Beam Selection and Merging Using Noise-Floor and SNR Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing beam selection techniques in audio systems often misidentify desired speech under noisy conditions, particularly when significant non-stationary noise is present, leading to ineffective wakeword detection and speech processing performance.
Innovation Solution
Devices determine beam-specific signal quality metrics, including a minimum noise floor and signal-to-noise ratio, to select a group of beams for beam merging, prioritizing low background noise and high SNR, and perform weighted sum calculations to generate a combined output signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional beam selection techniques are used, then the system operates with simple selection logic, but beam selection accuracy deteriorates under noisy conditions with non-stationary noise
Solution Approach 1:
The patent changes the selection criteria from simple signal strength to a composite metric incorporating signal-to-noise ratio and noise stationarity detection. This parameter transformation enables the system to distinguish between stationary background noise and non-stationary interfering noise, thereby improving beam selection accuracy in noisy environments.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring noise characteristics and adjusting beam selection in real-time. The noise stationarity detector provides feedback about the nature of background noise, allowing the beam selector to adapt its strategy dynamically, which improves performance under varying noise conditions.
2Reliability
If multiple beamformed audio data are combined, then speech processing robustness improves, but computational complexity increases
Solution Approach 1:
The patent merges multiple beamformed audio data streams into a single composite signal after selecting beams based on noise characteristics. This combining approach enhances speech processing robustness by integrating information from multiple directional sources while suppressing noise through the selective merging process.
Solution Approach 2:
The system segments the audio processing into distinct stages: beamforming, noise characterization, beam selection, and final combining. This segmentation allows each stage to be optimized independently, managing computational complexity by breaking down the overall processing into manageable, specialized sub-tasks.
3Measurement precision
If beam-specific noise floor analysis is performed, then noise characterization accuracy improves, but processing time increases
Solution Approach 1:
The system performs noise floor analysis on a subset of beams rather than all available beams. By applying noise characterization selectively to candidate beams identified through initial screening, the system achieves sufficient noise characterization accuracy while reducing the overall processing time and computational burden.
Data Source
AI summary
A system that performs beam selection and beam merging using beam-specific signal quality metrics corresponding to a minimum noise floor. For example, a device may track a minimum noise floor for each beam, determine a highest minimum noise floor across the beams, and determine a noise floor ratio between the beam-specific minimum noise floor and the highest minimum noise floor. Using a combination of the noise floor ratio and signal-to-noise ratio (SNR) values, the device may perform beam selection by prioritizing low background noise as well as high SNR to select a pre-defined beam group. In addition, the device may use the noise floor ratio to perform beam merging and generate single-channel output audio data using the selected beam group. For example, the device may scale the beams based on a combination of the SNR value and the noise floor ratio.