Invalid language model path error

Deep Speech
#1

So I’m trying to train the deepspeech using this.

python ../DeepSpeech.py \
  --train_files data/dataset.csv \
  --dev_files data/dataset.csv \
  --test_files data/dataset.csv \
  --alphabet data/alphabet.txt \

but got this error after finish training.

I FINISHED optimization in 0:10:15.965783
[scorer.cpp:77] FATAL: "(access(filename, 4)) == (0)" check failed. Invalid language model path

I’ve check the csv and alphabet.txt file and everythings fine.

Any ideas on how to fix this?

#2

As the error states, your language model path is invalid. We can’t help more, but I see you are using ../DeepSpeech.py. Maybe you don’t want to do that. Please check documentation and --help about --lm_binary_path and --trie_binary_path.