
If(mmcssHandle = NULL) MessageBox(NULL,TEXT("Unable to enable MMCSS on capture thread!"), TEXT("Error"), MB_OK) ĭWORD waitResult = WaitForMultipleObjects(2, waitArray, FALSE, INFINITE) MmcssHandle = AvSetMmThreadCharacteristics("Audio", &mmcssTaskIndex) If (FAILED(hr)) MessageBox(NULL,TEXT("Unable to initialize COM in render thread!"), TEXT("Error"), MB_OK) Hr = CoInitializeEx(NULL, COINIT_MULTITHREADED)

I can modify the WAVEFORMATEX and change the driver format but I still only get different kind of loud static noises or no sound at all.If I use Audacity to capture, it works, so the problem must be at my code. I'd be happy for any further insights to this problem.If I open the captured PCM data with a Hex eitor they are only combinations of 00 and FF bytes.It seems to capture only static noise on my laptop and pc.On my pc, the recording device (Stereo Mix - Realtek High Definiton Audio) and on my laptop I use the recording device Stereo Mix (Realtek Audio). Also all settings of Soundquality in RDP have no effect. We set up a testversion of TS2022, the problem persists. Using Windows RDP Clients or connecting macos Clients to a TS2012 remedies the distorsions.

Setting the VM to latency sensitive seems to have minimal effect.

We tried different Bandwith Settings (22 - 48 kHz, Mono/Stereo) in macos and audacity to no avail. I cannot tell our RIS to use WASAPI though. As long as I use Directplay or MME as Audio host recordings gets distorted, as long as I use WASAPI it's fine.

I installed Audacity to look further into this. This happens with all macos and IOS clients I could get my hands on, running Monterey to ventura and Intel and Apple silicone as well as year old Microsoft remote desktop clients to the newest beta. The sound quality on TS2019 is really bad, voice sounds too fast with a lot of skips in between. We use Macs as PACS Clients to connect to a Terminal Server 2019 that is run on a VMWare vSphere server.Īfter switching from TS2012 to TS2019 we cannot use the speech recognition of our RIS anymore.
