Microphone Array Algorithm on SigmaDSP

Hi,

      One of my customer is interested in designing a Microphone Array for their mulit-channel conference system to capture the human voice. I was wondering if we have some SigmaDSP products are suitable for this kind of application. Though ADSP-21584 is used by the customer, the MIPS resource is limited.

    the customer would like the algorithm function as below:

     1.  Echo Cancellation

     2.  Noise Reduction

              We definitely have these two algorithm on Sigma300/Sigma350. What I am not sure is that if it is suitable for Microphone Array along with Beamforming. 

     3.  Beamforming 

            We have a adaptive Beam former in ADAU1761 in this link:

            https://ez.analog.com/dsp/sigmadsp/f/discussions/66727/sigmadsp-two-microphone-beamformer

           However, the customer would like to capture the voice from 360°. It means the circle/ triangle Microphone Array(at least 4 microphones) may be more suitable. So maybe the Adaptive Beam Former we have now in SigmaStudio is not suitable for this customer.  

           And I also notice we have EVAL-BOARD for Microphone Array along with EVAL-ADAU1467Z.

          https://www.analog.com/media/en/technical-documentation/user-guides/EVAL-MICCANVASZ-UG-1296.pdf

          I am curious that if we have another beamforming designing tool for this board. Or is that possible design a adaptive beam former with the fundamental block in the SigmaStudio.  

   

     4.  Human voice location detection

         I suppose this is kind of algorithm like VAD and DOA. I notice that we have a VAD module in SigmaStudio but it seems cannot recognize the if the voice is from human beings. And do we have DOA algorithm in SigmaStudio?

     So do you have some suggestion DSP products with reference algorithm for this kind of application:  Microphone array with Voice Capture Front-end Processing Algorithm.  

B.R.

Yosen