voxceleb_list = "Data/voxceleb.npy"
voxceleb_dev_list = "Data/voxceleb_dev.npy"
voxceleb_dev_10k_list = "Data/voxceleb_dev_10k.npy"
if os.path.isfile(voxceleb_list):
voxceleb = np.load(voxceleb_list, allow_pickle=True)
else:
voxceleb = read_my_voxceleb_structure(args.dataroot)
np.save(voxceleb_list, voxceleb)
# Make fbank feature if not yet.
if args.makemfb:
#pbar = tqdm(voxceleb)
for datum in voxceleb:
# print(datum['filename'])
mk_MFB((args.dataroot + '/' + datum['filename'] + '.wav'))
print("Complete convert")
# Create file loader for dataset
if args.mfb:
transform = transforms.Compose([truncatedinputfromMFB(), totensor()])
transform_T = transforms.Compose([
truncatedinputfromMFB(input_per_file=args.test_input_per_file),
totensor()
])
file_loader = read_MFB
else:
transform = transforms.Compose([
truncatedinput(),
toMFB(),
totensor(),
# create logger
logger = Logger(LOG_DIR)
# Define visulaize SummaryWriter instance
writer = SummaryWriter('Log/amsoftmax_res10', comment='margin0.3')
kwargs = {'num_workers': 0, 'pin_memory': True} if args.cuda else {}
if args.cos_sim:
l2_dist = nn.CosineSimilarity(dim=1, eps=1e-6)
else:
l2_dist = PairwiseDistance(2)
voxceleb, voxceleb_dev = wav_list_reader(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([
concateinputfromMFB(),
# truncatedinputfromMFB(),
totensor()
])
transform_T = transforms.Compose([
concateinputfromMFB(input_per_file=args.test_input_per_file),
# truncatedinputfromMFB(input_per_file=args.test_input_per_file),
totensor()
])
file_loader = read_MFB
else:
os.makedirs(EXT_DIR)
# create logger
logger = Logger(LOG_DIR)
kwargs = {'num_workers': 0, 'pin_memory': True} if args.cuda else {}
l2_dist = PairwiseDistance(2)
audio_set = []
audio_set = if_load_npy(dataroot, data_set_list)
if args.makemfb:
#pbar = tqdm(voxceleb)
for datum in audio_set:
# print(datum['filename'])
mk_MFB((datum['filename']+'.wav'))
print("Complete convert")
if args.mfb:
transform = transforms.Compose([
concateinputfromMFB(),
to4tensor()
# truncatedinputfromMFB(),
# totensor()
])
transform_T = transforms.Compose([
truncatedinputfromMFB(input_per_file=args.test_input_per_file),
totensor()
])
file_loader = read_MFB
else: