novoapi_functions.py is adjusted to use convert_phoneset.py.
This commit is contained in:
parent
b444b70af9
commit
2004399179
Binary file not shown.
@ -14,18 +14,20 @@ def multi_character_tokenize(line, multi_character_tokens):
|
|||||||
line = line[1:]
|
line = line[1:]
|
||||||
|
|
||||||
|
|
||||||
def split_word(word, multi_character_phones):
|
def split_word(word, phoneset):
|
||||||
"""
|
"""
|
||||||
split a line by given phoneset.
|
split a line by given phoneset.
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
word (str): a word written in given phoneset.
|
word (str): a word written in given phoneset.
|
||||||
multi_character_phones (list): the list of multicharacter phones which is considered as one phone. this can be obtained with phoneset definition such as fame_ipa.py.
|
#multi_character_phones (list): the list of multicharacter phones which is considered as one phone. this can be obtained with phoneset definition such as fame_ipa.py.
|
||||||
|
phoneset (list): the list of phones.
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
(word_seperated) (list): the word splitted in given phoneset.
|
(word_seperated) (list): the word splitted in given phoneset.
|
||||||
|
|
||||||
"""
|
"""
|
||||||
|
multi_character_phones = extract_multi_character_phones(phoneset)
|
||||||
return [phone
|
return [phone
|
||||||
for phone in multi_character_tokenize(word.strip(), multi_character_phones)
|
for phone in multi_character_tokenize(word.strip(), multi_character_phones)
|
||||||
]
|
]
|
||||||
@ -44,3 +46,13 @@ def phone_reduction(phones, reduction_key):
|
|||||||
multi_character_tokenize(wo.strip(), multi_character_phones)
|
multi_character_tokenize(wo.strip(), multi_character_phones)
|
||||||
return [reduction_key.get(i, i) for i in phones
|
return [reduction_key.get(i, i) for i in phones
|
||||||
if not i in phones_to_be_removed]
|
if not i in phones_to_be_removed]
|
||||||
|
|
||||||
|
|
||||||
|
def extract_multi_character_phones(phoneset):
|
||||||
|
"""
|
||||||
|
Args:
|
||||||
|
phoneset (list):
|
||||||
|
"""
|
||||||
|
multi_character_phones = [i for i in phoneset if len(i) > 1]
|
||||||
|
multi_character_phones.sort(key=len, reverse=True)
|
||||||
|
return multi_character_phones
|
@ -352,6 +352,9 @@ def fix_lexicon(lexicon_file):
|
|||||||
return
|
return
|
||||||
|
|
||||||
|
|
||||||
|
#def add_sp_to_lexicon(lexicon_file):
|
||||||
|
|
||||||
|
|
||||||
def word2htk(word):
|
def word2htk(word):
|
||||||
return ''.join([fame_asr.translation_key_word2htk.get(i, i) for i in word])
|
return ''.join([fame_asr.translation_key_word2htk.get(i, i) for i in word])
|
||||||
|
|
||||||
|
@ -16,50 +16,53 @@ import defaultfiles as default
|
|||||||
sys.path.append(default.toolbox_dir)
|
sys.path.append(default.toolbox_dir)
|
||||||
import file_handling as fh
|
import file_handling as fh
|
||||||
from htk import pyhtk
|
from htk import pyhtk
|
||||||
|
#from scripts import run_command
|
||||||
|
|
||||||
|
|
||||||
## ======================= user define =======================
|
## ======================= user define =======================
|
||||||
# procedure
|
# procedure
|
||||||
|
combine_all = 1
|
||||||
|
|
||||||
make_lexicon = 0
|
make_lexicon = 0
|
||||||
make_label = 0 # it takes roughly 4800 sec on Surface pro 2.
|
make_label = 0 # it takes roughly 4800 sec on Surface pro 2.
|
||||||
make_mlf = 0
|
make_mlf = 0
|
||||||
extract_features = 0
|
extract_features = 0
|
||||||
flat_start = 0
|
flat_start = 1
|
||||||
train_monophone_without_sp = 0
|
train_monophone_without_sp = 1
|
||||||
add_sp = 0
|
add_sp = 1
|
||||||
train_monophone_with_re_aligned_mlf = 0
|
train_monophone_with_re_aligned_mlf = 1
|
||||||
|
increase_mixture = 1
|
||||||
train_triphone = 0
|
train_triphone = 0
|
||||||
train_triphone_tied = 1
|
train_triphone_tied = 0
|
||||||
|
|
||||||
|
|
||||||
# pre-defined values.
|
# pre-defined values.
|
||||||
dataset_list = ['devel', 'test', 'train']
|
dataset_list = ['devel', 'test', 'train']
|
||||||
feature_size = 39
|
feature_size = 30
|
||||||
improvement_threshold = 0.3
|
improvement_threshold = 0.3
|
||||||
|
|
||||||
hmmdefs_name = 'hmmdefs'
|
|
||||||
proto_name = 'proto'
|
|
||||||
|
|
||||||
lexicon_asr = os.path.join(default.fame_dir, 'lexicon', 'lex.asr')
|
lexicon_asr = os.path.join(default.fame_dir, 'lexicon', 'lex.asr')
|
||||||
lexicon_oov = os.path.join(default.fame_dir, 'lexicon', 'lex.oov')
|
lexicon_oov = os.path.join(default.fame_dir, 'lexicon', 'lex.oov')
|
||||||
|
|
||||||
config_dir = os.path.join(default.htk_dir, 'config')
|
config_dir = os.path.join(default.htk_dir, 'config')
|
||||||
phonelist_full_txt = os.path.join(config_dir, 'phonelist_full.txt')
|
phonelist_full_txt = os.path.join(config_dir, 'phonelist_full.txt')
|
||||||
tree_hed = os.path.join(config_dir, 'tree.hed')
|
tree_hed = os.path.join(config_dir, 'tree.hed')
|
||||||
quest_hed = os.path.join(config_dir, 'quests.hed')
|
quests_hed = os.path.join(config_dir, 'quests.hed')
|
||||||
|
|
||||||
model_dir = os.path.join(default.htk_dir, 'model')
|
model_dir = os.path.join(default.htk_dir, 'model')
|
||||||
model_mono0_dir = os.path.join(model_dir, 'mono0')
|
model_mono0_dir = os.path.join(model_dir, 'mono0')
|
||||||
model_mono1_dir = os.path.join(model_dir, 'mono1')
|
model_mono1_dir = os.path.join(model_dir, 'mono1')
|
||||||
model_mono1sp_dir = os.path.join(model_dir, 'mono1sp')
|
model_mono1sp_dir = os.path.join(model_dir, 'mono1sp')
|
||||||
model_mono1sp2_dir = os.path.join(model_dir, 'mono1sp2')
|
model_mono1sp2_dir = os.path.join(model_dir, 'mono1sp2')
|
||||||
model_tri1_dir = os.path.join(model_dir, 'tri1')
|
model_tri1_dir = os.path.join(model_dir, 'tri1')
|
||||||
|
model_tri1tied_dir = os.path.join(model_dir, 'tri1tied')
|
||||||
|
|
||||||
# directories / files to be made.
|
# directories / files to be made.
|
||||||
lexicon_dir = os.path.join(default.htk_dir, 'lexicon')
|
lexicon_dir = os.path.join(default.htk_dir, 'lexicon')
|
||||||
lexicon_htk_asr = os.path.join(lexicon_dir, 'lex.htk_asr')
|
lexicon_htk_asr = os.path.join(lexicon_dir, 'lex.htk_asr')
|
||||||
lexicon_htk_oov = os.path.join(lexicon_dir, 'lex.htk_oov')
|
lexicon_htk_oov = os.path.join(lexicon_dir, 'lex.htk_oov')
|
||||||
lexicon_htk = os.path.join(lexicon_dir, 'lex.htk')
|
lexicon_htk = os.path.join(lexicon_dir, 'lex.htk')
|
||||||
|
lexicon_htk_with_sp = os.path.join(lexicon_dir, 'lex_with_sp.htk')
|
||||||
lexicon_htk_triphone = os.path.join(lexicon_dir, 'lex_triphone.htk')
|
lexicon_htk_triphone = os.path.join(lexicon_dir, 'lex_triphone.htk')
|
||||||
|
|
||||||
feature_dir = os.path.join(default.htk_dir, 'mfc')
|
feature_dir = os.path.join(default.htk_dir, 'mfc')
|
||||||
@ -71,10 +74,20 @@ fh.make_new_directory(label_dir, existing_dir='leave')
|
|||||||
|
|
||||||
|
|
||||||
## training
|
## training
|
||||||
hcompv_scp_train = os.path.join(tmp_dir, 'train.scp')
|
if combine_all:
|
||||||
mlf_file_train = os.path.join(label_dir, 'train_phone.mlf')
|
hcompv_scp_train = os.path.join(tmp_dir, 'all.scp')
|
||||||
mlf_file_train_with_sp = os.path.join(label_dir, 'train_phone_with_sp.mlf')
|
mlf_file_train = os.path.join(label_dir, 'all_phone.mlf')
|
||||||
mlf_file_train_aligned = os.path.join(label_dir, 'train_phone_aligned.mlf')
|
mlf_file_train_word = os.path.join(label_dir, 'all_word.mlf')
|
||||||
|
mlf_file_train_with_sp = os.path.join(label_dir, 'all_phone_with_sp.mlf')
|
||||||
|
mlf_file_train_aligned = os.path.join(label_dir, 'all_phone_aligned.mlf')
|
||||||
|
triphone_mlf = os.path.join(label_dir, 'all_triphone.mlf')
|
||||||
|
else:
|
||||||
|
hcompv_scp_train = os.path.join(tmp_dir, 'train.scp')
|
||||||
|
mlf_file_train = os.path.join(label_dir, 'train_phone.mlf')
|
||||||
|
mlf_file_train_word = os.path.join(label_dir, 'train_word.mlf')
|
||||||
|
mlf_file_train_with_sp = os.path.join(label_dir, 'train_phone_with_sp.mlf')
|
||||||
|
mlf_file_train_aligned = os.path.join(label_dir, 'train_phone_aligned.mlf')
|
||||||
|
triphone_mlf = os.path.join(label_dir, 'train_triphone.mlf')
|
||||||
hcompv_scp_train_updated = hcompv_scp_train.replace('.scp', '_updated.scp')
|
hcompv_scp_train_updated = hcompv_scp_train.replace('.scp', '_updated.scp')
|
||||||
|
|
||||||
## testing
|
## testing
|
||||||
@ -104,19 +117,18 @@ if make_lexicon:
|
|||||||
print('>>> fixing the lexicon...')
|
print('>>> fixing the lexicon...')
|
||||||
fame_functions.fix_lexicon(lexicon_htk)
|
fame_functions.fix_lexicon(lexicon_htk)
|
||||||
|
|
||||||
## add sp to the end of each line.
|
## adding sp to the lexicon for HTK.
|
||||||
#print('>>> adding sp...')
|
print('>>> adding sp to the lexicon...')
|
||||||
#with open(lexicon_htk) as f:
|
with open(lexicon_htk) as f:
|
||||||
# lines = f.read().split('\n')
|
lines = f.read().split('\n')
|
||||||
#lines = [line + ' sp' for line in lines]
|
with open(lexicon_htk_with_sp, 'wb') as f:
|
||||||
#with open(lexicon_htk_with_sp, 'wb') as f:
|
f.write(bytes(' sp\n'.join(lines), 'ascii'))
|
||||||
# f.write(bytes('\n'.join(lines), 'ascii'))
|
|
||||||
|
|
||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
## intialize the instance for HTK.
|
## intialize the instance for HTK.
|
||||||
chtk = pyhtk.HTK(config_dir, fame_asr.phoneset_htk, lexicon_htk, feature_size)
|
chtk = pyhtk.HTK(config_dir, fame_asr.phoneset_htk, lexicon_htk_with_sp, feature_size)
|
||||||
|
|
||||||
|
|
||||||
## ======================= make label files =======================
|
## ======================= make label files =======================
|
||||||
@ -152,7 +164,7 @@ if make_label:
|
|||||||
shutil.move(dictionary_file, os.path.join(label_dir_, filename + '.dic'))
|
shutil.move(dictionary_file, os.path.join(label_dir_, filename + '.dic'))
|
||||||
|
|
||||||
label_file = os.path.join(label_dir_, filename + '.lab')
|
label_file = os.path.join(label_dir_, filename + '.lab')
|
||||||
chtk.create_label_file(sentence_htk, label_file)
|
chtk.make_label_file(sentence_htk, label_file)
|
||||||
else:
|
else:
|
||||||
os.remove(dictionary_file)
|
os.remove(dictionary_file)
|
||||||
|
|
||||||
@ -174,7 +186,6 @@ if make_mlf:
|
|||||||
os.remove(empty_dic_file)
|
os.remove(empty_dic_file)
|
||||||
|
|
||||||
for dataset in dataset_list:
|
for dataset in dataset_list:
|
||||||
#wav_dir_ = os.path.join(default.fame_dir, 'fame', 'wav', dataset)
|
|
||||||
feature_dir_ = os.path.join(feature_dir, dataset)
|
feature_dir_ = os.path.join(feature_dir, dataset)
|
||||||
label_dir_ = os.path.join(label_dir, dataset)
|
label_dir_ = os.path.join(label_dir, dataset)
|
||||||
mlf_word = os.path.join(label_dir, dataset + '_word.mlf')
|
mlf_word = os.path.join(label_dir, dataset + '_word.mlf')
|
||||||
@ -183,11 +194,11 @@ if make_mlf:
|
|||||||
|
|
||||||
print(">>> generating a word level mlf file for {}...".format(dataset))
|
print(">>> generating a word level mlf file for {}...".format(dataset))
|
||||||
chtk.label2mlf(label_dir_, mlf_word)
|
chtk.label2mlf(label_dir_, mlf_word)
|
||||||
|
|
||||||
print(">>> generating a phone level mlf file for {}...".format(dataset))
|
print(">>> generating a phone level mlf file for {}...".format(dataset))
|
||||||
chtk.mlf_word2phone(mlf_phone, mlf_word, with_sp=False)
|
chtk.mlf_word2phone(mlf_phone, mlf_word, with_sp=False)
|
||||||
chtk.mlf_word2phone(mlf_phone_with_sp, mlf_word, with_sp=True)
|
chtk.mlf_word2phone(mlf_phone_with_sp, mlf_word, with_sp=True)
|
||||||
|
|
||||||
|
|
||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
@ -197,7 +208,7 @@ if extract_features:
|
|||||||
timer_start = time.time()
|
timer_start = time.time()
|
||||||
print('==== extract features on dataset {} ===='.format(dataset))
|
print('==== extract features on dataset {} ===='.format(dataset))
|
||||||
|
|
||||||
wav_dir_ = os.path.join(default.fame_dir, 'fame', 'wav', dataset)
|
wav_dir_ = os.path.join(default.fame_dir, 'fame', 'wav', dataset)
|
||||||
label_dir_ = os.path.join(label_dir, dataset)
|
label_dir_ = os.path.join(label_dir, dataset)
|
||||||
feature_dir_ = os.path.join(feature_dir, dataset)
|
feature_dir_ = os.path.join(feature_dir, dataset)
|
||||||
fh.make_new_directory(feature_dir_, existing_dir='delete')
|
fh.make_new_directory(feature_dir_, existing_dir='delete')
|
||||||
@ -217,8 +228,8 @@ if extract_features:
|
|||||||
+ os.path.join(feature_dir_, os.path.basename(lab_file).replace('.lab', '.mfc'))
|
+ os.path.join(feature_dir_, os.path.basename(lab_file).replace('.lab', '.mfc'))
|
||||||
for lab_file in lab_list]
|
for lab_file in lab_list]
|
||||||
|
|
||||||
if os.path.exists(empty_mfc_file):
|
#if os.path.exists(empty_mfc_file):
|
||||||
os.remove(empty_mfc_file)
|
# os.remove(empty_mfc_file)
|
||||||
with open(hcopy_scp.name, 'wb') as f:
|
with open(hcopy_scp.name, 'wb') as f:
|
||||||
f.write(bytes('\n'.join(feature_list), 'ascii'))
|
f.write(bytes('\n'.join(feature_list), 'ascii'))
|
||||||
|
|
||||||
@ -235,9 +246,64 @@ if extract_features:
|
|||||||
with open(hcompv_scp, 'wb') as f:
|
with open(hcompv_scp, 'wb') as f:
|
||||||
f.write(bytes('\n'.join(mfc_list) + '\n', 'ascii'))
|
f.write(bytes('\n'.join(mfc_list) + '\n', 'ascii'))
|
||||||
|
|
||||||
|
print(">>> extracting features on stimmen...")
|
||||||
|
chtk.wav2mfc(os.path.join(htk_stimmen_dir, 'hcopy.scp'))
|
||||||
|
|
||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
|
## ======================= flat start monophones =======================
|
||||||
|
if combine_all:
|
||||||
|
# script files.
|
||||||
|
fh.concatenate(
|
||||||
|
os.path.join(tmp_dir, 'devel.scp'),
|
||||||
|
os.path.join(tmp_dir, 'test.scp'),
|
||||||
|
hcompv_scp_train
|
||||||
|
)
|
||||||
|
fh.concatenate(
|
||||||
|
hcompv_scp_train,
|
||||||
|
os.path.join(tmp_dir, 'train.scp'),
|
||||||
|
hcompv_scp_train
|
||||||
|
)
|
||||||
|
|
||||||
|
# phone level mlfs.
|
||||||
|
fh.concatenate(
|
||||||
|
os.path.join(label_dir, 'devel_phone.mlf'),
|
||||||
|
os.path.join(label_dir, 'test_phone.mlf'),
|
||||||
|
mlf_file_train
|
||||||
|
)
|
||||||
|
fh.concatenate(
|
||||||
|
mlf_file_train,
|
||||||
|
os.path.join(label_dir, 'train_phone.mlf'),
|
||||||
|
mlf_file_train
|
||||||
|
)
|
||||||
|
|
||||||
|
# phone level mlfs with sp.
|
||||||
|
fh.concatenate(
|
||||||
|
os.path.join(label_dir, 'devel_phone_with_sp.mlf'),
|
||||||
|
os.path.join(label_dir, 'test_phone_with_sp.mlf'),
|
||||||
|
mlf_file_train_with_sp
|
||||||
|
)
|
||||||
|
fh.concatenate(
|
||||||
|
mlf_file_train_with_sp,
|
||||||
|
os.path.join(label_dir, 'train_phone_with_sp.mlf'),
|
||||||
|
mlf_file_train_with_sp
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
# word level mlfs.
|
||||||
|
fh.concatenate(
|
||||||
|
os.path.join(label_dir, 'devel_word.mlf'),
|
||||||
|
os.path.join(label_dir, 'test_word.mlf'),
|
||||||
|
mlf_file_train_word
|
||||||
|
)
|
||||||
|
fh.concatenate(
|
||||||
|
mlf_file_train_word,
|
||||||
|
os.path.join(label_dir, 'train_word.mlf'),
|
||||||
|
mlf_file_train_word
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
## ======================= flat start monophones =======================
|
## ======================= flat start monophones =======================
|
||||||
if flat_start:
|
if flat_start:
|
||||||
timer_start = time.time()
|
timer_start = time.time()
|
||||||
@ -246,17 +312,14 @@ if flat_start:
|
|||||||
|
|
||||||
chtk.flat_start(hcompv_scp_train, model_mono0_dir)
|
chtk.flat_start(hcompv_scp_train, model_mono0_dir)
|
||||||
|
|
||||||
# create macros.
|
# make macros.
|
||||||
vFloors = os.path.join(model_mono0_dir, 'vFloors')
|
vFloors = os.path.join(model_mono0_dir, 'vFloors')
|
||||||
if os.path.exists(vFloors):
|
if os.path.exists(vFloors):
|
||||||
chtk.create_macros(vFloors)
|
chtk.make_macros(vFloors)
|
||||||
|
|
||||||
# allocate mean & variance to all phones in the phone list
|
# allocate mean & variance to all phones in the phone list
|
||||||
print('>>> allocating mean & variance to all phones in the phone list...')
|
print('>>> allocating mean & variance to all phones in the phone list...')
|
||||||
chtk.create_hmmdefs(
|
chtk.make_hmmdefs(model_mono0_dir)
|
||||||
os.path.join(model_mono0_dir, proto_name),
|
|
||||||
os.path.join(model_mono0_dir, 'hmmdefs')
|
|
||||||
)
|
|
||||||
|
|
||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
@ -320,8 +383,9 @@ if train_monophone_with_re_aligned_mlf:
|
|||||||
os.path.join(modeln_dir, 'macros'),
|
os.path.join(modeln_dir, 'macros'),
|
||||||
os.path.join(modeln_dir, 'hmmdefs'),
|
os.path.join(modeln_dir, 'hmmdefs'),
|
||||||
mlf_file_train_aligned,
|
mlf_file_train_aligned,
|
||||||
os.path.join(label_dir, 'train_word.mlf'),
|
mlf_file_train_word,
|
||||||
hcompv_scp_train)
|
hcompv_scp_train)
|
||||||
|
chtk.fix_mlf(mlf_file_train_aligned)
|
||||||
|
|
||||||
print('>>> updating the script file... ')
|
print('>>> updating the script file... ')
|
||||||
chtk.update_script_file(
|
chtk.update_script_file(
|
||||||
@ -349,24 +413,55 @@ if train_monophone_with_re_aligned_mlf:
|
|||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
## ======================= train triphone =======================
|
## ======================= increase mixture =======================
|
||||||
if train_triphone:
|
if increase_mixture:
|
||||||
print('==== traina triphone model ====')
|
print('==== increase mixture ====')
|
||||||
timer_start = time.time()
|
timer_start = time.time()
|
||||||
|
for nmix in [2, 4, 8, 16]:
|
||||||
|
if nmix == 2:
|
||||||
|
modeln_dir_ = model_mono1sp2_dir
|
||||||
|
else:
|
||||||
|
modeln_dir_ = os.path.join(model_dir, 'mono'+str(nmix_))
|
||||||
|
modeln_dir = os.path.join(model_dir, 'mono'+str(nmix))
|
||||||
|
|
||||||
triphonelist_txt = os.path.join(config_dir, 'triphonelist.txt')
|
print('mixture: {}'.format(nmix))
|
||||||
triphone_mlf = os.path.join(default.htk_dir, 'label', 'train_triphone.mlf')
|
fh.make_new_directory(modeln_dir, existing_dir='delete')
|
||||||
|
niter = chtk.get_niter_max(modeln_dir_)
|
||||||
|
chtk.increase_mixture(
|
||||||
|
os.path.join(modeln_dir_, 'iter'+str(niter), 'hmmdefs'),
|
||||||
|
nmix,
|
||||||
|
os.path.join(modeln_dir, 'iter0'),
|
||||||
|
model_type='monophone_with_sp')
|
||||||
|
shutil.copy2(os.path.join(modeln_dir_, 'iter'+str(niter), 'macros'),
|
||||||
|
os.path.join(modeln_dir, 'iter0', 'macros'))
|
||||||
|
|
||||||
print('>>> making triphone list... ')
|
#improvement_threshold = -10
|
||||||
chtk.make_triphonelist(
|
niter = chtk.re_estimation_until_saturated(
|
||||||
triphonelist_txt,
|
modeln_dir,
|
||||||
triphone_mlf,
|
os.path.join(modeln_dir_, 'iter0'),
|
||||||
mlf_file_train_aligned)
|
improvement_threshold,
|
||||||
|
hcompv_scp_train_updated,
|
||||||
|
os.path.join(htk_stimmen_dir, 'mfc'),
|
||||||
|
'mfc',
|
||||||
|
os.path.join(htk_stimmen_dir, 'word_lattice.ltc'),
|
||||||
|
mlf_file=mlf_file_train_aligned,
|
||||||
|
lexicon=os.path.join(htk_stimmen_dir, 'lexicon_recognition.dic'),
|
||||||
|
model_type='monophone_with_sp'
|
||||||
|
)
|
||||||
|
nmix_ = nmix
|
||||||
|
|
||||||
print('>>> making triphone header... ')
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
chtk.make_tri_hed(
|
|
||||||
os.path.join(config_dir, 'mktri.hed')
|
|
||||||
)
|
## ======================= train triphone =======================
|
||||||
|
print('>>> making triphone list... ')
|
||||||
|
chtk.make_triphonelist(
|
||||||
|
mlf_file_train_aligned,
|
||||||
|
triphone_mlf)
|
||||||
|
|
||||||
|
if train_triphone:
|
||||||
|
print('==== train triphone model ====')
|
||||||
|
timer_start = time.time()
|
||||||
|
|
||||||
print('>>> init triphone model... ')
|
print('>>> init triphone model... ')
|
||||||
niter = chtk.get_niter_max(model_mono1sp2_dir)
|
niter = chtk.get_niter_max(model_mono1sp2_dir)
|
||||||
@ -377,8 +472,8 @@ if train_triphone:
|
|||||||
)
|
)
|
||||||
|
|
||||||
print('>>> re-estimation... ')
|
print('>>> re-estimation... ')
|
||||||
# I wanted to train until satulated:
|
## I wanted to train until satulated:
|
||||||
# #niter = chtk.re_estimation_until_saturated(
|
#niter = chtk.re_estimation_until_saturated(
|
||||||
# model_tri1_dir,
|
# model_tri1_dir,
|
||||||
# os.path.join(model_tri1_dir, 'iter0'),
|
# os.path.join(model_tri1_dir, 'iter0'),
|
||||||
# improvement_threshold,
|
# improvement_threshold,
|
||||||
@ -395,7 +490,6 @@ if train_triphone:
|
|||||||
# ERROR [+8231] GetHCIModel: Cannot find hmm [i:-]r[+???]
|
# ERROR [+8231] GetHCIModel: Cannot find hmm [i:-]r[+???]
|
||||||
# therefore only two times re-estimation is performed.
|
# therefore only two times re-estimation is performed.
|
||||||
output_dir = model_tri1_dir
|
output_dir = model_tri1_dir
|
||||||
|
|
||||||
for niter in range(1, 4):
|
for niter in range(1, 4):
|
||||||
hmm_n = 'iter' + str(niter)
|
hmm_n = 'iter' + str(niter)
|
||||||
hmm_n_pre = 'iter' + str(niter-1)
|
hmm_n_pre = 'iter' + str(niter-1)
|
||||||
@ -414,18 +508,59 @@ if train_triphone:
|
|||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
## ======================= train triphone =======================
|
## ======================= train tied-state triphones =======================
|
||||||
if train_triphone_tied:
|
if train_triphone_tied:
|
||||||
print('==== traina tied-state triphone ====')
|
print('==== train tied-state triphones ====')
|
||||||
timer_start = time.time()
|
timer_start = time.time()
|
||||||
|
|
||||||
print('>>> making lexicon for triphone... ')
|
print('>>> making lexicon for triphone... ')
|
||||||
chtk.make_triphone_full(phonelist_full_txt, lexicon_htk_triphone)
|
chtk.make_lexicon_triphone(phonelist_full_txt, lexicon_htk_triphone)
|
||||||
|
chtk.combine_phonelists(phonelist_full_txt)
|
||||||
|
|
||||||
print('>>> making headers... ')
|
print('>>> making a tree header... ')
|
||||||
chtk.make_tree_header(tree_hed)
|
fame_phonetics.make_quests_hed(quests_hed)
|
||||||
fame_phonetics.make_quests_hed(quest_hed)
|
stats = os.path.join(r'c:\OneDrive\Research\rug\experiments\acoustic_model\fame\htk\model\tri1\iter3', 'stats')
|
||||||
|
chtk.make_tree_header(tree_hed, quests_hed, stats, config_dir)
|
||||||
|
|
||||||
|
print('>>> init triphone model... ')
|
||||||
|
niter = chtk.get_niter_max(model_tri1_dir)
|
||||||
|
fh.make_new_directory(os.path.join(model_tri1tied_dir, 'iter0'), existing_dir='leave')
|
||||||
|
chtk.init_triphone(
|
||||||
|
os.path.join(model_tri1_dir, 'iter'+str(niter)),
|
||||||
|
os.path.join(model_tri1tied_dir, 'iter0'),
|
||||||
|
tied=True)
|
||||||
|
|
||||||
|
# I wanted to train until satulated:
|
||||||
|
#niter = chtk.re_estimation_until_saturated(
|
||||||
|
# model_tri1tied_dir,
|
||||||
|
# os.path.join(model_tri1tied_dir, 'iter0'),
|
||||||
|
# improvement_threshold,
|
||||||
|
# hcompv_scp_train_updated,
|
||||||
|
# os.path.join(htk_stimmen_dir, 'mfc'),
|
||||||
|
# 'mfc',
|
||||||
|
# os.path.join(htk_stimmen_dir, 'word_lattice.ltc'),
|
||||||
|
# mlf_file=triphone_mlf,
|
||||||
|
# lexicon=os.path.join(htk_stimmen_dir, 'lexicon_recognition.dic'),
|
||||||
|
# model_type='triphone'
|
||||||
|
# )
|
||||||
|
#
|
||||||
|
# but because the data size is limited, some triphone cannot be trained and received the error:
|
||||||
|
# ERROR [+8231] GetHCIModel: Cannot find hmm [i:-]r[+???]
|
||||||
|
# therefore only 3 times re-estimation is performed.
|
||||||
|
output_dir = model_tri1tied_dir
|
||||||
|
for niter in range(1, 4):
|
||||||
|
hmm_n = 'iter' + str(niter)
|
||||||
|
hmm_n_pre = 'iter' + str(niter-1)
|
||||||
|
_modeln_dir = os.path.join(output_dir, hmm_n)
|
||||||
|
_modeln_dir_pre = os.path.join(output_dir, hmm_n_pre)
|
||||||
|
|
||||||
|
fh.make_new_directory(_modeln_dir, 'leave')
|
||||||
|
chtk.re_estimation(
|
||||||
|
os.path.join(_modeln_dir_pre, 'hmmdefs'),
|
||||||
|
_modeln_dir,
|
||||||
|
hcompv_scp_train_updated,
|
||||||
|
mlf_file=triphone_mlf,
|
||||||
|
macros=os.path.join(_modeln_dir_pre, 'macros'),
|
||||||
|
model_type='triphone')
|
||||||
|
|
||||||
print("elapsed time: {}".format(time.time() - timer_start))
|
print("elapsed time: {}".format(time.time() - timer_start))
|
||||||
|
|
||||||
|
|
||||||
|
@ -1,20 +1,19 @@
|
|||||||
## this script should be used only by Aki Kunikoshi.
|
## this script should be used only by Aki Kunikoshi.
|
||||||
|
|
||||||
|
import os
|
||||||
|
|
||||||
import numpy as np
|
import numpy as np
|
||||||
|
import pandas as pd
|
||||||
import argparse
|
import argparse
|
||||||
import json
|
import json
|
||||||
|
|
||||||
from novoapi.backend import session
|
from novoapi.backend import session
|
||||||
|
|
||||||
import os
|
|
||||||
#os.chdir(r'C:\Users\Aki\source\repos\acoustic_model\acoustic_model')
|
|
||||||
import defaultfiles as default
|
import defaultfiles as default
|
||||||
|
import convert_phoneset
|
||||||
|
|
||||||
|
|
||||||
def load_phonset():
|
def load_novo70_phoneset():
|
||||||
translation_key_ipa2novo70 = dict()
|
|
||||||
translation_key_novo702ipa = dict()
|
|
||||||
|
|
||||||
#phonelist_novo70_ = pd.ExcelFile(default.phonelist_novo70_xlsx)
|
#phonelist_novo70_ = pd.ExcelFile(default.phonelist_novo70_xlsx)
|
||||||
#df = pd.read_excel(phonelist_novo70_, 'list')
|
#df = pd.read_excel(phonelist_novo70_, 'list')
|
||||||
## *_simple includes columns which has only one phone in.
|
## *_simple includes columns which has only one phone in.
|
||||||
@ -23,21 +22,23 @@ def load_phonset():
|
|||||||
# print('{0}:{1}'.format(ipa, novo70))
|
# print('{0}:{1}'.format(ipa, novo70))
|
||||||
# translation_key[ipa] = novo70
|
# translation_key[ipa] = novo70
|
||||||
#phonelist_novo70 = np.unique(list(df['novo70_simple']))
|
#phonelist_novo70 = np.unique(list(df['novo70_simple']))
|
||||||
|
novo70_phoneset = pd.read_csv(default.novo70_phoneset, delimiter='\t', header=None, encoding="utf-8")
|
||||||
|
novo70_phoneset.rename(columns={0: 'novo70', 1: 'ipa', 2: 'description'}, inplace=True)
|
||||||
|
|
||||||
phoneset_ipa = []
|
#phoneset_ipa = []
|
||||||
phoneset_novo70 = []
|
#phoneset_novo70 = []
|
||||||
with open(default.novo70_phoneset, "rt", encoding="utf-8") as fin:
|
#with open(default.novo70_phoneset, "rt", encoding="utf-8") as fin:
|
||||||
lines = fin.read()
|
# lines = fin.read()
|
||||||
lines = lines.split('\n')
|
# lines = lines.split('\n')
|
||||||
for line in lines:
|
# for line in lines:
|
||||||
words = line.split('\t')
|
# words = line.split('\t')
|
||||||
if len(words) > 1:
|
# if len(words) > 1:
|
||||||
novo70 = words[0]
|
# novo70 = words[0]
|
||||||
ipa = words[1]
|
# ipa = words[1]
|
||||||
phoneset_ipa.append(ipa)
|
# phoneset_ipa.append(ipa)
|
||||||
phoneset_novo70.append(novo70)
|
# phoneset_novo70.append(novo70)
|
||||||
translation_key_ipa2novo70[ipa] = novo70
|
# translation_key_ipa2novo70[ipa] = novo70
|
||||||
translation_key_novo702ipa[novo70] = ipa
|
# translation_key_novo702ipa[novo70] = ipa
|
||||||
|
|
||||||
# As per Nederlandse phoneset_aki.xlsx recieved from David
|
# As per Nederlandse phoneset_aki.xlsx recieved from David
|
||||||
# [ɔː] oh / ohr # from ipa->novo70, only oh is used.
|
# [ɔː] oh / ohr # from ipa->novo70, only oh is used.
|
||||||
@ -48,14 +49,25 @@ def load_phonset():
|
|||||||
# [w] wv in IPA written as ʋ.
|
# [w] wv in IPA written as ʋ.
|
||||||
extra_ipa = ['ɔː', 'ɪː', 'iː', 'œː', 'ɛː', 'ʋ']
|
extra_ipa = ['ɔː', 'ɪː', 'iː', 'œː', 'ɛː', 'ʋ']
|
||||||
extra_novo70 = ['oh', 'ih', 'iy', 'uh', 'eh', 'wv']
|
extra_novo70 = ['oh', 'ih', 'iy', 'uh', 'eh', 'wv']
|
||||||
for ipa, novo70 in zip(extra_ipa, extra_novo70):
|
|
||||||
phoneset_ipa.append(ipa)
|
phoneset_ipa = list(novo70_phoneset['ipa'])
|
||||||
phoneset_novo70.append(novo70)
|
phoneset_ipa.extend(extra_ipa)
|
||||||
|
phoneset_ipa = [i.replace('ː', ':') for i in phoneset_ipa]
|
||||||
|
|
||||||
|
phoneset_novo70 = list(novo70_phoneset['novo70'])
|
||||||
|
phoneset_novo70.extend(extra_novo70)
|
||||||
|
phoneset_novo70 = [i.replace('ː', ':') for i in phoneset_novo70]
|
||||||
|
|
||||||
|
translation_key_ipa2novo70 = dict()
|
||||||
|
translation_key_novo702ipa = dict()
|
||||||
|
for ipa, novo70 in zip(phoneset_ipa, phoneset_novo70):
|
||||||
|
#phoneset_ipa.append(ipa)
|
||||||
|
#phoneset_novo70.append(novo70)
|
||||||
translation_key_ipa2novo70[ipa] = novo70
|
translation_key_ipa2novo70[ipa] = novo70
|
||||||
translation_key_novo702ipa[novo70] = ipa
|
translation_key_novo702ipa[novo70] = ipa
|
||||||
|
|
||||||
translation_key_novo702ipa['ohr'] = 'ɔː'
|
translation_key_novo702ipa['ohr'] = 'ɔ:'
|
||||||
translation_key_novo702ipa['ihr'] = 'ɪː'
|
translation_key_novo702ipa['ihr'] = 'ɪ:'
|
||||||
|
|
||||||
phoneset_ipa = np.unique(phoneset_ipa)
|
phoneset_ipa = np.unique(phoneset_ipa)
|
||||||
phoneset_novo70 = np.unique(phoneset_novo70)
|
phoneset_novo70 = np.unique(phoneset_novo70)
|
||||||
@ -63,25 +75,6 @@ def load_phonset():
|
|||||||
return phoneset_ipa, phoneset_novo70, translation_key_ipa2novo70, translation_key_novo702ipa
|
return phoneset_ipa, phoneset_novo70, translation_key_ipa2novo70, translation_key_novo702ipa
|
||||||
|
|
||||||
|
|
||||||
def multi_character_tokenize(line, multi_character_tokens):
|
|
||||||
"""
|
|
||||||
Tries to match one of the tokens in multi_character_tokens at each position of line,
|
|
||||||
starting at position 0,
|
|
||||||
if so tokenizes and eats that token. Otherwise tokenizes a single character.
|
|
||||||
|
|
||||||
Copied from forced_alignment.convert_phone_set.py
|
|
||||||
"""
|
|
||||||
while line != '':
|
|
||||||
for token in multi_character_tokens:
|
|
||||||
if line.startswith(token) and len(token) > 0:
|
|
||||||
yield token
|
|
||||||
line = line[len(token):]
|
|
||||||
break
|
|
||||||
else:
|
|
||||||
yield line[:1]
|
|
||||||
line = line[1:]
|
|
||||||
|
|
||||||
|
|
||||||
def split_ipa(line):
|
def split_ipa(line):
|
||||||
"""
|
"""
|
||||||
Split a line by IPA phones.
|
Split a line by IPA phones.
|
||||||
@ -89,13 +82,16 @@ def split_ipa(line):
|
|||||||
:param string line: one line written in IPA.
|
:param string line: one line written in IPA.
|
||||||
:return string lineSeperated: the line splitted in IPA phone.
|
:return string lineSeperated: the line splitted in IPA phone.
|
||||||
"""
|
"""
|
||||||
|
phoneset_ipa, _, _, _ = load_novo70_phoneset()
|
||||||
|
#multi_character_phones = [i for i in phoneset_ipa if len(i) > 1]
|
||||||
|
#multi_character_phones.sort(key=len, reverse=True)
|
||||||
|
#multi_character_phones = [
|
||||||
|
# # IPAs in CGN.
|
||||||
|
# u'ʌu', u'ɛi', u'œy', u'aː', u'eː', u'iː', u'oː', u'øː', u'ɛː', u'œː', u'ɔː', u'ɛ̃ː', u'ɑ̃ː', u'ɔ̃ː', u'œ̃', u'ɪː'
|
||||||
|
# ]
|
||||||
|
#return [phone for phone in multi_character_tokenize(line.strip(), multi_character_phones)]
|
||||||
|
|
||||||
multi_character_phones = [
|
return convert_phoneset.split_word(line, phoneset_ipa)
|
||||||
# IPAs in CGN.
|
|
||||||
u'ʌu', u'ɛi', u'œy', u'aː', u'eː', u'iː', u'oː', u'øː', u'ɛː', u'œː', u'ɔː', u'ɛ̃ː', u'ɑ̃ː', u'ɔ̃ː', u'œ̃', u'ɪː'
|
|
||||||
]
|
|
||||||
|
|
||||||
return [phone for phone in multi_character_tokenize(line.strip(), multi_character_phones)]
|
|
||||||
|
|
||||||
|
|
||||||
def split_novo70(line):
|
def split_novo70(line):
|
||||||
@ -104,29 +100,32 @@ def split_novo70(line):
|
|||||||
:param string line: one line written in novo70.
|
:param string line: one line written in novo70.
|
||||||
:return string lineSeperated: the line splitted by novo70 phones.
|
:return string lineSeperated: the line splitted by novo70 phones.
|
||||||
"""
|
"""
|
||||||
_, phoneset_novo70, _, _ = load_phonset()
|
_, phoneset_novo70, _, _ = load_novo70_phoneset()
|
||||||
multi_character_phones = [p for p in phoneset_novo70 if len(p) > 1]
|
#multi_character_phones = [p for p in phoneset_novo70 if len(p) > 1]
|
||||||
multi_character_phones = sorted(multi_character_phones, key=len, reverse=True)
|
#multi_character_phones = sorted(multi_character_phones, key=len, reverse=True)
|
||||||
|
multi_character_phones = convert_phoneset.extract_multi_character_phones(phoneset_novo70)
|
||||||
|
|
||||||
return ['sp' if phone == ' ' else phone
|
return ['sp' if phone == ' ' else phone
|
||||||
for phone in multi_character_tokenize(line.strip(), multi_character_phones)]
|
for phone in multi_character_tokenize(line.strip(), multi_character_phones)]
|
||||||
|
|
||||||
|
|
||||||
def novo702ipa(tokens):
|
def novo702ipa(line):
|
||||||
pronunciation = []
|
#pronunciation = []
|
||||||
_, _, _, translation_key = load_phonset()
|
_, _, _, translation_key = load_novo70_phoneset()
|
||||||
for phone in split_novo70(tokens):
|
#for phone in split_novo70(tokens):
|
||||||
pronunciation.append(translation_key.get(phone, phone))
|
# pronunciation.append(translation_key.get(phone, phone))
|
||||||
return ' '.join(pronunciation)
|
#return ' '.join(pronunciation)
|
||||||
|
return ' '.join(convert_phoneset.convert_phoneset(split_novo70(line), translation_key))
|
||||||
|
|
||||||
|
|
||||||
# numbering of novo70 should be checked.
|
# numbering of novo70 should be checked.
|
||||||
def ipa2novo70(tokens):
|
def ipa2novo70(line):
|
||||||
pronunciation = []
|
#pronunciation = []
|
||||||
_, _, translation_key, _ = load_phonset()
|
_, _, translation_key, _ = load_novo70_phoneset()
|
||||||
for phone in split_ipa(tokens):
|
#for phone in split_ipa(tokens):
|
||||||
pronunciation.append(translation_key.get(phone, phone))
|
# pronunciation.append(translation_key.get(phone, phone))
|
||||||
return ' '.join(pronunciation)
|
#return ' '.join(pronunciation)
|
||||||
|
return ' '.join(convert_phoneset.convert_phoneset(split_ipa(line), translation_key))
|
||||||
|
|
||||||
|
|
||||||
def make_grammar(word, pronunciation_ipa):
|
def make_grammar(word, pronunciation_ipa):
|
||||||
@ -174,6 +173,7 @@ def forced_alignment(wav_file, word, pronunciation_ipa):
|
|||||||
p = argparse.ArgumentParser()
|
p = argparse.ArgumentParser()
|
||||||
p.add_argument("--user", default='martijn.wieling')
|
p.add_argument("--user", default='martijn.wieling')
|
||||||
p.add_argument("--password", default='xxxxxx')
|
p.add_argument("--password", default='xxxxxx')
|
||||||
|
|
||||||
args = p.parse_args()
|
args = p.parse_args()
|
||||||
|
|
||||||
rec = session.Recognizer(grammar_version="1.0", lang="nl", snodeid=101, user=args.user, password=args.password, keepopen=True) # , modeldir=modeldir)
|
rec = session.Recognizer(grammar_version="1.0", lang="nl", snodeid=101, user=args.user, password=args.password, keepopen=True) # , modeldir=modeldir)
|
||||||
@ -196,4 +196,5 @@ def result2pronunciation(result, word):
|
|||||||
|
|
||||||
if __name__ == 'main':
|
if __name__ == 'main':
|
||||||
pronunciation_ipa = ['rø:s', 'mɑn', 'mɑntsjə']
|
pronunciation_ipa = ['rø:s', 'mɑn', 'mɑntsjə']
|
||||||
grammar = make_grammar('reus', pronunciation_ipa)
|
#grammar = make_grammar('reus', pronunciation_ipa)
|
||||||
|
phoneset_ipa, phoneset_novo70, translation_key_ipa2novo70, translation_key_novo702ipa = load_novo70_phoneset()
|
Loading…
Reference in New Issue
Block a user