Mboshi MMS Q-Former Aya

This checkpoint was trained for Mboshi automatic speech recognition. During training, the MMS speech encoder and Q-Former were trainable, while the Aya decoder remained frozen.

Evaluation

The reported evaluation results are:

Metric Score
Character error rate (CER) 8.48%
Word error rate (WER) 28.4%
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support