How to tune alpha (lm_weight) and beta(word_count_weight) for our custom training set?


(Deepak Banka) #1

Hi, I have trained using custom dataset of 1000 hours. I wanted to know how are the values of lm_weight, word_count_weight and valid_word_count_weight decided/optimized?


(Lissyx) #2

I tend to recall that @reuben did that, he might be able to weight some knowledges :slight_smile:


(Reuben Morais) #3

They are decided/optimized by doing a search over the possible values and choosing the value that yields the best WER on the validation set.