Edit model card

First Automatic Speech Recognition Model for Luxembourgish

Training pipeline based on the tutorials:

Trained on custom data set (~8 hours of Luxembourgish audio+transcript data)

This a first shot and purely experimental. The inferences are thus based only on the trained acoustic model. No language model has been used yet.

WER: 0.18.

Downloads last month
86
Hosted inference API
or or
This model can be loaded on the Inference API on-demand.