Microsoft patents AI that rewrites livestream speech in real time, including lip movements
kevinnbass · x · 2026-11-22
Microsoft has filed a patent application for an AI system that can rewrite what people say during livestreams in near real time. It flags words based on preset rules or viewer profiles (age, language, region), then uses generative AI to replace or remove them, smooth the audio, and adjust the speaker's mouth and facial movements for a seamless edit. Example uses: live translation, removing filler words and profanity, fixing mispronunciations, and age/region-based content adaptation. The original stream would be kept as a "ground truth" record. Commenters compared it to Soviet-style speech censorship.
More from Companies & People
- Dario Amodei once pitched immersive video games as a fix for AI-induced joblessness at OpenAI — zetalyrae · 2026-11-22
- Vaibhav: Most Companies Must Be 'Refounded' for the Post-AI World — vaibhavbetter · 2026-11-22
- Leaked chats claim Anthropic employees are dismissive of existential risk — teortaxesTex · 2026-11-22
- Ex-Qwen post-training researcher launches AlpsLab, an RSI startup eyeing $100M+ valuation — 量子位 · 2026-11-22
- Designer KOL: Google's model pace and quality suggest it will soon 'mog them all' — AIandDesign · 2026-11-22
- Apple reportedly planning biggest iPhone redesign in a decade for 2027 20th anniversary — aftahi_ai · 2026-11-22