def __init__(self, args):
super(RawAudioDataset, self).__init__()
self._input_tarball = args["input_tarball"]
self._transcripts = None
self._translations = None
self._feature_extractor = build_feature_extractor(args)
try:
import soundfile
_ = soundfile
except ImportError:
raise ImportError(
"Please install soundfile with: pip3 install soundfile")
except OSError:
raise OSError("sndfile library not found. Please install with: "
"apt-get install libsndfile1")
try:
from pydub import AudioSegment
_ = AudioSegment
except ImportError:
raise ImportError("Please install pydub with: pip3 install pydub")
excluded_file = args["excluded_file"]
self._excluded_str = None
if excluded_file is not None:
if not tf.io.gfile.exists(excluded_file):
raise ValueError(f"`excluded_file` not found: {excluded_file}")
with tf.io.gfile.GFile(excluded_file) as fp:
self._excluded_str = " ".join(
set([x.strip().lower() for x in fp]))
def _main(_):
# define and parse program flags
arg_parser = flags_core.define_flags(FLAG_LIST, with_config_file=True)
args, remaining_argv = flags_core.intelligent_parse_flags(
FLAG_LIST, arg_parser)
flags_core.verbose_flags(FLAG_LIST, args, remaining_argv)
dataset = build_dataset(args)
feature_extractor = build_feature_extractor(args)
if dataset is None:
raise ValueError("dataset must be provided.")
main(dataset, feature_extractor)
def __init__(self, args):
super(RawAudioDataset, self).__init__()
self._input_tarball = args["input_tarball"]
self._transcripts = None
self._translations = None
self._feature_extractor = build_feature_extractor(args)
try:
import soundfile
_ = soundfile
except ImportError:
raise ImportError(
"Please install soundfile with: pip3 install soundfile")
except OSError:
raise OSError("sndfile library not found. Please install with: "
"apt-get install libsndfile1")
def __init__(self, args):
super(RawAudioDataset, self).__init__()
self._input_tarball = args["input_tarball"]
self._transcripts = None
self._translations = None
self._feature_extractor = build_feature_extractor(args)
try:
import sox
self._sox_transformer = sox.Transformer()
self._sox_transformer.set_output_format(rate=16000,
file_type="wav")
except (ImportError, ModuleNotFoundError):
self._sox_transformer = None
excluded_file = args["excluded_file"]
self._excluded_str = None
if excluded_file is not None:
if not tf.io.gfile.exists(excluded_file):
raise ValueError(f"`excluded_file` not found: {excluded_file}")
with tf.io.gfile.GFile(excluded_file) as fp:
self._excluded_str = " ".join(
set([x.strip().lower() for x in fp]))