November 5, 2026
Microsoft is working on developing an advanced technology based on artificial intelligence, which allows people to modify their speech and lip movement during a live broadcast, so that the speaker appears as if he uttered phrases other than his original words.
This was stated in a patent application submitted by the company under the title “Artificial Intelligence enhancement of video or augmented reality displays for live events in near-real time,” as the technology aims to process sound and image simultaneously and without the need for the recording to end.
According to what was reported by the Patientlyz website, the system relies on artificial intelligence agents that monitor the live broadcast to determine the words to be changed based on specific criteria or user preferences, so that generative artificial intelligence produces alternative wording and integrates it into the audio and video tracks.
The system’s capabilities are not limited only to replacing words, but also include reshaping video frames and modifying lip movements to match new phrases. This ensures that the edits appear naturally to the viewer, and the processing time ranges between approximately one second and ten seconds.
Suggested uses for the technology include eliminating filler words and hesitations such as “ah” and “um,” correcting stuttering and mispronunciations, removing offensive or inappropriate phrases, as well as replacing colloquial expressions and translating speech into other languages with lip-syncing to match the new language.
According to the patent document, the technology can be operated across a variety of devices including smart glasses, phones, tablets, and video cameras, and its potential applications include meetings, lectures, television programs, sports broadcasts, concerts, and social media platforms.
On the other hand, this technology raises serious questions about the credibility of visual content, as it is difficult for the viewer to distinguish between real words and those added by artificial intelligence, especially with voice modifications synchronizing with changing lip movement. In this context, the document indicated the possibility of retaining the original audio and video recording for comparison when needed.
It is noteworthy that the mere fact that Microsoft has filed a patent application does not necessarily mean launching the technology commercially or setting a date for its official launch.
