Yunchao He

Yes, It takes a longer time about 220 seconds per epoch.

Yes, I changed the value into 4.

I forgot to increase the value of `batch_size` by multiplying with `num_gpus` when training.

@veqtor Thanks for your sharing. Do you know if there is a TensorFlow-version griffin lim?

@Kyubyong The dimension is updated. However, this change leads to a new issue in [highway net](https://github.com/Kyubyong/tacotron/blob/master/modules.py#L278): ``` outputs = H * T + inputs * C ``` The dimension of...

@AzamRabiee How is your synthesized waves like at epoch 40? Could you share some of them?

More choice is always better. I will try it later. Thanks for your sharing.

@ggsonic Nice work! Looking forward to your code modification.

@ggsonic In the latest [pull request](https://github.com/Kyubyong/tacotron/pull/43/files), I have added this feature. Could you please help check whether it is right?

中文example没有提供。