To linearly cancel and suppress the returned echo, both Acoustic Echo Cancellation (AEC) and Acoustic Echo Suppression (AES) functions are required. The difference between AEC and AES is AEC can cancel the linearly coupled echo by maintaining the desired near-end speech. With the better AEC performance, the user can easily achieve full-duplex speech communication. But nonlinear echo generated by speakerphone housing, imperfect speaker/microphone devices and undesired echo environment are not canceled by AEC. Hence, the function of AES is to suppress the input signal mixed with the desired signal and unwanted echo signal in the frequency domain based on the information of the VAD. Therefore, the user must carefully select the parameters for AES and Double-Talk Threshold carefully to prevent from the degradation of the desired speech quality in the case of strong nonlinear echo or close placement between the microphone and speakers. The following figure illustrates the AEC tuning interface in the DSP configuration tool.Figure 4-3. AEC Tuning Interface in DSP Configuration Tool
The following sections explain the adjustable parameters to fine tune the performance of echo cancellation.
The online versions of the documents are provided as a courtesy. Verify all content and data in the device’s PDF documentation found on the device product page.
