Multi-Device Audio Capture Relevance Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for capturing multi-view audio struggle with accurately combining audio signals from multiple recording devices, especially in scenarios with inaccurate or unavailable device position measurements, leading to compromised audio quality and increased complexity, particularly in large-scale events.

Innovation Solution

A method that selects recording devices based on relevance levels determined by their perceptual importance, combining only the audio signals from relevant devices to generate high-quality combined audio signals, regardless of device positions, using techniques like similarity analysis and relevance ranking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If audio signals from all recording devices are combined, then audio coverage is improved, but computational complexity increases significantly

Engineering Contradiction:
Improveaudio coverageVSAvoidcomputational complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and combines only the most relevant audio signals from recording devices based on their proximity to the listening position, rather than combining all available signals. This selective extraction approach maintains audio coverage quality while significantly reducing computational complexity by processing only a subset of signals.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by prioritizing recording devices based on their spatial relationship to the desired listening position. Devices closer to the listening position are given higher relevance and their signals are combined with greater weight, creating a locally optimized audio mixture that improves efficiency without sacrificing coverage.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If device position measurements are used to select recording devices, then selection accuracy is improved, but measurement precision requirements increase system complexity

Engineering Contradiction:
Improveposition measurement accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements self-service by allowing recording devices to autonomously determine their own relevance based on simple position measurements and the listening position information. Each device calculates its own relevance metric without requiring complex centralized processing or high-precision measurement infrastructure, reducing overall system complexity.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent changes the parameter of position measurement precision requirements by using relative position information and relevance calculations that work with moderate-precision measurements. Instead of requiring high-precision absolute positioning, the system uses parameter transformations that convert position data into relevance scores, allowing accurate device selection with simpler measurement systems.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If relevance levels are calculated for all recording devices, then selection accuracy is improved, but processing time increases

Engineering Contradiction:
Improvedevice selection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the set of recording devices into relevance groups based on simplified criteria such as distance thresholds or signal strength ranges. Instead of calculating detailed relevance levels for all devices, the system divides them into segments (e.g., near-field, mid-field, far-field devices) and processes each segment with appropriate selection criteria, reducing overall processing time while maintaining selection accuracy.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS9913067B2Processing of multi device audio capture
Publication Date: 2018.03.06 NOKIA TECHNOLOGIES OY
  • US9913067B2 patent drawing
  • US9913067B2 patent drawing
  • US9913067B2 patent drawing

AI summary

It is inter alia disclosed to select, by a first apparatus, one or more recording devices to obtain a set of selected recording devices, or to receive, at the first apparatus, information on a set of one or more selected recording devices. The selected recording devices in the set of selected recording devices are selected from a set of recording devices at least based on relevance levels determined for the recording devices of the set of recording devices. A combination of audio signals recorded by the selected recording devices of the set of selected recording devices in an audio scene to obtain one or more combined audio signals for rendering is performed or caused by the first apparatus.