Best way to get a baseline WER with the pre-trained model on my own test set


(Dwhettam) #1

I’m wanting to measure the performance of the pre-trained model on my own test set, before training it further with my own data. What’s the best way to go about this?

I’ve seen one post on here where OP has manually tested on each .wav file and averaged the WER. Is there a better way to go about it?