Smart Loudspeaker Echo Cancellation via Shared Filter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Smart loudspeakers with voice control face challenges in processing power demands for multi-channel audio and echo cancellation while maintaining audio quality and portability, particularly in suppressing background noise and directing sound effectively in a room.
Innovation Solution
A smart loudspeaker design featuring a circular array of speaker elements and microphones with a digital signal processor that uses finite input response filters, rotation matrices, and adaptive acoustic echo cancellers to generate steerable audio beams and suppress echo, allowing for efficient processing and high-quality audio playback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple AEC filter pairs are used for each microphone element, then echo cancellation performance is improved, but processing complexity and power consumption increase
Solution Approach 1:
The patent combines multiple AEC filter pairs into a single shared AEC filter that serves all microphone elements. Instead of having separate AEC filters for each of the 6 microphone elements (which would require 6 filter pairs), a single AEC filter processes signals from all microphones, significantly reducing processing complexity and power consumption while maintaining effective echo cancellation across the array.
Solution Approach 2:
The single AEC filter is designed to perform multiple functions simultaneously - it cancels echo for all microphone elements in the array and operates across multiple frequency bands. This universal filter handles the acoustic echo cancellation task for the entire microphone array rather than requiring dedicated filters for each element.
2Volume of moving object
If a small diameter microphone array is used, then device portability is improved, but microphone array performance may be degraded
Solution Approach 1:
The patent changes the spatial parameter of the microphone array by reducing the diameter of the circular array to approximately 10mm, making the device more portable. Despite this reduced physical size, the array maintains effective performance through careful design of the beamforming algorithms and signal processing techniques that compensate for the smaller aperture.
3Reliability
If complex signal processing is used for multi-channel audio and beamforming, then audio quality is improved, but processing power demands increase
Solution Approach 1:
The patent segments the audio processing into distinct frequency bands using multiple AEC filters operating at different frequency ranges. This segmentation allows for more efficient processing by handling different frequency components separately, reducing the overall computational burden while maintaining high audio quality across the full frequency spectrum.
Solution Approach 2:
The patent combines multiple processing functions into unified algorithms - the beamforming operation simultaneously performs spatial filtering, echo cancellation, and audio enhancement. By merging these functions rather than executing them as separate processing stages, the system reduces redundant computations and lowers power demands while delivering high-quality multi-channel audio output.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A digital signal processor is programmed to extract a center channel from a stereo input, apply the center channel to an array of speaker elements using a first set of finite impulse response filters and a first rotation matrix to generate a first beam of audio content at a target angle about the axis, apply a left channel of the stereo input to the array of speaker elements using a second set of finite impulse response filters and a second rotation matrix to generate a second beam of audio content at a first offset angle from the target angle about the axis, and apply a right channel of the stereo input to the array of speaker elements using a third set of finite impulse response filters and third rotation matrix to generate a third beam of audio content at a second offset angle from the target angle about the axis.