Echo eliminator and echo cancellation method

Abstract

The related echo eliminator comprises: a linear predicted decoder to decode data into remote speech signal and extract the short activating signal and linear prediction coding parameter, a decorrelation filter to receive the predicted coding parameter and error signal for decorrelation and generate decorrelated error signal, an adaptive filter to receive all of the remote speech and short activating signal and decorrelated error signal and generate the estimated echo, and a subtracter to generate the error signal. This invention combines with LPC device, reduces algorithm complexity, and prompts algorithm convergence.

Description

【Technical field】 [0001] The invention relates to an echo canceller and an echo cancelling method, in particular to an echo canceller and an echo cancelling method used in a communication system and a terminal using a linear predictive coding (LPC) vocoder. 【Background technique】 [0002] In recent years, the development of the Internet has promoted the wide application of VoIP (Voice over Internet Protocol) technologies such as VoIP and video conferencing. At the same time, its voice quality is also increasingly concerned, and one of the most critical factors is the impact of echo. Because the sound played by the earpiece speaker is picked up by the microphone and sent back to the far end, coupled with the effects of various delays, the far-end talker can hear his own echo, which seriously affects the quality of the call. [0003] At present, the most effective way to solve this problem is to use an adaptive filter to construct the acoustic echo cancellation system shown i...

AI Technical Summary

Problems solved by technology

When the signal input to the adaptive filter is a highly correlated signal such as speech, the convergence speed of the algorithm will be significantly reduced, which will affect the quality of echo cancellation, so that the echo cannot be effectively eliminated

Other improvement methods, such as the commonly used method of preprocessing the input signal through the forward prediction filter to reduce the correlation of the signal, can significantly improve the convergence speed of the algorithm, but it needs to continuously calculate and update the filter coefficients. Added extra computation

In addition, in a time-varying, non-stable speech environment, the NLMS algorithm is not ideal for the control of the step size factor. It replaces the expected value in the steepest gradient method with the instantaneous value, which will inevitably lead to the randomness of the steady-state misalignment error. Fluctuations, which will affect the performance of the algorithm and reduce the effect of echo cancellation

Images