This model is mbart-large-50-many-to-many-mmt model fine-tuned on the text part of PortMEDIA-Lang spoken language understanding dataset.

The scores on the test set are 40.76% and 43.93% for CER and CVER respectively.

Downloads last month
7
Safetensors
Model size
0.6B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support