# create logger
logger = Logger(LOG_DIR)
kwargs = {'num_workers': 0, 'pin_memory': True} if args.cuda else {}
l2_dist = PairwiseDistance(2)
voxceleb = read_voxceleb_structure(args.dataroot)
if args.makemfb:
#pbar = tqdm(voxceleb)
for datum in voxceleb:
mk_MFB(
(args.dataroot + '/voxceleb1_wav/' + datum['filename'] + '.wav'))
print("Complete convert")
if args.mfb:
transform = transforms.Compose([truncatedinputfromMFB(), totensor()])
transform_T = transforms.Compose([truncatedinputfromMFB(), totensor()])
file_loader = read_MFB
else:
transform = transforms.Compose([
truncatedinput(),
toMFB(),
totensor(),
#tonormal()
])
file_loader = read_audio
voxceleb_dev = [datum for datum in voxceleb if datum['subset'] == 'dev']
train_dir = DeepSpeakerDataset(voxceleb=voxceleb_dev,
dir=args.dataroot,
n_triplets=args.n_triplets,
loss = torch.mean(dist_hinge)
return loss
kwargs = {'num_workers': 0, 'pin_memory': True} if args.cuda else {}
l2_dist = PairwiseDistance(2)
voxceleb = read_voxceleb_structure(args.dataroot)
if args.makemfb:
for datum in voxceleb.iterrows():
mk_MFB((args.dataroot +'/voxceleb1_wav/' + datum[1]['filename']+'.wav'))
if args.mfb:
transform = transforms.Compose([
truncatedinputfromMFB(),
totensor()
])
file_loader = read_MFB
else:
transform = transforms.Compose([
truncatedinput(),
toMFB(),
totensor(),
#tonormal()
])
file_loader = read_audio
voxceleb_dev = voxceleb.loc[lambda voxceleb: voxceleb.subset == 'dev']
elif args.makeif:
print('==> Started IF')
num_features = c.IF_FEATURES
print('==> Started converting wav to npy')
parallel_function(mk_if, [datum['file_path'] for datum in voxceleb_test], num_threads)
print('===> Converting test set is done')
if not args.test_only:
parallel_function(mk_if, [datum['file_path'] for datum in voxceleb_dev], num_threads)
print('===> Converting dev set is done')
print("==> Complete converting")
# Data
transform_train = transforms.Compose([
totensor(permute=False),
truncatedinput(c.NUM_FRAMES),
])
transform_test = transforms.Compose([
totensor(permute=False),
truncatedinput(c.NUM_FRAMES),
])
file_loader = read_npy
train_dir = DeepSpeakerDataset(voxceleb=voxceleb_dev,
dir=args.dataroot,
n_triplets=args.n_triplets,
loader=file_loader,
transform=transform_train)
test_dir = VoxcelebTestset(dir=args.dataroot,
pairs_path=args.test_pairs_path,
loader=file_loader,