New Language Training
Hi, your work is fascinating, especially in terms of response time. I’d like to know if I can train this model on a new language?
+1
Followed finetuning tutorial but I get that error on Colab:
numba.cuda.cudadrv.driver.LinkerError: [222] Call to cuLinkAddData results in CUDA_ERROR_UNSUPPORTED_PTX_VERSION ptxas application ptx input, line 9; fatal : Unsupported .version 8.5; current version is '8.4
Name: numba
Version: 0.61.0
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Thu_Jun__6_02:18:23_PDT_2024
Cuda compilation tools, release 12.5, V12.5.82
Build cuda_12.5.r12.5/compiler.34385749_0
If anyone is thinking of training/finetuning this model, it would be very helpful if you could share your notebook.
Yes either you can finetune: https://docs.nvidia.com/nemo-framework/user-guide/latest/nemotoolkit/asr/configs.html#fine-tuning-configurations
or train from scratch: https://docs.nvidia.com/nemo-framework/user-guide/latest/nemotoolkit/asr/examples/kinyarwanda_asr.html
It’s work.
Just using download dataset script and run fine tuning script