PyTorch Implementation of ByteDance's Cross-speaker Emotion Transfer Based on Speaker Condition Layer Normalization and Semi-Supervised Training in Text-To-Speech
MIT License
181
stars
26
forks
source link
'NoneType' object has no attribute 'unsqueeze' because mask is None in evaluating. #2
When in evaluating,'self.training' is False,so mel_masks is None。Decoder encounters this error because mel_mask is None.