I am trying synthesize wave from some predicted mel(.npy) file i got from tts. I follow the read file that synthesis.py may not work (in factlly I also get error when using this ) So I use the evalutae.py. But errors occur:
~/AI_/wavenet_vocoder# python evaluate.py --data-root=./output_mel/ ./wavenet_premodel/20180510_mixture_lj_checkpoint_step000320000_ema.pth generated
/root/anaconda3/envs/keras/lib/python3.6/site-packages/sklearn/utils/deprecation.py:143: FutureWarning: The sklearn.preprocessing.data module is deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.preprocessing. Anything that cannot be imported from sklearn.preprocessing is now part of the private API.
warnings.warn(message, FutureWarning)
Using TensorFlow backend.
Command line args:
{'--data-root': './output_mel/',
'--file-name-suffix': '',
'--help': False,
'--hparams': '',
'--initial-value': None,
'--length': '32000',
'--num-utterances': '-1',
'--output-html': False,
'--speaker-id': None,
'': './wavenet_premodel/20180510_mixture_lj_checkpoint_step000320000_ema.pth',
'': 'generated'}
Traceback (most recent call last):
File "evaluate.py", line 70, in
test_data_loader = get_data_loaders(data_root, speaker_id, testshuffle=False)["test"]
File "/root/AI/wavenet_vocoder/train.py", line 861, in get_data_loaders
random_state=hparams.random_state))
File "/root/anaconda3/envs/keras/lib/python3.6/site-packages/nnmnkwii/datasets/init.py", line 111, in init
collected_files = self.file_data_source.collectfiles()
File "/root/AI/wavenet_vocoder/train.py", line 127, in collect_files
with open(meta, "rb") as f:
FileNotFoundError: [Errno 2] No such file or directory: './output_mel/train.txt'
Why it needs train.txt? I mean 'wavenet_vocoder ' should generate the wav if I just give the mel files input?
If you see this isues please help me about this ,Thank you.
I am trying synthesize wave from some predicted mel(.npy) file i got from tts. I follow the read file that synthesis.py may not work (in factlly I also get error when using this ) So I use the evalutae.py. But errors occur:
~/AI_/wavenet_vocoder# python evaluate.py --data-root=./output_mel/ ./wavenet_premodel/20180510_mixture_lj_checkpoint_step000320000_ema.pth generated /root/anaconda3/envs/keras/lib/python3.6/site-packages/sklearn/utils/deprecation.py:143: FutureWarning: The sklearn.preprocessing.data module is deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.preprocessing. Anything that cannot be imported from sklearn.preprocessing is now part of the private API. warnings.warn(message, FutureWarning) Using TensorFlow backend. Command line args: {'--data-root': './output_mel/', '--file-name-suffix': '', '--help': False, '--hparams': '', '--initial-value': None, '--length': '32000', '--num-utterances': '-1', '--output-html': False, '--speaker-id': None, '': './wavenet_premodel/20180510_mixture_lj_checkpoint_step000320000_ema.pth',
'': 'generated'}
Traceback (most recent call last):
File "evaluate.py", line 70, in
test_data_loader = get_data_loaders(data_root, speaker_id, testshuffle=False)["test"]
File "/root/AI/wavenet_vocoder/train.py", line 861, in get_data_loaders
random_state=hparams.random_state))
File "/root/anaconda3/envs/keras/lib/python3.6/site-packages/nnmnkwii/datasets/init.py", line 111, in init
collected_files = self.file_data_source.collectfiles()
File "/root/AI/wavenet_vocoder/train.py", line 127, in collect_files
with open(meta, "rb") as f:
FileNotFoundError: [Errno 2] No such file or directory: './output_mel/train.txt'
Why it needs train.txt? I mean 'wavenet_vocoder ' should generate the wav if I just give the mel files input?
If you see this isues please help me about this ,Thank you.