RIVA Quick start for Embedded Platforms Failed to run on Jetson

Hardware - Jetson NX 8GB
Riva Version - 2.9.0

I am following the instructions from https://catalog.ngc.nvidia.com/orgs/nvidia/teams/riva/resources/riva_quickstart_arm64

As per the docs jetson device is running at 15W 6 core mode, and the default runtime for docker is set to nvidia.

When I execute bash riva_init.sh command it shows

Logging into NGC docker registry if necessary...
Pulling required docker images if necessary...
Note: This may take some time, depending on the speed of your Internet connection.
> Pulling Riva Speech Server images.
  > Image nvcr.io/nvidia/riva/riva-speech:2.9.0-l4t-aarch64 exists. Skipping.

Downloading models (RMIRs) from NGC...
Note: this may take some time, depending on the speed of your Internet connection.
To skip this process and use existing RMIRs set the location and corresponding flag in config.sh.
Error: No such container: riva-models-download
Error in downloading models.

Here is my config file: config.sh (12.5 KB)

Before this, I also tried Riva Version 2.0.0 Read this thread

Please help me resolve the issue.


Hi @NitinRai

Thanks for your interest in Riva

Apologies you are facing the Issue

I will check with the internal team on this error and get back


HI @NitinRai

I have some updates from Riva Team

your config.sh seems perfect, only simple container start is still failing for the user.

as next steps, please try the below and share the feedback with us

  1. Run docker info and share the output with us (confirm that the default runtime shows as nvidia.)
  2. Start basic container and see if works with nvidia runtime: docker run -it --rm --runtime nvidia ubuntu:20.04
  3. Please check if any stale Riva container is open from previous trials (run docker ps -a if any riva named/related container, please remove using docker container stop [container_id] and docker container rm [container_id]) and remove it before running riva_init script.

let us know the feedback