Latest Microsoft patent shows new Copilot feature that generates live images as you speak

Microsoft has been betting big on AI. There’s no secret about that. The Redmond company has been at the forefront of the latest AI race, thanks to Copilot—Windows’ AI assistant—& its billions of investment in ChatGPT maker OpenAI.

And now, a recently published patent shows that Microsoft may be working on a new Copilot-powered feature. The US Patent and Trademark Office (USPTO) posted a 20-page document recently on October 10, 2024, about an AI-powered system for a system to convert live audio into images.

Initially filed on April 5 last year, Microsoft mentions in the document that this AI-powered system converts live audio (during meetings or conferences, for example), and then a language model summarizes that. And then, with this summary, the system creates an AI-generated image.

Here are some of the sketches from the technical document:

In the process, as users speak, their words are captured by a microphone and converted into a text transcript. The transcript is then segmented into phrases, and each segment is summarized using a language model to generate prompts for image generation.

We’re expecting this feature to launch for Teams, because these AI-generated images are then displayed on a screen in real-time as the live audio continues. So, when the speaker moves to another topic or speaks about something else, the live images also change.

“When images are used to complement verbal communication, they can help to clarify concepts and make them easier to understand, which can be especially beneficial for individuals who learn better through visual aids,” Microsoft says when describing the idea behind this tech.

Copilot in Teams comes in the Microsoft 365 Copilot subscription plan, which caters more to businesses and enterprises than the personal-only Copilot Pro plan. It also has the AI assistant in all MS365 apps like Word, PowerPoint, Excel, Outlook, and more, besides enterprise-grade security and Copilot Studio.

But still, though, being a patented tech doesn’t always mean that the feature is launching. It’s a common sight in the tech industry where companies file a patent application to prevent competitors from having the same tech, so, it’s best to take this with a grain of salt.