Online Speaker Calibration for Echo Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Speakerphones face challenges in maintaining high-quality acoustic echo cancellation due to aging microphones and speakers, and interference from noise sources like fans and air conditioning, which affect voice discernment in conference calls.
Innovation Solution
A method for online calibration that involves computing midrange and lowpass sensitivities from input and output signal spectra, performing iterative searches for model parameters, and updating averages to improve echo cancellation, while also applying notch filters to remove resonant frequencies and monitoring signal power to detect potential issues with speakers or microphones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional echo cancellation is used with aging microphones and speakers, then the system can maintain basic communication functionality, but the acoustic echo cancellation quality deteriorates over time
Solution Approach 1:
The system dynamically updates the transfer function model parameters of the speaker and microphone in real-time during operation. Instead of using fixed calibration data, the system continuously adapts the model to account for aging and environmental changes, maintaining accurate echo cancellation performance throughout the device's service life
Solution Approach 2:
The system uses feedback from the actual acoustic measurements during conference calls to refine and update the transfer function models. By monitoring the residual echo and adjusting the model parameters accordingly, the system maintains optimal echo cancellation performance despite component aging
2Reliability
If the system processes audio signals continuously to maintain echo cancellation, then communication quality is maintained, but computational resources and processing time are consumed
Solution Approach 1:
The system performs preliminary calibration and model estimation during periods when audio processing is not critical, such as during silence periods or when the conference call is not active. This pre-computation reduces the computational burden during active communication phases
Solution Approach 2:
The system applies partial updates to the transfer function models only when necessary, such as when detecting significant changes in acoustic conditions or when calibration data becomes available. This selective updating approach reduces unnecessary computational processing while maintaining communication quality
3Measurement precision
If the system uses multiple microphones to improve voice discernment, then the ability to distinguish participants improves, but the complexity of processing and noise filtering increases
Solution Approach 1:
The system segments the audio processing task by separating direct echo cancellation from noise filtering operations. The transfer function models handle the echo cancellation for each microphone channel independently, while a unified noise filtering stage processes the combined signals, reducing the complexity of multi-microphone processing
Data Source
AI summary
A system may include a processor and memory. The processor may be configured to perform calibration measurements on the speaker even when the speaker is being used to conduct a live conversation. The processor may be configured to: provide a live output signal for transmission from a speaker; receive an input signal corresponding to the output signal; compute a midrange sensitivity and a lowpass sensitivity for a transfer function derived from a spectrum of the input signal and a spectrum of the output signal; subtract the midrange sensitivity from the lowpass sensitivity to obtain a speaker-related sensitivity; perform an iterative search for current parameters of a speaker model using the input signal spectrum, the output signal spectrum and the speaker-related sensitivity; and update averages of the speaker model parameters using the current parameter values. The parameter averages may be used to perform echo cancellation.