Information Needed to Troubleshoot Issues
| DOWNLOAD | DOCS | MIN SPECS |
Hey,
I’m new to Omniverse. I have been able to get the Audio2Face 2022.2 installed and working and it animates the default mesh face just fine on my hardware. Good so far. However, I wanted to follow the tutorials that use OmniGraph, however, I don’t see the OmniGraph pulldown menu. I thought maybe there was an OmniGraph plugin, but nope. I consulted the OmniGraph documentation and there is nothing about installing OmniGraph. So, I’m perplexed. If the only app I have installed is Audio2Face, how do I get OmniGraph?
Thanks!
Oh, just found this discussion: OmniGraph in Audio2Face is nowhere to be found - Apps / Audio2Face - NVIDIA Developer Forums
So, it seems like perhaps OmniGraph comes from Create, which I suspected, but couldn’t confirm. I’ll just go install that and see if suddenly things makes sense. Assuming this solves my problem, you guys should update the tutorials to say something like this. ;-)
Not sure if I got it from Create, or if I already had it and just needed to open the correct Window. However, I do not have the OmniGraph pulldown menu. Should I?
Hi @daniel.levine OmniGraph is a collection of other graph types, such as ActionGraph or GenericGrpah, etc. Each of these come from a separate extension. If you don’t see them in your UI, you can always load them from Window -> Extensions.
Thanks. It seems like those are turned on for me. I go to Windows->Visual Scripting->General Node (?) and then there’s a tab at the bottom with Console I can switch to and I can click on Edit and then I see the kinds of stuff I expected to see (connected nodes).
Essentially I’m trying to reproduce this demo: Audio2Face Streaming Audio Player - Riva Text to Speech Integration Example | NVIDIA On-Demand
I have a Riva Server running TTS with a female voice as the default using the Riva Quickstart. I’m not sure how to get the list of voices I could use (instead of using the default).
I was able to get the Riva TTS extension installed in the Audio2Face, type in the ip address of my Riva TTS server, and type text to be sent to the Riva TTS server to produce the audio that the Audio2Face would then play and animate the face with. However, in my UI, there’s a voice that’s specified and both the male and female voices produce an error that the specific model is not available on the Riva server. In the above demo, the UI doesn’t specify a voice and so it probably uses the default TTS voice and works. Mine produces and error because it specifies a voice.
I was hoping to use nodes to let me customize these connections a little more and work around these issues, but when I look at the OmniGraph nodes, I don’t see anything feeding into the Streaming Player (I was expecting it to be the output of the Riva TTS call). So I also don’t see how I could easily change the Riva TTS call (or make my own) to not specify a voice using the visual scripting. :-/
Great to see graphs are loaded and working. Regarding Riva, I suggest creating a new thread in the forum so it can receive better view and answers.
Thanks, but eventually, I’ll want to “code” something with OmniGraph. How do I add a node that would make a call to my Riva Server instead of how it’s done with the plug-in?
To receive better answers, I suggest creating a new thread in the proper forum for each separate question, so they get viewed by respective people.
Here’s the page showing all app forums: Latest Omniverse/Apps topics - NVIDIA Developer Forums
Use “New Topic” button in each forum to create a thread.
