluence draft from a spoken briefCreate with Rovo speech-to-textIt turns a prompt and optional connected context into a reviewable draft or canvas.
Write prose in a chosen editor fieldmacOS Dictation or IraVoiceThey can provide text to a supported focused field without selecting a space or publishing.
Speak a prompt to the desktop assistantRovo Desktop dictationIts microphone creates or edits the assistant prompt and may use optional screen context.
Preserve a recording and transcriptLoom in ConfluenceThe recording remains in Loom and carries separate sharing controls.
Collaborate with immediate visibilityConfluence live docLive docs have no draft state, so recognized edits appear to permitted collaborators immediately.
Publish, update, add, or commentReview Confluence’s explicit controlSpeech input does not confirm the destination, audience, accuracy, or account action.

Primary sources

Use the current documentation.

This guide was reviewed July 26, 2026. Product versions, settings, policies, and availability can change.

  1. Atlassian: Create and edit content in Confluence

    Current content types, Create with Rovo, prompt, sources, templates, speech-to-text helper, preview, editing, and Add to Confluence workflow.

  2. Atlassian: Create, update, and manage published content

    Current page and live-doc states, collaborative editing, autosave, permissions, publish shortcuts, unpublished changes, history, comparison, and restore workflow.

  3. Atlassian: Keyboard shortcuts, markdown, and autocomplete

    Current edit, quick-insert, comment, publish, publish-with-note, markdown, and slash-command shortcuts.

  4. Atlassian: Get started with Rovo Desktop

    Current Atlassian and Rovo requirement, microphone and screen permissions, prompt dictation, and screenshot context.

  5. Atlassian: Use Loom in Confluence Cloud

    Current recording surfaces, Loom library storage, Confluence restrictions, Loom permissions, and sharing boundary.

  6. Atlassian: Loom transcription and closed captions

    Current post-recording transcript and closed-caption workflow.

  7. Atlassian: AI availability and administration

    Current Standard, Premium and Enterprise availability, administrator controls, Government organization, and sandbox limitations.

  8. Apple: Dictate messages and documents on Mac

    The separate system Dictation shortcut, focused-field workflow, stop controls, and configuration-specific processing notice.

  9. Apple: If Dictation on Mac does not work as expected

    Current settings, shortcut, language, microphone, connection, Voice Control, and destination-field troubleshooting.

Questions

What readers usually ask next.

Does Confluence have voice typing?

Confluence currently lists speech-to-text inside Create with Rovo. For ordinary page, live-doc, title, or comment text on Mac, use a separate supported focused-field Dictation path.

What does Confluence Create with Rovo speech-to-text do?

It supplies a spoken prompt to an AI-assisted creation workflow. Rovo can use linked or attached context, prepare a draft or canvas, and show a preview before you choose Add to Confluence.

Can Rovo Desktop dictate into any Confluence field?

Its documented dictation control creates or edits prompts inside Rovo Desktop. It is not a generic microphone for every Confluence editor field.

Are dictated changes immediately visible in a Confluence live doc?

Yes, to people who can access that live doc. Live docs have no draft state, so changes are immediately visible. Pages instead use draft and publish or update states.

Is a Loom transcript the same as Confluence Dictation?

No. Loom starts with a retained recording stored in the Loom library and can provide a transcript after processing. Dictation supplies live text to a supported focused editor field.

Do Confluence and Loom use the same permissions?

Not necessarily. A Loom recording embedded in Confluence can be affected by both the Confluence content restrictions and the Loom recording permissions. Review both before sharing.

Does IraVoice integrate with Confluence or Rovo?

No. IraVoice can provide local text to supported focused fields or create reviewable Markdown. It does not access spaces, pages, Rovo, Loom, permissions, connected sources, or publication controls.

macOS 26+ · Apple-silicon

Try the local Mac workflow for yourself.

Dictate ordinary text or turn a rough technical thought into reviewable Markdown during the 14-day full trial.