After training, using some script like this: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/tools/freeze_graph.py ?

What should be the input node name(s) and output node names used for freeze and inference later?

Thanks!

Due to the 2GB contiguous block allocation limitation of 32 bit systems, this line:

https://github.com/readbeyond/aeneas/blob/master/aeneas/audiofile.py#L447

might throw a MemoryError, causing an "unexpected exception".

We should:

1. catch it and log a meaningful error
2. document this limitation somewhere

I'd like to train this model on 8 V100 GPUs - does it support multi GPU training?

After testing this code, I found out that following dependencies must be updated otherwise Exception occurs :

Pytorch > 1.0 # F.one_hot function isn't available in PyTorch v1.0 it requires 1.1 or 1.2
librosa==0.7.0 # for Griffin-Lim -> librosa.feature.inverse introduced in librosa 0.7.0

Also for running soundfile following OS dependency required

sudo apt-get install libsndfile1

It looks like the default initialization scheme with tf.get_variable is uniform Glorot , but the paper uses He's Gaussian.