How Do I Get The VocalFusion Microphone Inputs Into My Development PC When Using I2S ?

Voice related projects and technical discussions
User avatar
johned
XCore Addict
Posts: 158
Joined: Tue Mar 26, 2013 12:10 pm
Contact:

How Do I Get The VocalFusion Microphone Inputs Into My Development PC When Using I2S ?

Postby johned » Tue Oct 02, 2018 12:02 pm

When I'm tuning a VocalFusion product that uses I2S to communicate to an Applications Processor, how do i get the vocalfusion microphone inputs into my development pc ?
User avatar
johned
XCore Addict
Posts: 158
Joined: Tue Mar 26, 2013 12:10 pm
Contact:

Postby johned » Tue Oct 02, 2018 12:03 pm

You can use xSCOPE.

Edit sw_vocalfusion/app_vf_stereo_base/src/extensions/config.xscope as follows :

Code: Select all

<!-- Probes for channel activity debug -->
<xSCOPEconfig ioMode="basic" enabled="true">
  <Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_0"/>
  <Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_1"/>
  <Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_2"/>
  <Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_3"/>
</xSCOPEconfig>


For VocalFusion Stereo, add the following lines to sw_vocalfusion/src/beclear_int.xc, at line 421 (Just after "inputNear[i] = attenuator(samp, *ptr_mul);") :

Code: Select all

               xscope_int(i, inputNear[i]);


For VocalFusion Mono, add the following lines to sw_vocalfusion/src/beclear_int.xc, at line 276 (Just after "inputNear[i] = sampsFromAudioToUsb[PDM_MIC_INDEX+i] << MIC_EXTRA_GAIN_SHIFT;") :

Code: Select all

               xscope_int(i, inputNear[i]);


Now run the application, ensuring that xSCOPE "Real-Time Mode" is enabled in the Run Configuration.

You will see the microphone input signals plotted in the xSCOPE graph.

To record the xscope audio streams to a .wav file, please see xvu here : https://github.com/ed-xmos/xvu.
tfwfc
Junior Member
Posts: 4
Joined: Thu Mar 17, 2016 12:31 pm

Postby tfwfc » Mon Oct 22, 2018 1:04 pm

If you're using VocalFusion Speaker with xTIMEcomposer here's how to enable xSCOPE output of the microphone signals

Copy app_smarthome_init/config.xscope to app_vf_spk_base/src/extensions/config.xscope:
Overwrite the code with the lines below. (Note - Channel naming means xSCOPE signal names will match silkscreen marking on lin33 board)
 

Code: Select all

<!-- Probes for channel activity debug -->
<xSCOPEconfig ioMode="basic" enabled="true">
<Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_0"/>
<Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_3"/>
<Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_1"/>
<Probe type="CONTINUOUS" datatype="INT" units="-" enabled="true" name="Ch Mic_2"/>
</xSCOPEconfig>


Add the following line to module_vocalfusion/src/beclear_int.xc, at line 277 (Just after "inputNear[i] = sampsFromAudioToUsb[PDM_MIC_INDEX+i] << MIC_EXTRA_GAIN_SHIFT;") :

Code: Select all

xscope_int(i, inputNear[i]);

 
Ensure you have -DXSCOPE=1 and -fxscope in the build configuration
Build your application
Run the application, ensuring that xSCOPE "Real-Time Mode" is enabled in the Run Configuration.
You will see the microphone input signals plotted in the xSCOPE graph.

Who is online

Users browsing this forum: No registered users and 29 guests