Error when training new model for another language

[EuphoriaCelestial]

Hi, I followed your guide in README and successfully trained an English model from LJS dataset. And I am trying to load my own dataset into it and change symbols list in /text/symbols.py file to train a new language.
The pre-proccess part have no problem, since my dataset have the same structure as LJS. But when I run train.py it give the following error:

`Use FastSpeech
Model Has Been Defined
Number of TTS Parameters: 25367169
Load data to buffer
100%|███████████████████████████████████████████████████████████████████████████| 9559/9559 [00:04<00:00, 1973.04it/s]
cost 4.85s to load all data into buffer.
Defined Optimizer and Loss Function.

---Start New Training---

Traceback (most recent call last):
File "train.py", line 193, in
main(args)
File "train.py", line 106, in main
length_target=duration)
File "/media/hangtg/Data/fastspeeech/venv-fast-speech/lib/python3.6/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/media/hangtg/Data/fastspeeech/venv-fast-speech/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 150, in forward
return self.module(*inputs[0], **kwargs[0])
File "/media/hangtg/Data/fastspeeech/venv-fast-speech/lib/python3.6/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/media/hangtg/Data/fastspeeech/FastSpeech/model.py", line 42, in forward
mel_max_length=mel_max_length)
File "/media/hangtg/Data/fastspeeech/venv-fast-speech/lib/python3.6/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/media/hangtg/Data/fastspeeech/FastSpeech/modules.py", line 82, in forward
output = self.LR(x, target, mel_max_length=mel_max_length)
File "/media/hangtg/Data/fastspeeech/FastSpeech/modules.py", line 72, in LR
output = alignment @ x
RuntimeError: invalid argument 6: wrong matrix size at /pytorch/aten/src/THC/generic/THCTensorMathBlas.cu:499
`

what else I need to change in the original code in order to train a new language?
so far, I just change to basic cleaner and symbols list