Residual echo detection method and apparatus

Abstract

The invention relates to the field of voice communication, and discloses a residual echo detection method and apparatus. In the invention, according to a voice spectrum statistical characteristic, within a frequency range with a voice signal-to-noise ratio, an average coherent coefficient of several frequency ranges is utilized to replace a conventional full-frequency range average coherent coefficient as a residual echo detection statistical amount. Since in a frequency range with quite low voice energy, the voice signal-to-noise ratio is quite low, a coherent coefficient is greatly influenced by noise; however, in a frequency range with quite high voice energy, the voice signal-to-noise ratio is quite high, and the coherent coefficient is less influenced by the noise. Therefore, the interference resistance of the full-frequency range average coherent coefficient is lower than the frequency range with the quite high voice energy, so that in actual application, the detection method of a selected frequency range can be provided with a quite high interference resistance capability. At the same time, since the calculation of the full-frequency range average coherent coefficient is unnecessary, the complexity of the detection method is greatly reduced.

Description

technical field [0001] The invention relates to the field of voice communication, in particular to a method and device for detecting residual echo. Background technique [0002] In a voice communication system, after the remote input signal arrives at the local signal receiving device, it is processed by the local signal receiving device and reaches the earpiece. During this process, echoes are often generated. In the prior art, an acoustic echo canceller is usually used to cancel the echo, and the acoustic echo canceller involves an adaptive filter, double-ended detection and residual echo suppression, such as figure 1 shown. [0003] Usually, the acoustic echo canceller applies an adaptive filter to synthesize the echo, and adjusts the filter coefficient adaptively according to the error signal to eliminate the linear echo. After adaptive filtering, some echoes still remain in y(n), mainly because: [0004] 1) The echo generated when the adaptive filter does not reach a...

AI Technical Summary

Problems solved by technology

[0005] 2) Even in the stable state of the filter, due to the consideration of algorithm complexity, the echo generated by the filter length is smaller than the echo path; and the echo generated by sacrificing tracking accuracy and other factors to speed up the convergence speed

[0006] 3) The adaptive filter can only eliminate linear echo, but in the actual communication environment, in the echo transmission path, the nonlinear distortion of microphones, speakers, hardware amplifiers, A / D, D / A and other devices will bring nonlinear distortion echo

[0017] Method 1 and Method 2 can accurately detect the residual echo when the near-end background noise is small; as the near-end background noise increases, the accuracy of residual echo detection gradually decreases

Method 3 has increased anti-interference ability compared with the first two methods, but the computational complexity is higher

Images