I’m trying to run Isaac SDK via Windows 10 WSL2 functionality since I don’t have spare Linux PC with Nvidia GPU for my fun. Beyond expectations that seems to work in general. I have installed SDK manager, Isaac SDK and all dependencies, GPU is successfully recognized and utilized by sample applications, applications compilation on host PC and target Jetson Nano is ok as well. The only issue I have now is no audio device on host. And because I want to build voice recognition application for Jetson Nano it is blocking issue for me :) Sound is not directly supported in WSL so I’m trying to use PulseAudio (server on Windows and client in WSL). It works in general, via standard PulseAudio/ALSA utils like arecord/aplay/parecord/paplay I can record and play audio without problem despite device list parameter output shows no soundcard. But when running Isaac sample audio apps I got error like
“2020-07-23 10:34:45.598 ERROR packages/audio/AudioCapture.cpp@45: No capture devices found”. The rest of compilation goes OK and app is started, by the way.
I’m pretty new to all this stuff and need some advice. Maybe somebody already have experience with similar task?
Or at least where to dig, how Isaac SDK is accessing sound hardware, which engine is used, etc? Is any way to befriend it with PulseAudio?