Spatial Comfort Noise Generation for Conference Audio Continuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conferencing systems face challenges in maintaining perceptual continuity due to the buildup of comfort noise when discontinuous transmission occurs, especially in systems that combine multiple audio streams, which can lead to an unpleasant absence of far-end presence for listeners.
Innovation Solution
The system generates and renders spatial comfort noise at the receiving endpoint, with spectral and spatial properties matching those of typical comfort noise, ensuring continuous audio presence by combining it with received audio signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comfort noise is added to maintain perceptual continuity during discontinuous transmission, then the sense of presence is improved, but noise buildup occurs when multiple audio streams are combined
Solution Approach 1:
The patent extracts the comfort noise generation function from the conference server and relocates it to individual endpoints. Each endpoint independently generates its own comfort noise locally rather than receiving mixed comfort noise from the server, thereby preventing noise buildup from multiple streams while maintaining perceptual continuity during discontinuous transmission
Solution Approach 2:
Each endpoint serves itself by generating comfort noise locally at the receiving endpoint rather than relying on the server to provide mixed comfort noise from multiple streams. This self-service approach eliminates the harmful noise buildup effect while maintaining the beneficial perceptual continuity
2Reliability
If spatial properties are used in conferencing systems, then audio realism is improved, but maintaining continuity between intended and synthetic audio segments becomes more difficult
Solution Approach 1:
The patent applies local quality by generating comfort noise with specific spatial properties (such as interaural time differences and interaural level differences) that match the characteristics of the intended audio stream. Each endpoint configures the spatial properties of its locally generated comfort noise to be consistent with the spatial characteristics of the active audio streams, thereby maintaining audio realism while simplifying continuity management
Solution Approach 2:
The system dynamically adjusts the spatial parameters (such as interaural time difference and interaural level difference) of the generated comfort noise to match the characteristics of the intended audio stream. By changing these parameters locally at each endpoint, the system maintains continuity between intended and synthetic audio segments while preserving spatial audio realism
Data Source
AI summary
A method, an apparatus, logic (e.g., executable instructions encoded in a non-transitory computer-readable medium to carry out a method), and a non-transitory computer-readable medium configured with such instructions. The method is to generate and spatially render spatial comfort noise at a receiving endpoint of a conference system, such that the comfort noise has target spectral characteristics typical of comfort noise, and at least one spatial property that at least substantially matches at least one target spatial property. On version includes receiving one or more or more audio signals from other endpoints, combining the received audio signals with the spatial comfort noise signals, and rendering the combination of the received audio signals and the spatial comfort noise signals to a set of output signals for loudspeakers, such that the spatial comfort noise signals are continually in the output signal sin addition to output from the received audio signals.