[Solved] int() argument must be a string, a bytes-like ofject or a number, not ´NoneType´

Deep Speech
#1

Hi! I am starting to use deep speech to train models for Spanish language.

I make some changes like include ‘ñ’ to alphabet.txt, but there is a problem when I’m training the model:

Exception in thread Thread-13:
Traceback (most recent call last):
File “/usr/lib/python3.5/threading.py”, line 914, in _bootstrap_inner
self.run()
File “/usr/lib/python3.5/threading.py”, line 862, in run
self._target(*self._args, **self._kwargs)
File “/home/enigma-admin/DeepSpeech/util/feeding.py”, line 156, in _populate_batch_queue
self._model_feeder.ph_y_length: target_len })
File “/home/enigma-admin/deepspeech_env/lib/python3.5/site-packages/tensorflow/python/client/session.py”, line 905, in run
run_metadata_ptr)
File “/home/enigma-admin/deepspeech_env/lib/python3.5/site-packages/tensorflow/python/client/session.py”, line 1106, in _run
np_val = np.asarray(subfeed_val, dtype=subfeed_dtype)
File “/home/enigma-admin/deepspeech_env/lib/python3.5/site-packages/numpy/core/numeric.py”, line 501, in asarray
return array(a, dtype, copy=False, order=order)
TypeError: int() argument must be a string, a bytes-like object or a number, not ‘NoneType’

What do I need to modify to be able to train a model in Spanish?

Thanks in advance.

#2

I’d guess you have something looking like a number in your dataset and the feeding process automagically triggers some numpy conversion to int(), and that it is unrelated to ñ.

#3

Thanks! I had modified the return value of text_to_char_array() function, so for this reason I get the error.

#4

Hi David, I am interested with to create a spanish model too, how can contact you?