NewEricWang

where is directory ./models/?

1

Hi, Yu: I can't find the directory "./models/" mentioned in your README file.

bad voice when setting tacotron_dropout_rate=0.0 in synthesis

4

Hi, I use the master at 2019-03-26. I set tacotron_dropout_rate=0.5 in hparams.py when training. I can get good voice if I set tacotron_dropout_rate=0.5 in hparams.pywhen synthesizing, but different quality and...

ASR perfomance of enhanced speech data

1

Hi, Do anybody try to use the enhanced speech data for ASR? Can you get performance improvement using the enhancement method? I failed to get good results.

Hi, I use my owner data to train teacher model and student model. The teacher model is normal; but the KL loss of student model becomes nan at about 50k...

Why is teacher model used when inference?

1

Hi, Why is teacher model loaded when inference? Is it necessary? How to disable it?

Hi, Do anybody try to use the enhanced speech data for ASR? Can you get performance improvement using the enhancement method? I failed to get good results.

Hi, I have tried to train Stacked Time Distributed 2D CNN-LSTM model on RAVDESS database by using default hyperparameters. But I can't reproduce the result(acc=94%) reported by you. Could you...