// AI news · 2026-11-07 · cnBeta
Microsoft patents real-time speech rewriting for live broadcasts
Microsoft has filed a patent for a real-time AI system that modifies live speech content and synchronizes facial movements in video streams.
Published 2026-11-07 22:13 (Beijing time)Source cnBetaAuto-summarised, fact-checked against source
01Summary
The system detects and edits spoken phrases using generative AI, updating lip and face animations to match revised audio. It supports translation, removal of filler words, error correction, and content filtering. Raw audio and video are preserved as baseline records with digital markers for traceability. This is a conceptual patent, not indicating current product deployment or commercialization.
Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.
02Key points
- Modifies live speech in real time and syncs facial movements to updated audio.
- Supports translation, filler word removal, error correction, and content filtering.
- Preserves original audio-video as baseline with digital modification markers.
- Patent is conceptual; no product launch or commercialization confirmed.
03Source
微软申请“实时改写发言”AI专利 可修改直播内容并同步调整口型
cnBeta · published 2026-11-07 22:13 Beijing time
05Related news
From AIMap AI news: collected from 1,100+ official and media sources. Summaries only restate the original article; follow the link for the full story. All stories of 2026-11-07 · Archive by date · All AI news · Today