Audio2Face 2022.2.1 hotfix is available now

[RonanDB]

Hi Everyone, This release of Audio2Face 2022.2.1 is focused around improved VGPU support. Installed on a physical GPU in a cloud or enterprise data center server, NVIDIA vGPU software creates virtual GPUs that can be shared across multiple virtual machines, accessed by any device, anywhere.

OM-90168 Fix - Pickup kit-sdk patch 104.1 with vgpu support.

OM-90171 Fix - Visibility toggle state does not update as expected.

OM-90170 Fix - Reset Bind State doesn’t work.

OM-90169 Fix - Riva TTS ext female model name typo.

OM-90166 Fix - A2F/A2E GetSettings – internal Server Error.

[daniel.levine]

I’ve got the Omniverse Launcher installed on a fresh Ubuntu 20.04 x86_64 system with dual RTX 2080Ti in it. I was able to download the 2022.2.1 version using the launcher just fine (as far as I can tell). However, when I launch Audio2Face it thinks, but then nothing comes up. :-/

I had a Windows 10 OS on this same computer using another disk and it was able to run the same just fine, so the hardware is capable of doing Audio2Face in Windows.

I found a log here:
/home//.local/share/ov/pkg/audio2face-2022.2.1/logs/Kit/Audio2Face/2022.2
kit_20230411_193413.log (145.7 KB)

Any idea what the problem could be?

[Ehsan.HM]

@daniel.levine

kit_20230411_193413.log (145.7 KB)

This log seems to be from (2023-04-11) before the latest version was released. Can you please try the latest 2022.2.1 and see if you still have this issue?

[daniel.levine]

Hmmm, I just installed the the OS from scratch today along with Omniverse Launcher and Audio2Face.

In launcher it says I have Audio2Face 2022.2.1

Perhaps I’m looking in the wrong place for the log?

[Ehsan.HM]

hmm, can you delete all the log folder completely and re-open Audio2Face. This should create that folder again.

[daniel.levine]

As you can see I also got Omniverse Code just now as well. It also fails to come up on this system. Perhaps there’s a log from Omniverse Launcher that we need to consult? If so, where is that?

[daniel.levine]

After deleting the logs/ directory, re-launching Audio2Face did not regenerate the logs directory. So perhaps Audio2Face never gets going.

[daniel.levine]

Under ~/.local/share/ov/data/Kit/Audio2Face/2022.2 I have a bunch of crash files like this:
crash_2023-04-19_17-00-22_16441.txt
crash_2023-04-19_17-00-22_16441.txt (6.4 KB)

[Ehsan.HM]

can you try this folder ~/.nvidia-omniverse/logs/Kit/Audio2Face?

[daniel.levine]

I’ll hve to take a look at it tomorrow. But I’ll share what I see there first thing when I get in.

[daniel.levine]

I have that file and this looks more promising! The relevant part of the log file for today’s Audio2Voice test log show this:

023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Networking disabled.
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Device 0:
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Render queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Compute queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Copy queues: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - CUDA device index: 0
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Free BAR1: 244 MB
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Device 1:
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Render queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Compute queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Copy queues: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - CUDA device index: 0
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Free BAR1: 244 MB
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Device 2:
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Render queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Compute queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Copy queues: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - CUDA device index: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Free BAR1: 246 MB
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Device 3:
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Render queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Compute queues: 2
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Copy queues: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - CUDA device index: 1
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin]  - Free BAR1: 246 MB
2023-04-20 13:31:19 [10,842ms] [Info] [gpu.foundation.plugin] Realm command line: realm4kit -ll:zsize 0 -ll:fsize 0 -ll:ib_zsize 0 -ll:gpu_bar1_free 0:255852544,0:255852544,1:258605056,1:258605056 -ll:gpu_ids 0,0,1,1 -ll:csize 0 -cuda:libcuda libcuda.so -ll:cpu 4 -ll:util 0 -cuda:libcudart /home/immersioncentral01/.local/share/ov/pkg/deps/073dc6c2c27a380c2ae0c71cf919e247/extscore/omni.gpu_foundation/bin/deps/libcudart.so -ll:ib_fsize 0 -ll:ht_sharing 0
2023-04-20 13:31:20 [11,240ms] [Warning] [realm:gpu] 4 GPUs requested, but only 2 available!
2023-04-20 13:31:20 [11,801ms] [Error] [carb.crashreporter-breakpad.plugin] crash detected
2023-04-20 13:31:20 [11,802ms] [Warning] [carb.crashreporter-breakpad.plugin] A crash has occurred.  If a debugger should be attached, please set the '/crashreporter/debuggerAttachTimeoutMs' setting to a timeout in milliseconds.  This can be used to allow the crash reporter to wait for up to that long for a debugger to attach before processing or sending the crash report.
2023-04-20 13:31:20 [11,828ms] [Warning] [carb.crashreporter-breakpad.plugin] compressing 2 files into '/home/immersioncentral01/.local/share/ov/data/Kit/Audio2Face/2022.2/7d0a391c-a9d5-4d65-e006c485-09c51e28.dmp.zip' for upload.
2023-04-20 13:31:21 [11,937ms] [Warning] [carb.crashreporter-breakpad.plugin] Uploading minidump: file:'/home/immersioncentral01/.local/share/ov/data/Kit/Audio2Face/2022.2/7d0a391c-a9d5-4d65-e006c485-09c51e28.dmp.zip' svr:'https://services.nvidia.com/submit'

There are 2 physical RTX cards in this system.

[daniel.levine]

This error led me to this issue: NVIDIA driver 440.44 / Vulkan vulkaninfo reports too many GPUs

I also had files in both places, so I moved the recommended one out of the way. After this it came up just fine! Thanks for telling me where the log files are so I could help myself!