jbetker commited on
Commit
8effe35
1 Parent(s): 56b10cc

More updates

Browse files
Files changed (1) hide show
  1. do_tts.py +8 -8
do_tts.py CHANGED
@@ -122,14 +122,14 @@ if __name__ == '__main__':
122
  preselected_cond_voices = {
123
  # Male voices
124
  'dotrice': ['voices/dotrice/1.wav', 'voices/dotrice/2.wav'],
125
- 'harris': ['voices/male_harris1.wav', 'voices/male_harris2.wav'],
126
- 'lescault': ['voices/male_lescault1.wav', 'voices/male_lescault2.wav'],
127
- 'otto': ['voices/male_otto1.wav', 'voices/male_otto2.wav'],
128
  # Female voices
129
- 'atkins': ['voices/female_atkins1.wav', 'voices/female_atkins2.wav'],
130
- 'grace': ['voices/female_grace1.wav', 'voices/female_grace2.wav'],
131
- 'kennard': ['voices/female_kennard1.wav', 'voices/female_kennard2.wav'],
132
- 'mol': ['voices/female_mol1.wav', 'voices/female_mol2.wav'],
133
  }
134
 
135
  parser = argparse.ArgumentParser()
@@ -197,7 +197,7 @@ if __name__ == '__main__':
197
  print("Loading DVAE..")
198
  dvae = DiscreteVAE(positional_dims=1, channels=80, hidden_dim=512, num_resnet_blocks=3, codebook_dim=512, num_tokens=8192, num_layers=2,
199
  record_codes=True, kernel_size=3, use_transposed_convs=False).cuda().eval()
200
- dvae.load_state_dict(torch.load('.models/dvae.pth'))
201
  print("Loading Diffusion Model..")
202
  diffusion = DiscreteDiffusionVocoder(model_channels=128, dvae_dim=80, channel_mult=[1, 1, 1.5, 2, 3, 4, 6, 8, 8, 8, 8], num_res_blocks=[1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1],
203
  spectrogram_conditioning_resolutions=[2,512], attention_resolutions=[512,1024], num_heads=4, kernel_size=3, scale_factor=2,
 
122
  preselected_cond_voices = {
123
  # Male voices
124
  'dotrice': ['voices/dotrice/1.wav', 'voices/dotrice/2.wav'],
125
+ 'harris': ['voices/harris/1.wav', 'voices/harris/2.wav'],
126
+ 'lescault': ['voices/lescault/1.wav', 'voices/lescault/2.wav'],
127
+ 'otto': ['voices/otto/1.wav', 'voices/otto/2.wav'],
128
  # Female voices
129
+ 'atkins': ['voices/atkins/1.wav', 'voices/atkins/2.wav'],
130
+ 'grace': ['voices/grace/1.wav', 'voices/grace/2.wav'],
131
+ 'kennard': ['voices/kennard/1.wav', 'voices/kennard/2.wav'],
132
+ 'mol': ['voices/mol/1.wav', 'voices/mol/2.wav'],
133
  }
134
 
135
  parser = argparse.ArgumentParser()
 
197
  print("Loading DVAE..")
198
  dvae = DiscreteVAE(positional_dims=1, channels=80, hidden_dim=512, num_resnet_blocks=3, codebook_dim=512, num_tokens=8192, num_layers=2,
199
  record_codes=True, kernel_size=3, use_transposed_convs=False).cuda().eval()
200
+ dvae.load_state_dict(torch.load('.models/dvae.pth'), strict=False)
201
  print("Loading Diffusion Model..")
202
  diffusion = DiscreteDiffusionVocoder(model_channels=128, dvae_dim=80, channel_mult=[1, 1, 1.5, 2, 3, 4, 6, 8, 8, 8, 8], num_res_blocks=[1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1],
203
  spectrogram_conditioning_resolutions=[2,512], attention_resolutions=[512,1024], num_heads=4, kernel_size=3, scale_factor=2,