def main(plda, enroll_xvector, test_xvector, trials, scores):
with kaldi_io.open_or_fd(plda, 'rb') as f:
for key, vec in kaldi_io.read_vec_flt_ark(f):
if key == 'mean':
mean = vec.reshape(-1, 1)
dim = vec.shape[0]
elif key == 'within_var':
within_var = vec.reshape(dim, dim)
else:
between_var = vec.reshape(dim, dim)
within_var = within_var + 5e-5 * np.eye(within_var.shape[0])
Gamma, Lambda, c, k = CalculateVar(between_var, within_var, mean)
f_writer = open(scores, 'w')
enrollutt2vector = {}
for key, vector in kaldi_io.read_vec(enroll_xvector):
enrollutt2vector[key] = vector
testutt2vector = {}
for key, vector in kaldi_io.read_vec(test_xvector):
testutt2vector[key] = vector
with open(trials, 'r') as f:
for line in f:
enroll, test, _ = line.strip().split()
score = PLDAScoring(enrollutt2vector[enroll].reshape(-1,1),testutt2vector[test].reshape(-1,1),\
Gamma,Lambda,c,k)
f_writer.write(enroll + ' ' + test + ' ' + str(score) + '\n')
f_writer.close()
def plda_write(self, plda):
with kaldi_io.open_or_fd(plda, 'wb') as f:
kaldi_io.write_vec_flt(f, self.mean, key='mean')
kaldi_io.write_vec_flt(f,
self.within_var.reshape(-1, 1),
key='within_var')
kaldi_io.write_vec_flt(f,
self.between_var.reshape(-1, 1),
key='between_var')
def plda_read(self, plda):
with kaldi_io.open_or_fd(plda, 'rb') as f:
for key, vec in kaldi_io.read_vec_flt_ark(f):
if key == 'mean':
self.mean = vec.reshape(-1, 1)
self.dim = self.mean.shape[0]
elif key == 'within_var':
self.within_var = vec.reshape(self.dim, self.dim)
else:
self.between_var = vec.reshape(self.dim, self.dim)
def plda_read(self,plda):
with kaldi_io.open_or_fd(plda,'rb') as f:
for key,vec in kaldi_io.read_vec_flt_ark(f):
if key == 'mean':
mean = vec.reshape(-1,1)
dim = mean.shape[0]
elif key == 'within_var':
within_var = vec.reshape(dim, dim)
else:
between_var = vec.reshape(dim, dim)
return mean,between_var,within_var
model_blueprint, model_creation = utils.read_nnet_config(args.nnet_config)
elif args.model_blueprint is not None and args.model_creation is not None:
model_blueprint = args.model_blueprint
model_creation = args.model_creation
else:
raise ValueError("Expected nnet_config or (model_blueprint, model_creation) to exist.")
model = utils.create_model_from_py(model_blueprint, model_creation)
model.load_state_dict(torch.load(args.model_path, map_location='cpu'), strict=False)
# Select device
model = utils.select_model_device(model, args.use_gpu, gpu_id=args.gpu_id)
model.eval()
with kaldi_io.open_or_fd(args.feats_rspecifier, "rb") as r, \
kaldi_io.open_or_fd(args.vectors_wspecifier, 'wb') as w:
for line in r:
# (key, rxfile, chunk_start, chunk_end) = line.decode().split(' ')
# chunk=[chunk_start, chunk_end]
# print("Process utterance for key {0}".format(key))
# feats = kaldi_io.read_mat(rxfile, chunk=chunk)
(key, rxfile) = line.decode().split(' ')
print("Process utterance for key {0}".format(key))
feats = kaldi_io.read_mat(rxfile)
embedding = model.extract_embedding(feats)
kaldi_io.write_vec_flt(w, embedding.numpy(), key=key)
# while(True):
# key = kaldi_io.read_key(r)
# if not key: