Meta received 1 granted patent in Audio.
The patent covers spatial Audio technology for virtual and augmented reality environments, designed to enhance speech clarity by using perceptual cues that help users focus on target speakers while filtering out background noise. This approach addresses the challenge of maintaining conversational intelligibility in immersive settings where multiple Audio sources may be present.
The Audio patent takes a perceptual approach to the cocktail party problem in extended reality. Instead of relying on straightforward signal-to-noise ratio improvements, the system builds statistical models of how humans actually perceive and separate different speakers. It analyzes characteristics like pitch patterns and spatial positioning of sound sources, calculating interaural level differences and the dynamic range of interaural time differences to determine which speaker a user is trying to focus on. By modeling these biological cues computationally, the technology can selectively amplify the target speaker's voice in a way that aligns with natural human auditory perception.
All data sourced from USPTO patent filings. Google Patents may take several weeks to index recent publications. If a link is unavailable, search for the patent number at USPTO Patent Public Search.