This week's Audio & Sound category includes 12 filed patent applications from 5 companies: Sony (6), Microsoft (2), Voyetra Turtle Beach (2), Nvidia (1), and NetEase (1).
Sony's applications span AI systems that select music to score gameplay clips, use acoustic signals for indoor positioning, let audio inputs control NPC behavior, generate dynamic boot sounds based on context, detect speaker obstructions on handhelds, and enable event-driven communication adjustments in gaming headphones. Microsoft filed applications for AI-powered real-time audio transformation that allows players to replace or modify specific sounds like gunshots or explosions. Voyetra Turtle Beach's applications describe headsets that provide real-time coaching based on detected in-game audio and controllers with integrated screens for audio adjustments, while Nvidia filed an application for camera-based head tracking to deliver spatial 3D audio and NetEase described dynamic sound systems that adapt to player skill and scene context.
Microsoft filed 2 patent applications for AI-powered audio transformation systems that let players customize in-game sounds in real time. Both applications describe technology that uses generative AI to identify and modify specific sound classes, such as gunshots or explosions, to help players dealing with PTSD, misophonia, or phonophobia. The systems go beyond traditional volume controls by semantically understanding and transforming sounds within dynamic audio streams, without requiring individual game developers to add support. One application highlights a feedback loop that refines the model when sounds are missed or incorrectly processed, while both describe the ability to train personalized models using user-provided audio samples and sync them across devices via a global account.
Sony received 6 patent applications covering diverse audio technologies for gaming. One application describes a system that automatically selects songs from a player's personal music library to score gameplay clips, matching tempo and mood to on-screen action through machine learning-based similarity scoring. Another uses spread-spectrum sound signals and reflected waves to track headsets and smartphones for indoor positioning, distinguishing between direct and reflected sound paths using unique spreading codes. A third treats audio as a control input, allowing music, voice, and ambient sounds to drive NPC behavior and synchronize in-game interactions through sentiment and beat detection. Sony also filed an application for AI-generated boot sounds that change based on time, location, weather, and user preferences rather than playing a fixed startup chime. The remaining 2 applications address hardware-level audio challenges: one detects when players' hands obstruct speakers on handheld devices and intelligently reroutes priority audio to unobstructed speakers, while the other reconfigures voice communication channels in gaming headphones automatically based on in-game events and adds physical NFC attachments that unlock game powers.
Voyetra Turtle Beach filed 2 patent applications focused on enhancing player experience through integrated audio controls. One application describes a headset that analyzes in-game audio in real time and delivers contextual voice coaching based on detected sounds, their direction, and intensity, all without requiring games to provide structured event data. The other embeds headset audio controls directly onto a game controller screen, syncing with a companion app to enable on-the-fly adjustments during gameplay without interrupting the action or navigating system menus.
NetEase filed 1 patent application for a dynamic audio system that adjusts sound playback based on both player skill state and sub-scene context. The technology binds audio parameters directly to skill state data derived from gameplay, creating a two-layer control system that combines player preferences with real-time modulation driven by in-game performance rather than relying solely on fixed audio rules or manual settings.
Nvidia filed 1 patent application for a spatial audio system that uses camera-based head tracking to deliver immersive 3D sound. The technology links visual head-pose estimation directly to audio generation through an optimized pipeline designed to reduce the computational costs typically associated with head-tracking audio systems, making the approach more efficient for VR, gaming, and other real-time applications.
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.