hello , i tried to modify this code to add other models ( architectures), now i want that the results display to me the accuracy ( WER word error rate , or the CER caractere error rate ) to be able to evaluate if my architecture is good or not !! thank you very much

When i try to run the command powershell ./flac_to_wav.sh, it opens a git bash but does nothing to the data, and doesn't even show an error, and runs indefinitely if I don't close it,

Hello I am trying to replicate the code and everything is working fine with no error but when I try to run get_predictions function I get an empty predicted transcription any ideas what is happening? bellow is the output:

Layer (type) Output Shape Param #

the_input (InputLayer) (None, None, 13) 0

rnn (GRU) (None, None, 29) 3741

softmax (Activation) (None, None, 29) 0

Total params: 3,741
Trainable params: 3,741
Non-trainable params: 0

None

True transcription:

mister quilter is the apostle of the middle classes and we are glad to welcome his gospel

Predicted transcription: