def __call__(self, mfcc_file):
start, end = self.likelihoods[1][mfcc_file]
if VERBOSE:
print mfcc_file
print start, end
_, posteriorgrams = viterbi(self.likelihoods[0][start:end],
self.transitions,
self.map_states_to_phones,
using_bigram=self.using_bigram)
assert (not (posteriorgrams == np.NaN).any())
assert (not (self.likelihoods[0][start:end] == np.NaN).any())
self.write_file(mfcc_file, start, end, posteriorgrams)
def __call__(self, mfcc_file):
start, end = self.likelihoods[1][mfcc_file]
if VERBOSE:
print mfcc_file
print start, end
_, posteriorgrams = viterbi(self.likelihoods[0][start:end],
self.transitions,
self.map_states_to_phones,
using_bigram=self.using_bigram)
assert(not (posteriorgrams == np.NaN).any())
assert(not (self.likelihoods[0][start:end] == np.NaN).any())
self.write_file(mfcc_file, start, end, posteriorgrams)
dbn = cPickle.load(idbnf)
with open(sys.argv[4]) as idbndtf:
dbn_to_int_to_state_tuple = cPickle.load(idbndtf)
dbn_phones_to_states = dbn_to_int_to_state_tuple[0]
likelihoods_computer = functools.partial(
reconstruct_articulatory_features_likelihoods, dbn)
# TODO bigrams
transitions = initialize_transitions(transitions)
#print transitions
transitions = penalty_scale(transitions,
insertion_penalty=INSERTION_PENALTY,
scale_factor=SCALE_FACTOR)
dummy = np.ndarray((2, 2)) # to force only 1 compile of Viterbi's C
viterbi(dummy, [None, dummy], {}) # also for this compile's debug purposes
list_of_mfcc_files = []
for d, ds, fs in os.walk(sys.argv[1]):
for fname in fs:
if fname[-4:] != '.mfc':
continue
fullname = d.rstrip('/') + '/' + fname
list_of_mfcc_files.append(fullname)
#print list_of_mfcc_files
if dbn != None:
input_n_frames_mfcc = dbn.rbm_layers[0].n_visible / 39 # TODO generalize
input_n_frames_arti = dbn.rbm_layers[
1].n_visible / 59 # 60 # TODO generalize
print "this is a DBN with", input_n_frames_mfcc, "MFCC frames on the input layer"
from DBN_Gaussian_mocha_timit import DBN
with open(sys.argv[3]) as idbnf:
dbn = cPickle.load(idbnf)
with open(sys.argv[4]) as idbndtf:
dbn_to_int_to_state_tuple = cPickle.load(idbndtf)
dbn_phones_to_states = dbn_to_int_to_state_tuple[0]
likelihoods_computer = functools.partial(reconstruct_articulatory_features_likelihoods, dbn)
# TODO bigrams
transitions = initialize_transitions(transitions)
#print transitions
transitions = penalty_scale(transitions, insertion_penalty=INSERTION_PENALTY,
scale_factor=SCALE_FACTOR)
dummy = np.ndarray((2,2)) # to force only 1 compile of Viterbi's C
viterbi(dummy, [None, dummy], {}) # also for this compile's debug purposes
list_of_mfcc_files = []
for d, ds, fs in os.walk(sys.argv[1]):
for fname in fs:
if fname[-4:] != '.mfc':
continue
fullname = d.rstrip('/') + '/' + fname
list_of_mfcc_files.append(fullname)
#print list_of_mfcc_files
if dbn != None:
input_n_frames_mfcc = dbn.rbm_layers[0].n_visible / 39 # TODO generalize
input_n_frames_arti = dbn.rbm_layers[1].n_visible / 59 # 60 # TODO generalize
print "this is a DBN with", input_n_frames_mfcc, "MFCC frames on the input layer"
print "and", input_n_frames_arti, "articulatory frames on the other input layer"
