US20260341309
2026-10-29
Physics
G11B27/036
The patent application describes a system and method for enhancing video and augmented reality (AR) presentations of live events using artificial intelligence (AI). The technology aims to modify spoken words in live audio and video recordings based on user profiles and parameters, replacing them with alternative words. This process is performed by a computing system that utilizes a generative AI system to ensure smooth transitions in the audio and video, creating an enhanced viewing experience in near-real-time.
During live presentations, speakers often use filler words, personalized phrases, or colloquial terms. Traditional recording methods capture these verbatim, requiring skilled technicians to edit them out post-production. The current technology addresses this by providing an AI-driven solution that can modify and enhance recordings either during post-production or in near-real-time, thereby eliminating the need for extensive manual editing.
In post-production, the system identifies words in an audio file that need modification based on audience profiles. The AI modifies the script or transcript and subsequently alters the audio file to replace the identified words. The system also adjusts the corresponding video file to match the modified audio, ensuring a seamless integration of changes. This results in enhanced video and audio files that are presented to the audience with improved quality.
For live events, the system operates in near-real-time, identifying and modifying spoken words during the event. It replaces these words in the live audio and video recordings based on user profiles. The AI system generates smooth transitions for the modified content, allowing the audience to experience the enhanced presentation almost instantaneously. This real-time capability is particularly useful for live broadcasts or events where immediate processing is crucial.
The technology can be implemented using neural processing units (NPUs), graphics processing units (GPUs), and central processing units (CPUs) on user devices. It is applicable in various scenarios, including lectures, presentations, social media content, and live performances. By enhancing the quality of recordings, the system is beneficial for creating training materials and improving viewing experiences across different media platforms.