def sick_corpus(sick_dict, basic_dir, generated_dir):
output_directory = os.path.join(generated_dir, 'sickcorpus')
corpus = Corpus(basic_dir, output_directory, num_jobs = 2)
corpus.write()
corpus.create_mfccs()
corpus.setup_splits(sick_dict)
return corpus
def sick_corpus(sick_dict, basic_dir, generated_dir):
output_directory = os.path.join(generated_dir, 'sickcorpus')
corpus = Corpus(basic_dir, output_directory, num_jobs=2)
corpus.write()
corpus.create_mfccs()
corpus.setup_splits(sick_dict)
return corpus
def test_stereo(basic_dict_path, textgrid_directory, generated_dir):
temp = os.path.join(generated_dir, 'stereo')
dictionary = Dictionary(basic_dict_path, os.path.join(temp, 'basic'))
dictionary.write()
d = Corpus(os.path.join(textgrid_directory, 'stereo'), temp)
d.write()
d.create_mfccs()
d.setup_splits(dictionary)
assert (d.get_feat_dim() == '39')
def test_basic(basic_dict_path, basic_dir, generated_dir):
dictionary = Dictionary(basic_dict_path, os.path.join(generated_dir, "basic"))
dictionary.write()
output_directory = os.path.join(generated_dir, "basic")
d = Corpus(basic_dir, output_directory)
d.write()
d.create_mfccs()
d.setup_splits(dictionary)
assert d.get_feat_dim() == "39"
def test_stereo(basic_dict_path, textgrid_directory, generated_dir):
temp = os.path.join(generated_dir, "stereo")
dictionary = Dictionary(basic_dict_path, os.path.join(temp, "basic"))
dictionary.write()
d = Corpus(os.path.join(textgrid_directory, "stereo"), temp)
d.write()
d.create_mfccs()
d.setup_splits(dictionary)
assert d.get_feat_dim() == "39"
def test_basic(basic_dict_path, basic_dir, generated_dir):
dictionary = Dictionary(basic_dict_path,
os.path.join(generated_dir, 'basic'))
dictionary.write()
output_directory = os.path.join(generated_dir, 'basic')
d = Corpus(basic_dir, output_directory)
d.write()
d.create_mfccs()
d.setup_splits(dictionary)
assert (d.get_feat_dim() == '39')
def test_acoustic(basic_dir, generated_dir):
output_directory = os.path.join(generated_dir, 'acoustic')
d = Corpus(basic_dir, output_directory)
d.write()
d.create_mfccs()
n = no_dictionary(d, output_directory)
d.setup_splits(n)
assert n.words['should'] == [['s', 'h', 'o', 'u', 'l', 'd']]
assert '<vocnoise>' not in n.words
assert n.words['here\'s'] == [['h', 'e', 'r', 'e', 's']]
def test_vietnamese(textgrid_directory, generated_dir):
output_directory = os.path.join(generated_dir, 'vietnamese')
d = Corpus(os.path.join(textgrid_directory, 'vietnamese'), output_directory)
d.write()
d.create_mfccs()
n = no_dictionary(d, output_directory)
d.setup_splits(n)
assert n.words['chăn'] == [['c', 'h', 'ă', 'n']]
assert '<vocnoise>' not in n.words
assert n.words['tập'] == [['t','ậ','p']]
def align_corpus(corpus_dir, dict_path, output_directory, temp_dir,
output_model_path, args):
if temp_dir == '':
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(temp_dir)
corpus_name = os.path.basename(corpus_dir)
if corpus_name == '':
corpus_dir = os.path.dirname(corpus_dir)
corpus_name = os.path.basename(corpus_dir)
data_directory = os.path.join(temp_dir, corpus_name)
if args.clean:
shutil.rmtree(data_directory, ignore_errors=True)
shutil.rmtree(output_directory, ignore_errors=True)
os.makedirs(data_directory, exist_ok=True)
os.makedirs(output_directory, exist_ok=True)
dictionary = Dictionary(dict_path, data_directory)
dictionary.write()
corpus = Corpus(corpus_dir,
data_directory,
args.speaker_characters,
num_jobs=args.num_jobs)
print(corpus.speaker_utterance_info())
corpus.write()
corpus.create_mfccs()
corpus.setup_splits(dictionary)
utt_oov_path = os.path.join(corpus.split_directory, 'utterance_oovs.txt')
if os.path.exists(utt_oov_path):
shutil.copy(utt_oov_path, output_directory)
oov_path = os.path.join(corpus.split_directory, 'oovs_found.txt')
if os.path.exists(oov_path):
shutil.copy(oov_path, output_directory)
mono_params = {'align_often': not args.fast}
tri_params = {'align_often': not args.fast}
tri_fmllr_params = {'align_often': not args.fast}
a = TrainableAligner(corpus,
dictionary,
output_directory,
temp_directory=data_directory,
mono_params=mono_params,
tri_params=tri_params,
tri_fmllr_params=tri_fmllr_params,
num_jobs=args.num_jobs)
a.verbose = args.verbose
a.train_mono()
a.export_textgrids()
a.train_tri()
a.export_textgrids()
a.train_tri_fmllr()
a.export_textgrids()
if output_model_path is not None:
a.save(output_model_path)
def align_corpus(model_path, corpus_dir, output_directory, temp_dir, args, debug = False):
all_begin = time.time()
if temp_dir == '':
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(temp_dir)
corpus_name = os.path.basename(corpus_dir)
if corpus_name == '':
corpus_dir = os.path.dirname(corpus_dir)
corpus_name = os.path.basename(corpus_dir)
data_directory = os.path.join(temp_dir, corpus_name)
if args.clean:
shutil.rmtree(data_directory, ignore_errors = True)
shutil.rmtree(output_directory, ignore_errors = True)
os.makedirs(data_directory, exist_ok = True)
os.makedirs(output_directory, exist_ok = True)
begin = time.time()
corpus = Corpus(corpus_dir, data_directory, args.speaker_characters, num_jobs = args.num_jobs)
print(corpus.speaker_utterance_info())
corpus.write()
if debug:
print('Wrote corpus information in {} seconds'.format(time.time() - begin))
begin = time.time()
corpus.create_mfccs()
if debug:
print('Calculated mfccs in {} seconds'.format(time.time() - begin))
archive = Archive(model_path)
begin = time.time()
a = PretrainedAligner(archive, corpus, output_directory,
temp_directory = data_directory, num_jobs = args.num_jobs, speaker_independent = args.no_speaker_adaptation)
if debug:
print('Setup pretrained aligner in {} seconds'.format(time.time() - begin))
a.verbose = args.verbose
begin = time.time()
corpus.setup_splits(a.dictionary)
if debug:
print('Setup splits in {} seconds'.format(time.time() - begin))
utt_oov_path = os.path.join(corpus.split_directory, 'utterance_oovs.txt')
if os.path.exists(utt_oov_path):
shutil.copy(utt_oov_path, output_directory)
oov_path = os.path.join(corpus.split_directory, 'oovs_found.txt')
if os.path.exists(oov_path):
shutil.copy(oov_path, output_directory)
begin = time.time()
a.do_align()
if debug:
print('Performed alignment in {} seconds'.format(time.time() - begin))
begin = time.time()
a.export_textgrids()
if debug:
print('Exported textgrids in {} seconds'.format(time.time() - begin))
print('Done! Everything took {} seconds'.format(time.time() - all_begin))
def align_corpus(corpus_dir, dict_path, output_directory, temp_dir,
output_model_path, args):
if temp_dir == '':
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(temp_dir)
corpus_name = os.path.basename(corpus_dir)
if corpus_name == '':
corpus_dir = os.path.dirname(corpus_dir)
corpus_name = os.path.basename(corpus_dir)
data_directory = os.path.join(temp_dir, corpus_name)
if args.clean:
shutil.rmtree(data_directory, ignore_errors = True)
shutil.rmtree(output_directory, ignore_errors = True)
os.makedirs(data_directory, exist_ok = True)
os.makedirs(output_directory, exist_ok = True)
corpus = Corpus(corpus_dir, data_directory, args.speaker_characters, num_jobs = args.num_jobs)
print(corpus.speaker_utterance_info())
corpus.write()
corpus.create_mfccs()
dictionary = Dictionary(dict_path, data_directory, word_set=corpus.word_set)
dictionary.write()
corpus.setup_splits(dictionary)
utt_oov_path = os.path.join(corpus.split_directory, 'utterance_oovs.txt')
if os.path.exists(utt_oov_path):
shutil.copy(utt_oov_path, output_directory)
oov_path = os.path.join(corpus.split_directory, 'oovs_found.txt')
if os.path.exists(oov_path):
shutil.copy(oov_path, output_directory)
mono_params = {'align_often': not args.fast}
tri_params = {'align_often': not args.fast}
tri_fmllr_params = {'align_often': not args.fast}
a = TrainableAligner(corpus, dictionary, output_directory,
temp_directory = data_directory,
mono_params = mono_params, tri_params = tri_params,
tri_fmllr_params = tri_fmllr_params, num_jobs = args.num_jobs)
a.verbose = args.verbose
a.train_mono()
a.export_textgrids()
a.train_tri()
a.export_textgrids()
a.train_tri_fmllr()
a.export_textgrids()
if output_model_path is not None:
a.save(output_model_path)
def test_extra(sick_dict, extra_dir, generated_dir):
output_directory = os.path.join(generated_dir, "extra")
corpus = Corpus(extra_dir, output_directory, num_jobs=2)
corpus.write()
corpus.create_mfccs()
corpus.setup_splits(sick_dict)