Page 1 of 1

Obtaining audio signals

Posted: Thu Feb 08, 2018 10:27 am
by antonimaco
Hello there,

I have seen in this forum and in the documentation that you use Audacity to obtain/record the audio signals. After some tests with my linear and circular xcore vocal fusion speaker. Now I am interested in obtaining the audio signals captured by the microphones of xcore device in my pc. In order to make this, I have a C++ program which is able to capture audio signals from an audio card using STK (Synthesis tool kit) library. So my question is: Is it possible to receive the audio signals captured by the 4 microphones in my program as if the xcore device was a audio card??

I have run my program but it doesn't recognize the xcore vocal fusion speaker as an audio device.

Thanks in advance.

Antonio

Re: Obtaining audio signals

Posted: Thu Feb 08, 2018 3:32 pm
by infiniteimprobability
if the xcore device was a audio card
When in USB mode, it *is* a sound card (and a USB Audio compliant one at that), so yes.

Re: Obtaining audio signals

Posted: Thu Feb 08, 2018 3:56 pm
by antonimaco
infiniteimprobability wrote:
if the xcore device was a audio card
When in USB mode, it *is* a sound card (and a USB Audio compliant one at that), so yes.
Thanks for answering. Could you please let me know how can I run the USB mode? I'm confused right now.

Re: Obtaining audio signals

Posted: Thu Feb 08, 2018 4:00 pm
by infiniteimprobability
There are instructions to obtain, build and download the firmware here:
https://www.xmos.com/download/private/s ... rc5%29.pdf

Any of the builds *without* "I2S" in them are USB Audio enabled.

Re: Obtaining audio signals

Posted: Thu Feb 08, 2018 5:58 pm
by antonimaco
infiniteimprobability wrote:There are instructions to obtain, build and download the firmware here:
https://www.xmos.com/download/private/s ... rc5%29.pdf

Any of the builds *without* "I2S" in them are USB Audio enabled.
I'm running "1i2o2_lin33_48khz" build and the STK (Synthesis ToolKit) library is not able to recognise the vocal fusion speaker as an available audio device.

Re: Obtaining audio signals

Posted: Mon Feb 12, 2018 10:35 am
by infiniteimprobability
What happens if you run the 1i4o2 build? This is basically just a sound card with 4 mics and stereo output to the DAC? No DSP of any sort other than decimators to get from PDM mics-> PCM data.

Re: Obtaining audio signals

Posted: Tue Feb 13, 2018 12:14 pm
by antonimaco
I have run 1i4o2 build and the program is not able to recognize the device as a sound card. Maybe I have a problem with the library I'm using (STK Syntheis Toolkit). Could you please recommend me a c or c++ library to capture audio from the vocal fusion speaker??

Thanks in advance.

Antonio

Re: Obtaining audio signals

Posted: Tue Feb 13, 2018 12:22 pm
by infiniteimprobability
Hi,
I am afraid supplying host side libraries is outside of the scope of xmos - we make embedded devices, IP to run on it and supply drivers for USB Audio Class 2. Our VF design running 1i4o2 or similar presents itself as USB Audio Class 1 compliant device and all OS support that in their standard drivers. The next level up is belongs to others in the ecosystem and shouldn't care whether the soundcard is from Xmos or any other vendor. If the soundcard can be seen and audio recorded OK (eg. using Audacity) then our (xmos) work is done..

I know colleagues have successfully used portaudio to capture audio from USB audio devices, if that helps.