Decoder and Method for Multi-Instance Spatial-Audio-Object-Coding Employing a Parametric Concept for Multichannel Downmix/upmix Cases

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent describes a decoder and method for reconstructing multi-channel, object-based audio from a downmix signal with three or more channels. It introduces a flexible channel-processing approach that routes various downmix channels to different processing units, each reconstructing parts of the original object audio based on supplementary side information. This results in a scalable and efficient way to recover individualized audio objects or scenes from complex downmixes, enabling improved spatial audio experiences.

Use CasesContent extracted from patent full text and abstract with AI.

• Home theater systems that reconstruct immersive surround sound from compressed object-based audio.
• Broadcast and streaming services delivering customizable audio scenes, such as adjusting commentator and crowd noise in sports broadcasts.
• Music production and education platforms where users can isolate, adjust, or remix specific instruments or vocals.
• Teleconferencing systems where speaker intelligibility and spatial separation of voices are enhanced for participants.
• Virtual or augmented reality (VR/AR) applications providing real-time, dynamic spatial audio rendering based on user orientation.

BenefitsContent extracted from patent full text and abstract with AI.

• Enables efficient storage and transmission of high-quality, multi-object and multi-channel audio by reducing required bitrates.
• Allows end-user customization of audio scenes, such as changing the levels or positions of specific audio objects (e.g., vocals, instruments, speakers).
• Increases flexibility for audio system designers and broadcasters to offer immersive and interactive audio on a wide range of playback devices.
• Is compatible with both current and future audio formats, supporting scalability up to many channels and objects.
• Reduces computational complexity by enabling selective and parallel processing of audio channels, with the ability to bypass unnecessary operations.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A decoder for generating an audio output signal comprising one or more audio output Channels from a downmix signal comprising three or more downmix Channels, wherein the downmix signal encodes three or more audio object Signals is provided. The decoder comprises an input Channel router (110) for receiving the three or more downmix Channels and for receiving side information, and at least two Channel processing units (121, 122) for generating at least two processed Channels to obtain the one or more audio output Channels. The input Channel router (110) is configured to feed each of at least two of the three or more downmix Channels into at least one of the at least two Channel processing units (121, 122), so that each of the at least two Channel processing units receives one or more of the three or more downmix Channels, and so that each of the at least two Channel processing units (121, 122) receives less than the total number of the three or more downmix Channels. Each Channel processing unit of the at least two Channel processing units (121, 122) is configured to generate one or more of the at least two processed Channels depending on the side information and depending on said one or more of the at least two of the three or more downmix Channels received by said Channel processing unit from the input Channel router.