def validate_corpus(args):
command = 'validate'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
data_directory = os.path.join(temp_dir, corpus_name)
shutil.rmtree(data_directory, ignore_errors=True)
os.makedirs(data_directory, exist_ok=True)
logger = setup_logger(command, data_directory)
corpus = AlignableCorpus(args.corpus_directory,
data_directory,
speaker_characters=args.speaker_characters,
num_jobs=getattr(args, 'num_jobs', 3),
logger=logger,
use_mp=not args.disable_mp)
dictionary = Dictionary(args.dictionary_path,
data_directory,
logger=logger)
if args.acoustic_model_path:
acoustic_model = AcousticModel(args.acoustic_model_path)
acoustic_model.validate(dictionary)
a = CorpusValidator(corpus,
dictionary,
temp_directory=data_directory,
ignore_acoustics=getattr(args, 'ignore_acoustics',
False),
test_transcriptions=getattr(args,
'test_transcriptions',
False),
use_mp=not args.disable_mp,
logger=logger)
begin = time.time()
a.validate()
logger.debug('Validation took {} seconds'.format(time.time() - begin))
logger.info('All done!')
logger.debug('Done! Everything took {} seconds'.format(time.time() -
all_begin))
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
def train_lm(args):
command = 'train_lm'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
if args.config_path:
train_config = train_lm_yaml_to_config(args.config_path)
else:
train_config = load_basic_train_lm()
corpus_name = os.path.basename(args.source_path)
if corpus_name == '':
args.source_path = os.path.dirname(args.source_path)
corpus_name = os.path.basename(args.source_path)
source = args.source_path
dictionary = None
if args.source_path.lower().endswith('.arpa'):
corpus_name = os.path.splitext(corpus_name)[0]
data_directory = os.path.join(temp_dir, corpus_name)
else:
data_directory = os.path.join(temp_dir, corpus_name)
logger = setup_logger(command, data_directory)
if not args.source_path.lower().endswith('.arpa'):
source = AlignableCorpus(args.source_path, data_directory, num_jobs=args.num_jobs, use_mp=args.num_jobs>1)
if args.dictionary_path is not None:
dictionary = Dictionary(args.dictionary_path, data_directory)
else:
dictionary = None
trainer = LmTrainer(source, train_config, args.output_model_path, dictionary=dictionary,
temp_directory=data_directory,
supplemental_model_path=args.model_path, supplemental_model_weight=args.model_weight)
begin = time.time()
trainer.train()
logger.debug('Training took {} seconds'.format(time.time() - begin))
logger.info('All done!')
logger.debug('Done! Everything took {} seconds'.format(time.time() - all_begin))
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
def transcribe_corpus(args):
command = 'transcribe'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
if args.config_path:
transcribe_config = transcribe_yaml_to_config(args.config_path)
else:
transcribe_config = load_basic_transcribe()
data_directory = os.path.join(temp_dir, corpus_name)
if getattr(args, 'clean', False) and os.path.exists(data_directory):
print('Cleaning old directory!')
shutil.rmtree(data_directory, ignore_errors=True)
logger = setup_logger(command, data_directory)
os.makedirs(data_directory, exist_ok=True)
os.makedirs(args.output_directory, exist_ok=True)
os.makedirs(data_directory, exist_ok=True)
conf_path = os.path.join(data_directory, 'config.yml')
if os.path.exists(conf_path):
with open(conf_path, 'r') as f:
conf = yaml.load(f, Loader=yaml.SafeLoader)
else:
conf = {
'dirty': False,
'begin': time.time(),
'version': __version__,
'type': 'transcribe',
'corpus_directory': args.corpus_directory,
'dictionary_path': args.dictionary_path,
'acoustic_model_path': args.acoustic_model_path,
'language_model_path': args.language_model_path,
}
if conf['dirty'] or conf['type'] != command \
or conf['corpus_directory'] != args.corpus_directory \
or conf['version'] != __version__ \
or conf['dictionary_path'] != args.dictionary_path \
or conf['language_model_path'] != args.language_model_path \
or conf['acoustic_model_path'] != args.acoustic_model_path:
logger.warning(
'WARNING: Using old temp directory, this might not be ideal for you, use the --clean flag to ensure no '
'weird behavior for previous versions of the temporary directory.')
if conf['dirty']:
logger.debug('Previous run ended in an error (maybe ctrl-c?)')
if conf['type'] != command:
logger.debug(
'Previous run was a different subcommand than {} (was {})'.
format(command, conf['type']))
if conf['corpus_directory'] != args.corpus_directory:
logger.debug('Previous run used source directory '
'path {} (new run: {})'.format(
conf['corpus_directory'], args.corpus_directory))
if conf['version'] != __version__:
logger.debug('Previous run was on {} version (new run: {})'.format(
conf['version'], __version__))
if conf['dictionary_path'] != args.dictionary_path:
logger.debug('Previous run used dictionary path {} '
'(new run: {})'.format(conf['dictionary_path'],
args.dictionary_path))
if conf['acoustic_model_path'] != args.acoustic_model_path:
logger.debug('Previous run used acoustic model path {} '
'(new run: {})'.format(conf['acoustic_model_path'],
args.acoustic_model_path))
if conf['language_model_path'] != args.language_model_path:
logger.debug('Previous run used language model path {} '
'(new run: {})'.format(conf['language_model_path'],
args.language_model_path))
try:
if args.evaluate:
corpus = AlignableCorpus(
args.corpus_directory,
data_directory,
speaker_characters=args.speaker_characters,
num_jobs=args.num_jobs,
use_mp=transcribe_config.use_mp)
else:
corpus = TranscribeCorpus(
args.corpus_directory,
data_directory,
speaker_characters=args.speaker_characters,
num_jobs=args.num_jobs,
use_mp=transcribe_config.use_mp)
print(corpus.speaker_utterance_info())
acoustic_model = AcousticModel(args.acoustic_model_path,
root_directory=data_directory)
language_model = LanguageModel(args.language_model_path,
root_directory=data_directory)
dictionary = Dictionary(args.dictionary_path, data_directory)
acoustic_model.validate(dictionary)
begin = time.time()
t = Transcriber(corpus,
dictionary,
acoustic_model,
language_model,
transcribe_config,
temp_directory=data_directory,
debug=getattr(args, 'debug', False),
evaluation_mode=args.evaluate)
if args.debug:
print('Setup pretrained aligner in {} seconds'.format(time.time() -
begin))
begin = time.time()
t.transcribe()
if args.debug:
print('Performed transcribing in {} seconds'.format(time.time() -
begin))
if args.evaluate:
t.evaluate(args.output_directory)
best_config_path = os.path.join(args.output_directory,
'best_transcribe_config.yaml')
save_config(t.transcribe_config, best_config_path)
t.export_transcriptions(args.output_directory)
else:
begin = time.time()
t.export_transcriptions(args.output_directory)
if args.debug:
print('Exported transcriptions in {} seconds'.format(
time.time() - begin))
print('Done! Everything took {} seconds'.format(time.time() -
all_begin))
except Exception as _:
conf['dirty'] = True
raise
finally:
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
if os.path.exists(data_directory):
with open(conf_path, 'w') as f:
yaml.dump(conf, f)
def train_ivector(args):
command = 'train_ivector'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
data_directory = os.path.join(temp_dir, corpus_name)
if args.config_path:
train_config, align_config = train_yaml_to_config(args.config_path)
else:
train_config, align_config = load_basic_train_ivector()
conf_path = os.path.join(data_directory, 'config.yml')
if getattr(args, 'clean', False) and os.path.exists(data_directory):
print('Cleaning old directory!')
shutil.rmtree(data_directory, ignore_errors=True)
logger = setup_logger(command, data_directory)
if os.path.exists(conf_path):
with open(conf_path, 'r') as f:
conf = yaml.load(f, Loader=yaml.SafeLoader)
else:
conf = {
'dirty': False,
'begin': all_begin,
'version': __version__,
'type': command,
'corpus_directory': args.corpus_directory,
'dictionary_path': args.dictionary_path,
'acoustic_model_path': args.acoustic_model_path,
}
if conf['dirty'] or conf['type'] != command \
or conf['corpus_directory'] != args.corpus_directory \
or conf['version'] != __version__ \
or conf['dictionary_path'] != args.dictionary_path \
or conf['acoustic_model_path'] != args.acoustic_model_path:
logger.warning(
'WARNING: Using old temp directory, this might not be ideal for you, use the --clean flag to ensure no '
'weird behavior for previous versions of the temporary directory.')
if conf['dirty']:
logger.debug('Previous run ended in an error (maybe ctrl-c?)')
if conf['type'] != command:
logger.debug(
'Previous run was a different subcommand than {} (was {})'.
format(command, conf['type']))
if conf['corpus_directory'] != args.corpus_directory:
logger.debug('Previous run used source directory '
'path {} (new run: {})'.format(
conf['corpus_directory'], args.corpus_directory))
if conf['version'] != __version__:
logger.debug('Previous run was on {} version (new run: {})'.format(
conf['version'], __version__))
if conf['dictionary_path'] != args.dictionary_path:
logger.debug('Previous run used dictionary path {} '
'(new run: {})'.format(conf['dictionary_path'],
args.dictionary_path))
if conf['acoustic_model_path'] != args.acoustic_model_path:
logger.debug('Previous run used acoustic model path {} '
'(new run: {})'.format(conf['acoustic_model_path'],
args.acoustic_model_path))
os.makedirs(data_directory, exist_ok=True)
try:
begin = time.time()
corpus = AlignableCorpus(args.corpus_directory,
data_directory,
speaker_characters=args.speaker_characters,
num_jobs=args.num_jobs,
debug=getattr(args, 'debug', False),
logger=logger,
use_mp=align_config.use_mp)
acoustic_model = AcousticModel(args.acoustic_model_path)
dictionary = Dictionary(args.dictionary_path,
data_directory,
word_set=corpus.word_set,
logger=logger)
acoustic_model.validate(dictionary)
a = PretrainedAligner(corpus,
dictionary,
acoustic_model,
align_config,
temp_directory=data_directory,
logger=logger)
logger.debug(
'Setup pretrained aligner in {} seconds'.format(time.time() -
begin))
a.verbose = args.verbose
begin = time.time()
a.align()
logger.debug('Performed alignment in {} seconds'.format(time.time() -
begin))
for identifier, trainer in train_config.items():
trainer.logger = logger
if identifier != 'ivector':
continue
begin = time.time()
trainer.init_training(identifier, data_directory, corpus,
dictionary, a)
trainer.train(call_back=print)
logger.debug('Training took {} seconds'.format(time.time() -
begin))
trainer.save(args.output_model_path)
logger.info('All done!')
logger.debug('Done! Everything took {} seconds'.format(time.time() -
all_begin))
except Exception as e:
conf['dirty'] = True
raise e
finally:
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
with open(conf_path, 'w') as f:
yaml.dump(conf, f)
def classify_speakers(args):
command = 'classify_speakers'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
data_directory = os.path.join(temp_dir, corpus_name)
conf_path = os.path.join(data_directory, 'config.yml')
if args.config_path:
classification_config = classification_yaml_to_config(args.config_path)
else:
classification_config = load_basic_classification()
classification_config.use_mp = not args.disable_mp
if getattr(args, 'clean', False) and os.path.exists(data_directory):
print('Cleaning old directory!')
shutil.rmtree(data_directory, ignore_errors=True)
logger = setup_logger(command, data_directory)
if os.path.exists(conf_path):
with open(conf_path, 'r') as f:
conf = yaml.load(f, Loader=yaml.SafeLoader)
else:
conf = {'dirty': False,
'begin': time.time(),
'version': __version__,
'type': command,
'corpus_directory': args.corpus_directory,
'ivector_extractor_path': args.ivector_extractor_path}
if conf['dirty'] or conf['type'] != command \
or conf['corpus_directory'] != args.corpus_directory \
or conf['version'] != __version__:
logger.warning(
'WARNING: Using old temp directory, this might not be ideal for you, use the --clean flag to ensure no '
'weird behavior for previous versions of the temporary directory.')
if conf['dirty']:
logger.debug('Previous run ended in an error (maybe ctrl-c?)')
if conf['type'] != command:
logger.debug('Previous run was a different subcommand than {} (was {})'.format(command, conf['type']))
if conf['corpus_directory'] != args.corpus_directory:
logger.debug('Previous run used source directory '
'path {} (new run: {})'.format(conf['corpus_directory'], args.corpus_directory))
if conf['version'] != __version__:
logger.debug('Previous run was on {} version (new run: {})'.format(conf['version'], __version__))
if conf['ivector_extractor_path'] != args.ivector_extractor_path:
logger.debug('Previous run used ivector extractor path {} '
'(new run: {})'.format(conf['ivector_extractor_path'], args.ivector_extractor_path))
os.makedirs(data_directory, exist_ok=True)
os.makedirs(args.output_directory, exist_ok=True)
try:
corpus = TranscribeCorpus(args.corpus_directory, data_directory,
num_jobs=args.num_jobs, logger=logger, use_mp=classification_config.use_mp)
ivector_extractor = IvectorExtractor(args.ivector_extractor_path, root_directory=data_directory)
begin = time.time()
a = SpeakerClassifier(corpus, ivector_extractor, classification_config,
temp_directory=data_directory,
debug=getattr(args, 'debug', False), logger=logger, num_speakers=args.num_speakers,
cluster=args.cluster)
logger.debug('Setup speaker classifier in {} seconds'.format(time.time() - begin))
a.verbose = args.verbose
begin = time.time()
a.classify()
logger.debug('Performed classification in {} seconds'.format(time.time() - begin))
begin = time.time()
a.export_classification(args.output_directory)
logger.debug('Exported classification in {} seconds'.format(time.time() - begin))
logger.info('Done!')
logger.debug('Done! Everything took {} seconds'.format(time.time() - all_begin))
except Exception as _:
conf['dirty'] = True
raise
finally:
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
with open(conf_path, 'w') as f:
yaml.dump(conf, f)
def train_dictionary(args):
command = 'train_dictionary'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
data_directory = os.path.join(temp_dir, corpus_name)
conf_path = os.path.join(data_directory, 'config.yml')
if args.config_path:
align_config = align_yaml_to_config(args.config_path)
else:
align_config = load_basic_align()
if getattr(args, 'clean', False) and os.path.exists(data_directory):
print('Cleaning old directory!')
shutil.rmtree(data_directory, ignore_errors=True)
logger = setup_logger(command, data_directory)
if os.path.exists(conf_path):
with open(conf_path, 'r') as f:
conf = yaml.load(f, Loader=yaml.SafeLoader)
else:
conf = {'dirty': False,
'begin': time.time(),
'version': __version__,
'type': command,
'corpus_directory': args.corpus_directory,
'dictionary_path': args.dictionary_path,
'acoustic_model_path': args.acoustic_model_path
}
if conf['dirty'] or conf['type'] != command \
or conf['corpus_directory'] != args.corpus_directory \
or conf['version'] != __version__ \
or conf['dictionary_path'] != args.dictionary_path:
logger.warning(
'WARNING: Using old temp directory, this might not be ideal for you, use the --clean flag to ensure no '
'weird behavior for previous versions of the temporary directory.')
if conf['dirty']:
logger.debug('Previous run ended in an error (maybe ctrl-c?)')
if conf['type'] != command:
logger.debug('Previous run was a different subcommand than {} (was {})'.format(command, conf['type']))
if conf['corpus_directory'] != args.corpus_directory:
logger.debug('Previous run used source directory '
'path {} (new run: {})'.format(conf['corpus_directory'], args.corpus_directory))
if conf['version'] != __version__:
logger.debug('Previous run was on {} version (new run: {})'.format(conf['version'], __version__))
if conf['dictionary_path'] != args.dictionary_path:
logger.debug('Previous run used dictionary path {} '
'(new run: {})'.format(conf['dictionary_path'], args.dictionary_path))
if conf['acoustic_model_path'] != args.acoustic_model_path:
logger.debug('Previous run used acoustic model path {} '
'(new run: {})'.format(conf['acoustic_model_path'], args.acoustic_model_path))
os.makedirs(data_directory, exist_ok=True)
try:
corpus = AlignableCorpus(args.corpus_directory, data_directory,
speaker_characters=args.speaker_characters,
num_jobs=args.num_jobs, use_mp=align_config.use_mp, logger=logger)
if corpus.issues_check:
logger.warning('WARNING: Some issues parsing the corpus were detected. '
'Please run the validator to get more information.')
logger.info(corpus.speaker_utterance_info())
acoustic_model = AcousticModel(args.acoustic_model_path)
dictionary = Dictionary(args.dictionary_path, data_directory, word_set=corpus.word_set, logger=logger)
acoustic_model.validate(dictionary)
begin = time.time()
a = PretrainedAligner(corpus, dictionary, acoustic_model, align_config,
temp_directory=data_directory,
debug=getattr(args, 'debug', False), logger=logger)
logger.debug('Setup pretrained aligner in {} seconds'.format(time.time() - begin))
a.verbose = args.verbose
begin = time.time()
a.align()
logger.debug('Performed alignment in {} seconds'.format(time.time() - begin))
a.generate_pronunciations(args.output_directory)
print('Done! Everything took {} seconds'.format(time.time() - all_begin))
except Exception as _:
conf['dirty'] = True
raise
finally:
with open(conf_path, 'w') as f:
yaml.dump(conf, f)
def align_corpus(args, unknown_args=None):
command = 'train_and_align'
all_begin = time.time()
if not args.temp_directory:
temp_dir = TEMP_DIR
else:
temp_dir = os.path.expanduser(args.temp_directory)
corpus_name = os.path.basename(args.corpus_directory)
if corpus_name == '':
args.corpus_directory = os.path.dirname(args.corpus_directory)
corpus_name = os.path.basename(args.corpus_directory)
data_directory = os.path.join(temp_dir, corpus_name)
logger = setup_logger(command, data_directory)
if args.config_path:
train_config, align_config = train_yaml_to_config(args.config_path)
else:
train_config, align_config = load_basic_train()
if unknown_args:
align_config.update_from_args(unknown_args)
conf_path = os.path.join(data_directory, 'config.yml')
if args.debug:
logger.warning(
'Running in DEBUG mode, may have impact on performance and disk usage.'
)
if getattr(args, 'clean', False) and os.path.exists(data_directory):
logger.info('Cleaning old directory!')
shutil.rmtree(data_directory, ignore_errors=True)
if os.path.exists(conf_path):
with open(conf_path, 'r') as f:
conf = yaml.load(f, Loader=yaml.SafeLoader)
else:
conf = {
'dirty': False,
'begin': time.time(),
'version': __version__,
'type': command,
'corpus_directory': args.corpus_directory,
'dictionary_path': args.dictionary_path
}
if conf['dirty'] or conf['type'] != command \
or conf['corpus_directory'] != args.corpus_directory \
or conf['version'] != __version__ \
or conf['dictionary_path'] != args.dictionary_path:
logger.warning(
'WARNING: Using old temp directory, this might not be ideal for you, use the --clean flag to ensure no '
'weird behavior for previous versions of the temporary directory.')
if conf['dirty']:
logger.debug('Previous run ended in an error (maybe ctrl-c?)')
if conf['type'] != command:
logger.debug(
'Previous run was a different subcommand than {} (was {})'.
format(command, conf['type']))
if conf['corpus_directory'] != args.corpus_directory:
logger.debug('Previous run used source directory '
'path {} (new run: {})'.format(
conf['corpus_directory'], args.corpus_directory))
if conf['version'] != __version__:
logger.debug('Previous run was on {} version (new run: {})'.format(
conf['version'], __version__))
if conf['dictionary_path'] != args.dictionary_path:
logger.debug('Previous run used dictionary path {} '
'(new run: {})'.format(conf['dictionary_path'],
args.dictionary_path))
os.makedirs(data_directory, exist_ok=True)
os.makedirs(args.output_directory, exist_ok=True)
try:
corpus = AlignableCorpus(args.corpus_directory,
data_directory,
speaker_characters=args.speaker_characters,
num_jobs=getattr(args, 'num_jobs', 3),
debug=getattr(args, 'debug', False),
logger=logger,
use_mp=align_config.use_mp)
if corpus.issues_check:
logger.warning('Some issues parsing the corpus were detected. '
'Please run the validator to get more information.')
logger.info(corpus.speaker_utterance_info())
dictionary = Dictionary(args.dictionary_path,
data_directory,
word_set=corpus.word_set,
logger=logger)
utt_oov_path = os.path.join(corpus.split_directory(),
'utterance_oovs.txt')
if os.path.exists(utt_oov_path):
shutil.copy(utt_oov_path, args.output_directory)
oov_path = os.path.join(corpus.split_directory(), 'oovs_found.txt')
if os.path.exists(oov_path):
shutil.copy(oov_path, args.output_directory)
a = TrainableAligner(corpus,
dictionary,
train_config,
align_config,
temp_directory=data_directory,
logger=logger,
debug=getattr(args, 'debug', False))
a.verbose = args.verbose
begin = time.time()
a.train()
logger.debug('Training took {} seconds'.format(time.time() - begin))
a.export_textgrids(args.output_directory)
if args.output_model_path is not None:
a.save(args.output_model_path)
logger.info('All done!')
logger.debug('Done! Everything took {} seconds'.format(time.time() -
all_begin))
except Exception as _:
conf['dirty'] = True
raise
finally:
handlers = logger.handlers[:]
for handler in handlers:
handler.close()
logger.removeHandler(handler)
with open(conf_path, 'w') as f:
yaml.dump(conf, f)