I use riva_quickstart:2.12.0, I download the “nvidia/riva/diarizer_vad_multilingual_marblenet” version check in the directory is: “diarizer_vad_multilingual_marblenet_vdeployable_v1.0”. Check trt-vad seen in the directory, however when running bash riva_start.sh error when launching model vad, error: “UNAVAILABLE: Internal: unable to create TensorRT engine”. Did I choose the wrong model VAD?
Hardware - GPU (L4)
Hardware - CPU
Operating System
Riva Version: riva_quickstart:2.12.0
TLT Version (if relevant)
How to reproduce the issue ? (This is for errors. Please share the command and the detailed log here)