DASH Client 3D Audio Adaptation Set Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in efficiently transporting and decoding three-dimensional (3D) audio data, particularly in adapting to different speaker geometries and acoustic conditions, using existing standards like MPEG-H 3D Audio, which struggles with accessing individual objects in a sound scene due to its reliance on spherical harmonic coefficients that convey spatial information implicitly.
Innovation Solution
The implementation of Dynamic Adaptive Streaming over HTTP (DASH) for streaming 3D audio data, where availability data is formatted according to MPEG-H 3D Audio, allowing for the selection and retrieval of scene-based, object-based, and channel-based audio adaptation sets, enabling flexible playback on various speaker configurations through an MPEG-H 3D Audio decoder that instructs a DASH client to retrieve specific audio data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If MPEG-H 3D Audio encoding is used to represent soundfield independently of speaker geometry, then adaptability to different speaker configurations is improved, but difficulty in accessing individual audio objects worsens
Solution Approach 1:
The patent segments the audio data into distinct adaptation sets: scene audio data (background noise), channel audio data (dedicated to particular channels/speakers), and object audio data (sound-producing objects in 3D space). This segmentation allows the system to maintain the adaptability of MPEG-H 3D Audio while enabling selective access to individual audio objects through separate adaptation set structures.
2Device complexity
If all audio data is provided in a single adaptation set, then device complexity is reduced, but loss of information about audio object locations and types increases
Solution Approach 1:
The patent adds a categorical dimension to the data structure by organizing audio data into three distinct adaptation sets (scene, channel, object) rather than a single flat structure. This dimensional organization preserves comprehensive audio object information while maintaining manageable complexity through structured separation.
3Reliability
If scene-based audio adaptation set is always retrieved, then completeness of audio reproduction is improved, but use of network bandwidth increases
Solution Approach 1:
The patent implements partial retrieval by allowing the DASH client to selectively retrieve only the necessary adaptation sets based on availability data and selection data. Instead of always retrieving all audio data (excessive action), the system retrieves only what is needed (partial action), optimizing bandwidth usage while maintaining audio reproduction quality.
Data Source
AI summary
In one example, a device for retrieving audio data includes one or more processors configured to receive availability data representative of a plurality of available adaptation sets, the available adaptation sets including a scene-based audio adaptation set and one or more object-based audio adaptation sets, receive selection data identifying which of the scene-based audio adaptation set and the one or more object-based audio adaptation sets are to be retrieved, and provide instruction data to a streaming client to cause the streaming client to retrieve data for each of the adaptation sets identified by the selection data, and a memory configured to store the retrieved data for the audio adaptation sets.