def main():
"""Phonemize a text from command-line arguments"""
args = parse_args()
if args.version:
print(version.version())
return
# configure logging according to --verbose/--quiet options
verbosity = 'normal'
if args.verbose:
verbosity = 'verbose'
elif args.quiet:
verbosity = 'quiet'
log = logger.get_logger(verbosity=verbosity)
# configure input as a readable stream
streamin = args.input
if isinstance(streamin, str):
streamin = codecs.open(streamin, 'r', encoding='utf8')
log.debug('reading from %s', streamin.name)
# configure output as a writable stream
streamout = args.output
if isinstance(streamout, str):
streamout = codecs.open(streamout, 'w', 'utf8')
log.debug('writing to %s', streamout.name)
# configure the separator for phonemes, syllables and words.
sep = separator.Separator(phone=args.phone_separator,
syllable=args.syllable_separator,
word=args.word_separator)
log.debug('separator is %s', sep)
# load the input text (python2 optionnally needs an extra decode)
text = streamin.read()
try:
text = text.decode('utf8')
except (AttributeError, UnicodeEncodeError):
pass
# phonemize the input text
out = phonemize.phonemize(text,
language=args.language,
backend=args.backend,
separator=sep,
strip=args.strip,
with_stress=args.with_stress,
use_sampa=args.sampa,
language_switch=args.language_switch,
njobs=args.njobs,
logger=log)
if len(out):
streamout.write(out + '\n')
def main():
"""Phonemize a text from command-line arguments"""
args = parse_args()
if args.version:
print(version.version())
return
# configure logging according to --verbose option. Init a logger to output
# on stderr. By default log only warnings, if verbose log all messages, if
# quiet do not log anything.
logger = logging.getLogger()
logger.handlers = []
handler = logging.StreamHandler(sys.stderr)
logger.setLevel(logging.WARNING)
if args.verbose:
logger.setLevel(logging.DEBUG)
if args.quiet:
handler = logging.NullHandler()
handler.setFormatter(logging.Formatter('[%(levelname)s] %(message)s'))
logger.addHandler(handler)
# configure input as a readable stream
streamin = args.input
if isinstance(streamin, str):
streamin = codecs.open(streamin, 'r', encoding='utf8')
logger.debug('reading from %s', streamin.name)
# configure output as a writable stream
streamout = args.output
if isinstance(streamout, str):
streamout = codecs.open(streamout, 'w', 'utf8')
logger.debug('writing to %s', streamout.name)
# configure the separator for phonemes, syllables and words.
sep = separator.Separator(
phone=args.phone_separator,
syllable=args.syllable_separator,
word=args.word_separator)
logger.debug('separator is %s', sep)
# load the input text (python2 optionnally needs an extra decode)
text = streamin.read()
try:
text = text.decode('utf8')
except (AttributeError, UnicodeEncodeError):
pass
# phonemize the input text
out = phonemize.phonemize(
text,
language=args.language,
backend=args.backend,
separator=sep,
strip=args.strip,
with_stress=args.with_stress,
use_sampa=args.sampa,
language_switch=args.language_switch,
njobs=args.njobs,
logger=logger)
if len(out):
streamout.write(out + '\n')
def main():
"""Phonemize a text from command-line arguments"""
args = parse_args()
# setup a custom path to espeak and festival if required (this must be done
# before generating the version message)
if args.espeak_path:
EspeakBackend.set_espeak_path(args.espeak_path)
if args.festival_path:
FestivalBackend.set_festival_path(args.festival_path)
# display version information and exit
if args.version:
print(version.version())
return
# list supported languages and exit
if args.list_languages:
backends = (['festival', 'segments', 'espeak', 'espeak-mbrola']
if not args.backend else [args.backend])
for backend in backends:
print(f'supported languages for {backend} are:\n' +
'\n'.join(f'\t{k}\t->\t{v}' for k, v in sorted(
BACKENDS_MAP[backend].supported_languages().items())))
return
# set default backend as espeak if not specified
args.backend = args.backend or 'espeak'
# configure logging according to --verbose/--quiet options
verbosity = 'normal'
if args.verbose:
verbosity = 'verbose'
elif args.quiet:
verbosity = 'quiet'
log = logger.get_logger(verbosity=verbosity)
# configure input as a readable stream
streamin = args.input
if isinstance(streamin, str):
streamin = codecs.open(streamin, 'r', encoding='utf8')
log.debug('reading from %s', streamin.name)
# configure output as a writable stream
streamout = args.output
if isinstance(streamout, str):
streamout = codecs.open(streamout, 'w', 'utf8')
log.debug('writing to %s', streamout.name)
# configure the separator for phonemes, syllables and words.
if args.backend == 'espeak-mbrola':
log.debug('using espeak-mbrola backend: ignoring word separator')
sep = separator.Separator(phone=args.phone_separator,
syllable=None,
word=None)
else:
sep = separator.Separator(phone=args.phone_separator,
syllable=args.syllable_separator,
word=args.word_separator)
log.debug('separator is %s', sep)
text = [line.strip() for line in streamin]
# phonemize the input text
out = phonemize(text,
language=args.language,
backend=args.backend,
separator=sep,
strip=args.strip,
preserve_punctuation=args.preserve_punctuation,
punctuation_marks=args.punctuation_marks,
with_stress=args.with_stress,
language_switch=args.language_switch,
njobs=args.njobs,
logger=log)
if out:
streamout.write('\n'.join(out) + '\n')
def main():
"""Phonemize a text from command-line arguments"""
args = parse_args()
# setup a custom path to espeak and festival if required (this must be done
# before generating the version message)
if args.espeak_library:
BACKENDS['espeak'].set_library(args.espeak_library)
if args.festival_executable:
BACKENDS['festival'].set_executable(args.festival_executable)
# display version information and exit
if args.version:
print(version.version())
return
# list supported languages and exit
if args.list_languages:
print(list_languages(args.backend))
return
# set default backend as espeak if not specified
args.backend = args.backend or 'espeak'
# configure logging according to --verbose/--quiet options
log = get_logger(args.verbose, args.quiet)
# configure input:output as a readable/writable streams
streamin = setup_stream(args.input, 'r')
log.debug('reading from %s', streamin.name)
streamout = setup_stream(args.output, 'w')
log.debug('writing to %s', streamout.name)
# configure the separator for phonemes, syllables and words.
if args.backend == 'espeak-mbrola':
log.debug('using espeak-mbrola backend: ignoring word separator')
sep = separator.Separator(phone=args.phone_separator,
syllable=None,
word=None)
else:
sep = separator.Separator(phone=args.phone_separator,
syllable=args.syllable_separator,
word=args.word_separator)
log.debug('separator is %s', sep)
if args.prepend_text:
input_output_separator = sep.input_output_separator(args.prepend_text)
log.debug('prepend input text to output, separator is "%s"',
input_output_separator)
else:
input_output_separator = False
# phonemize the input text
out = phonemize(streamin.readlines(),
language=args.language,
backend=args.backend,
separator=sep,
strip=args.strip,
prepend_text=args.prepend_text,
preserve_punctuation=args.preserve_punctuation,
punctuation_marks=args.punctuation_marks,
with_stress=args.with_stress,
tie=args.tie,
language_switch=args.language_switch,
words_mismatch=args.words_mismatch,
njobs=args.njobs,
logger=log)
if out and input_output_separator:
streamout.write(
os.linesep.join(f'{line[0]} {input_output_separator} {line[1]}'
for line in out) + os.linesep)
elif out:
streamout.write(os.linesep.join(out) + os.linesep)