Recording a mono track from 4 MEMS mics array (STEVAL-MKI129V3)

baggiomichele · ‎2014-10-14

Posted on October 15, 2014 at 06:04

Hi,

i'm an italian videomaker interested on MEMS mic

I want to record some audio using your Evaluation Kit (STEVAL-MKI126V1 + MKI129V3 connected to my laptop through the MKI138V1

).

I've installed APWorkbench on Windows 64bit running on Parallels Desktop.

FTDI missing but i can run the app for the STSmartvoice Demo Kit.

What i need is capturing sound from the 4 mics on the 129 board and mix them to a mono file.

The main reason is to get a higher SNR for a better audio quality.

Could you please show me how to send CH3 4 5 6 to only one OUT Channel?

More. I can control the volume only of mic 4 and 5.

It would be great to solve this problem before my shooting next week.

Thanks so much.

Michele

#mki138v1-apworkbench-studiohub

Marco VENERI · ‎2014-10-15

Posted on October 15, 2014 at 15:53

Hi Michele,

In order to downmix the 4 mic of STEVAL-MKI129Vx board into a mono stream you can use the embedded processing features of the STA321MP onboard device.

Simply set the pre-mixer coefficients for the ''Mix In 4'' and ''Mix In 5'' to the corresponding hex value for -12dB (ie. 2026F3h instead of the default 7FFFFFh value), this for the PDM3/4/5/6 inputs . Refer to the below picture.

I'm here assuming you're recording the mono stream from the HP output which, by default, gets the audio from the Ch4/5 - that's why this is the only volume you can control.

Let me know if this helps.

Marco

baggiomichele · ‎2014-10-21

Posted on October 21, 2014 at 16:03

Hi Marco,

thanks for your fast response.

I followed your instructions but i can make mic 4 and 5 working but not the 3 and 6.

Here's my config file. Could you please review it?

https://dl.dropboxusercontent.com/u/28754179/3456in1.apc.zip

Today i have a recording session and i need to solve:

- record a voice in mono adding mic 3 4 5 6 of the array to get the best SNR.

- record in 24bit.

I'm currently using Audacity but it exports in 16bit.

I've tried Adobe Audition but it gives me an error ''Input and Output sample rate are different. Cannot record''

I've select all the sample rates available on Audition and on APWorkbench without success.

Could you please give me the solution to this two issue i have?

Thanks so much

Really appreciated.

My email is baggiomichele@hotmail.com to contact me directly.

Marco VENERI · ‎2014-11-12

Posted on November 12, 2014 at 12:45

Michele,

If you're going to use the digital upstream via USB (instead of the HP output) then you should replicate the coefficients I suggested in my previous post on the CH1/CH2 pre-mixer controls.

Jsut to be clear: the headphone jack gets the output from CH4/CH5 of STA321MP whereas the digital upstreaming is done from CH1/CH2.

Marco