num_sample=args.num_sample,
normalize_vq=True,
noise_x=True,
noise_y=True).cuda()
dataset_type = 'multi'
elif model_type == 'wavernn':
print("Model type is wavernn")
model_fn = lambda dataset: wr.Model(rnn_dims=896,
fc_dims=896,
pad=2,
upsample_factors=(4, 4, 4),
feat_dims=80).cuda()
dataset_type = 'single'
elif model_type == 'nc':
print("Model type is nc")
model_fn = lambda dataset: nc.Model(rnn_dims=896, fc_dims=896).cuda()
dataset_type = 'single'
else:
sys.exit(f'Unknown model: {model_type}')
if dataset_type == 'multi':
data_path = config.multi_speaker_data_path
with open(f'{data_path}/index.pkl', 'rb') as f:
index = pickle.load(f)
logger.log(f"len of vctk index pkl object is {len(index)}"
) # should be equal to total number of speakers in the dataset
# logger.log(f"index.pkl file --- index[:5] {index[:5]}")
# logger.log(f"index.pkl file --- index[0][:5] {index[0][:5]}")
test_index = [
noise_x=True,
noise_y=True,
DEVICE=DEVICE).to(DEVICE)
dataset_type = 'multi'
elif model_type == 'wavernn':
raise ValueError("NYI wavernn")
model_fn = lambda dataset: wr.Model(rnn_dims=896,
fc_dims=896,
pad=2,
upsample_factors=(4, 4, 4),
feat_dims=80,
DEVICE=DEVICE).to(DEVICE)
dataset_type = 'single'
elif model_type == 'nc':
raise ValueError("NYI nc")
model_fn = lambda dataset: nc.Model(
rnn_dims=896, fc_dims=896, DEVICE=DEVICE).to(DEVICE)
dataset_type = 'single'
else:
sys.exit(f'Unknown model: {model_type}')
if dataset_type == 'multi':
"""
data_path = config.multi_speaker_data_path
data_path = "gt_data_dir"
with open(f'{data_path}/index.pkl', 'rb') as f:
index = pickle.load(f)
test_index = [x[-1:] if i < 2 * args.count else [] for i, x in enumerate(index)]
train_index = [x[:-1] if i < args.count else x for i, x in enumerate(index)]
dataset = env.MultispeakerDataset(train_index, data_path)
"""
elif dataset_type == 'single':