osalinasv / lipnet Goto Github PK

Hi! @osalinasv I would love to use your project, but I cannot clone. I get this error. Can you perhaps advise? Thank you!

I'm getting the following error: ModuleNotFoundError: No module named 'common.files'
I know the folder common is there, but I don't know why it's complaining about. Can't figure it out about this import issue, can you help me?

Traceback (most recent call last): File "train.py", line 154, in <module> main() File "train.py", line 88, in main train(name, config) File "train.py", line 119, in train use_multiprocessing=True File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/legacy/interfaces.py", line 91, in wrapper return func(*args, **kwargs) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/engine/training.py", line 1415, in fit_generator initial_epoch=initial_epoch) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/engine/training_generator.py", line 93, in fit_generator callbacks.set_model(callback_model) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/callbacks.py", line 52, in set_model callback.set_model(model) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/callbacks.py", line 746, in set_model self.sess = K.get_session() File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 184, in get_session _SESSION = tf.Session(config=config) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1494, in __init__ super(Session, self).__init__(target, graph, config=config) File "/root/anaconda3/envs/lip/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 626, in __init__ self._session = tf_session.TF_NewSession(self._graph._c_graph, opts) tensorflow.python.framework.errors_impl.InternalError: Failed to create session.

Hello I am new to this kinda projects and had a small doubt, possibly stupid. So the model you've trained will it be able to classify any person's lip movement irrespective of their accent , race etc? Or do I need to create a dataset of my own and then I need to train a new model? Because I've tried using available models but the predictions are wrong. Would be of great help if you could clearify this for me.

great work. This is not particularly an issue. So, if i want to train it with my own training data, how can i generate the align files for each video that i recorded manually? any sort of reply would be helpful. Thanks

how much computation power require to run this code ?