I’m trying very hard to get text-to-speech working locally on my Jetson Orin Nano via Nvidia Riva.
After going through the installation steps and finally running sudo bash riva_start.sh my system gets stuck on: Waiting for Riva server to load all models...retrying in 10 seconds until eventually erroring out.
What’s frustrating is that when I try to check the logs with docker logs riva-speech I get no output, error or otherwise.
I’ve tried this on Jetpack 6.2 and 6.1. Do you think I should try it on 6.0?
I’ve also tried running riva_clean and reinstalling with riva_init.
What do you mean by run from the docker container? I believe that’s what I’m doing. I’ve been following this tutorial.
It utilizes docker, and at the end “docker logs riva-speech” is the recommended command to debug the failure. But it doesn’t provide an output, like I was saying before.
I would like to try this. I do think it’s different from what the quickstart was telling me. Could you explain how? I have not done this before with jetson-containers
docker run --rm -it dustynv/riva-client:python-r36.2.0 bash
This seems to have at least run the docker container successfully though when I ran one of the provided Python test scripts, it couldn’t detect my USB microphone. My microphone does work outside of the Docker container.
