Dual Acoustic Echo Cancellation for Tone Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing networked microphone devices face challenges in effectively canceling acknowledgement tones to avoid interference with voice utterances, particularly when the Acoustic Echo Canceller (AEC) is inactive or has not converged in time.
Innovation Solution
Implementing two acoustic echo cancellation processes: a first AEC that cancels acoustic echoes during active audio playback, and a second AEC that converges faster to cancel acknowledgement tones when the speakers are idle, by processing only the specific frequency ranges where the acknowledgement tone has content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a single acoustic echo canceller is used to cancel all acoustic echoes, then the device complexity is reduced, but the convergence speed is insufficient when speakers are idle and acknowledgement tones need to be cancelled
Solution Approach 1:
The patent divides the acoustic echo cancellation function into two separate AEC processes: a first AEC that operates during active audio playback and a second AEC that operates when speakers are idle. This segmentation allows each AEC to be optimized for its specific operational context, with the second AEC converging faster since it only needs to handle acknowledgement tones rather than full audio content.
2Measurement precision
If the first AEC is used during idle periods to cancel acknowledgement tones, then the signal-to-noise ratio improves, but the AEC has not converged in time causing interference with voice utterances
Solution Approach 1:
The system activates the second AEC in advance during idle periods before voice utterances occur. This preliminary action ensures that the AEC has sufficient time to converge and cancel acknowledgement tones before the voice command is processed, improving the signal-to-noise ratio without causing interference.
3Reliability
If the second AEC processes all frequency ranges, then the cancellation is comprehensive, but the convergence time increases significantly
Solution Approach 1:
The second AEC is configured to process only specific frequency ranges where acknowledgement tones have content, rather than processing the entire audio spectrum. This localized approach maintains effective cancellation for the relevant frequencies while significantly reducing convergence time and computational complexity.
Data Source
AI summary
Example techniques involve systems with multiple acoustic echo cancellers. An example implementation captures first audio within an acoustic environment and detecting, within the captured first audio content, a wake-word. In response to the wake-word and before playing an acknowledgement tone, the implementation activates (a) a first sound canceller when one or more speakers are playing back audio content or (b) a second sound canceller when the one or more speakers are idle. In response to the wake-word and after activating either (a) the first sound canceller or (b) the second sound canceller, the implementation outputs the acknowledgement tone via the one or more speakers. The implementation captures second audio within the acoustic environment and cancelling the acoustic echo of the acknowledgement tone from the captured second audio using the activated sound canceller.