Fine Tune the hind Nvidia Nemo

I already covered this Error in a post.
The “riva-trt-conformer-en-US-asr-streaming-am-streaming” is broken.
Ryan Leary is aware of the issue. Something breaks the Triton all together. Cant get ASR working and going from NeMo to RIVA-Triton is not exactly a cake walk.

Thanks @rvinobha

kindly share the notebook.

Hi @Pappachuck_renan, @rvinobha

Do we have some solution for this?? @Pappachuck_renan I cannot locate the post can you please share the link?

We mostly figured it out on discord.
No theres no solution. Its something wrong with Tensor RT probably. Riva script is complex and there is no documentation.
Getting a models from NeMo and deploying into Tensor RT can get to level of complexity as high as messing with CUDA kernels.
I reccomend we open a Slack channel or talk to Edmar @edmar1 or Wendy @WendyGram to open a thread or discord so we can go over this.
We can go step by step until we hit the issue and explore solutions.

Hi @Pappachuck_renan

Can you share the discord/Slack channel link?

Thanks,
Garima

Usually the TTS stuff we use the Audio2Face channel.
You can find CUDA specific stuff on Warp. Lots of knowledge spread there.