synthesis.py Failed to load checkpoint at logs-Tacotron-2/wave_pretrained/

[DonggeunYu]

After learning Tacotron-2 up to 10000 steps, I tried text to speech.

How to train the wave_pretrained?

python synthesize.py

Error:

synthesized mel spectrograms at tacotron_output/eval
Synthesizing audio from mel-spectrograms.. (This may take a while)
Traceback (most recent call last):
  File "/home/modeep/Documents/GitHub/Text2Speech/wavenet_vocoder/synthesize.py", line 73, in wavenet_synthesize
    checkpoint_path = tf.train.get_checkpoint_state(checkpoint).model_checkpoint_path
AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "synthesize.py", line 101, in <module>
    main()
  File "synthesize.py", line 95, in main
    synthesize(args, hparams, taco_checkpoint, wave_checkpoint, sentences)
  File "synthesize.py", line 42, in synthesize
    wavenet_synthesize(args, hparams, wave_checkpoint)
  File "/home/modeep/Documents/GitHub/Text2Speech/wavenet_vocoder/synthesize.py", line 76, in wavenet_synthesize
    raise RuntimeError('Failed to load checkpoint at {}'.format(checkpoint))
RuntimeError: Failed to load checkpoint at logs-Tacotron-2/wave_pretrained/

[wan-wei]

Hi @Yudonggeun , if you only want to test the performance of your Tacotron-2 model (not together with WaveNet model, right?), you may run python synthesize.py --model=Tacotron. Furthermore, if you want to test your own sentences, you may run python synthesize.py --mode=live, and it will ignore WaveNet model when the mode is live. Hope that this information is useful to you :)

[DonggeunYu]

Thank you for you!! I fixed it.