PowerPoint users on Mac can now generate captions using speech recognition

The Closed Captions pane in PowerPoint for Mac

To mark Global Accessibility Awareness Day, Microsoft has announced a major accessibility upgrade for PowerPoint on Mac: users can now generate captions for embedded videos directly within the app using speech recognition.

This development builds upon previous accessibility enhancements from Microsoft. For instance, the company expanded SRT file support in PowerPoint back in January, making it easier for users across platforms to import professionally created or self-made caption files for videos in their slides.

Peter Wu, a Principal Software Engineer on the PowerPoint team, emphasized the need, noting that captions are essential for the 5% of the world's population who are deaf or hard of hearing, and that 50% of Americans actually prefer watching videos with captions most of the time. Here is how Microsoft says the new generation feature works:

Open a new or existing presentation in PowerPoint on your Mac device and embed a video.

Select the video, then select Playback > Edit Captions. You can also find the Edit Captions button in the context menu that appears when you right-click on the video, or in the Accessibility tab that appears when Accessibility Assistant is open.

Select the Generate Captions from: dropdown menu under the Closed Captions pane on the right side of your presentation, then select the language spoken in the video. When the captions are done generating, they will appear in the Closed Captions pane.

Select a text cue to view that part of the video. Compare the spoken dialogue with the text cue and edit if needed.

Translate captions into additional languages by selecting the Translate to: dropdown menu, then selecting the language you want to translate them into.
NOTE: The translated captions will appear in the Closed Captions pane when translation is done.

Select to see the list of all captions tracks for the video.

From here, insert more tracks, remove a track, and/or reorder the tracks.

If you have a favorite app or service for captioning videos that can save the captions as a WebVTT or SRT file, insert the file into your presentation by selecting Insert Captions from File in the Closed Captions pane.

NOTE: You can select multiple files at once.

Once generated, these captions are not just slapped on and forgotten. Users can review and edit them for accuracy or clarity, apply formatting like bold or italics, and even translate them into other languages. The system also supports adding speaker names or describing non-dialogue sounds, like "[phone ringing]". Microsoft also points out that presenting using PowerPoint Live in Teams allows each participant to select their preferred captions track independently.

This caption generation feature is currently rolling out to PowerPoint for Mac Beta Channel users running Version 16.98 (Build 25050401) or later. As is typical with beta features, it might take a bit for everyone in the channel to see it.

Speaking of accessibility, Google also, in celebration of yesterday's Global Accessibility Awareness Day, announced a slate of new AI-driven features for Android and Chrome. These included advancements like Gemini-powered image descriptions in TalkBack and more expressive real-time captions.