def __init__(self,
root_dir,
audio_query="*.h5",
mel_query="*.h5",
audio_load_fn=lambda x: read_hdf5(x, "wave"),
mel_load_fn=lambda x: read_hdf5(x, "feats"),
audio_length_threshold=None,
mel_length_threshold=None,
return_filename=False,
):
"""Initialize dataset.
Args:
root_dir (str): Root directory including dumped files.
audio_query (str): Query to find audio files in root_dir.
mel_query (str): Query to find feature files in root_dir.
audio_load_fn (func): Function to load audio file.
mel_load_fn (func): Function to load feature file.
audio_length_threshold (int): Threshold to remove short audio files.
mel_length_threshold (int): Threshold to remove short feature files.
return_filename (bool): Whether to return the filename with arrays.
"""
# find all of audio and mel files
audio_files = sorted(find_files(root_dir, audio_query))
mel_files = sorted(find_files(root_dir, mel_query))
# filter by threshold
if audio_length_threshold is not None:
audio_lengths = [audio_load_fn(f).shape[0] for f in audio_files]
idxs = [idx for idx in range(len(audio_files)) if audio_lengths[idx] > audio_length_threshold]
if len(audio_files) != len(idxs):
logging.info(f"some files are filtered by audio length threshold "
f"({len(audio_files)} -> {len(idxs)}).")
audio_files = [audio_files[idx] for idx in idxs]
mel_files = [mel_files[idx] for idx in idxs]
if mel_length_threshold is not None:
mel_lengths = [mel_load_fn(f).shape[0] for f in mel_files]
idxs = [idx for idx in range(len(mel_files)) if mel_lengths[idx] > mel_length_threshold]
if len(mel_files) != len(idxs):
logging.info(f"some files are filtered by mel length threshold "
f"({len(mel_files)} -> {len(idxs)}).")
audio_files = [audio_files[idx] for idx in idxs]
mel_files = [mel_files[idx] for idx in idxs]
# assert the number of files
assert len(audio_files) != 0, f"Not found any audio files in ${root_dir}."
assert len(audio_files) == len(mel_files), \
f"Number of audio and mel files are different ({len(audio_files)} vs {len(mel_files)})."
self.audio_files = audio_files
self.mel_files = mel_files
self.audio_load_fn = audio_load_fn
self.mel_load_fn = mel_load_fn
self.return_filename = return_filename
def __init__(
self,
root_dir,
mel_query="*-feats.npy",
mel_length_threshold=None,
mel_load_fn=np.load,
return_utt_id=False,
allow_cache=False,
):
"""Initialize dataset.
Args:
root_dir (str): Root directory including dumped files.
mel_query (str): Query to find feature files in root_dir.
mel_load_fn (func): Function to load feature file.
mel_length_threshold (int): Threshold to remove short feature files.
return_utt_id (bool): Whether to return the utterance id with arrays.
allow_cache (bool): Whether to allow cache of the loaded files.
"""
# find all of the mel files
mel_files = sorted(find_files(root_dir, mel_query))
# filter by threshold
if mel_length_threshold is not None:
mel_lengths = [mel_load_fn(f).shape[0] for f in mel_files]
idxs = [
idx for idx in range(len(mel_files))
if mel_lengths[idx] > mel_length_threshold
]
if len(mel_files) != len(idxs):
logging.warning(
f"Some files are filtered by mel length threshold "
f"({len(mel_files)} -> {len(idxs)}).")
mel_files = [mel_files[idx] for idx in idxs]
# assert the number of files
assert len(mel_files) != 0, f"Not found any mel files in ${root_dir}."
self.mel_files = mel_files
self.mel_load_fn = mel_load_fn
self.utt_ids = [
os.path.splitext(os.path.basename(f))[0] for f in mel_files
]
if ".npy" in mel_query:
self.utt_ids = [
os.path.basename(f).replace("-feats.npy", "")
for f in mel_files
]
else:
self.utt_ids = [
os.path.splitext(os.path.basename(f))[0] for f in mel_files
]
self.return_utt_id = return_utt_id
self.allow_cache = allow_cache
if allow_cache:
# NOTE(kan-bayashi): Manager is need to share memory in dataloader with num_workers > 0
self.manager = Manager()
self.caches = self.manager.list()
self.caches += [() for _ in range(len(mel_files))]
def __init__(self,
root_dir,
mel_query="*-feats.npy",
mel_length_threshold=None,
mel_load_fn=np.load,
return_filename=False,
):
"""Initialize dataset.
Args:
root_dir (str): Root directory including dumped files.
mel_query (str): Query to find feature files in root_dir.
mel_load_fn (func): Function to load feature file.
mel_length_threshold (int): Threshold to remove short feature files.
return_filename (bool): Whether to return the filename with arrays.
"""
# find all of the mel files
mel_files = sorted(find_files(root_dir, mel_query))
# filter by threshold
if mel_length_threshold is not None:
mel_lengths = [mel_load_fn(f).shape[0] for f in mel_files]
idxs = [idx for idx in range(len(mel_files)) if mel_lengths[idx] > mel_length_threshold]
if len(mel_files) != len(idxs):
logging.info(f"some files are filtered by mel length threshold "
f"({len(mel_files)} -> {len(idxs)}).")
mel_files = [mel_files[idx] for idx in idxs]
# assert the number of files
assert len(mel_files) != 0, f"Not found any mel files in ${root_dir}."
self.mel_files = mel_files
self.mel_load_fn = mel_load_fn
self.return_filename = return_filename
def main():
parser = argparse.ArgumentParser(
description=
'Convert filter banks to waveform using Griffin-Lim algorithm',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--conf',
type=str,
required=True,
help='Cofiguration file')
parser.add_argument('--rootdir',
type=str,
required=True,
help='Root directory of filter bank h5 files')
parser.add_argument('--outdir',
type=str,
required=True,
help='Output directory')
args = parser.parse_args()
# logging info
logging.basicConfig(
level=logging.INFO,
stream=sys.stdout,
format="%(asctime)s (%(module)s:%(lineno)d) "
"%(levelname)s: %(message)s",
)
# load configure files
conf = load_yaml(args.conf)
for k, v in conf.items():
logging.info("{}: {}".format(k, v))
# find h5 files
feats_files = find_files(args.rootdir, "*.h5")
feats = {
os.path.join(args.outdir,
os.path.basename(filename).replace(".h5", ".wav")):
read_feature(filename, "feats")
for filename in feats_files
}
# Main Griffin-Lim algorithm
Parallel(n_jobs=30)([
delayed(mlfb2wavf)(
feats[wavf],
wavf,
fs=conf["feature"]["fs"],
n_mels=conf["feature"]["mlfb_dim"],
fftl=conf["feature"]["fftl"],
hop_size=conf["feature"]["hop_size"],
plot=False,
) for wavf in list(feats.keys())
])
def __init__(
self,
root_dir,
audio_query="*-wave.npy",
audio_length_threshold=None,
audio_load_fn=np.load,
return_filename=False,
allow_cache=False,
):
"""Initialize dataset.
Args:
root_dir (str): Root directory including dumped files.
audio_query (str): Query to find audio files in root_dir.
audio_load_fn (func): Function to load audio file.
audio_length_threshold (int): Threshold to remove short audio files.
return_filename (bool): Whether to return the filename with arrays.
allow_cache (bool): Whether to allow cache of the loaded files.
"""
# find all of audio and mel files
audio_files = sorted(find_files(root_dir, audio_query))
# filter by threshold
if audio_length_threshold is not None:
audio_lengths = [audio_load_fn(f).shape[0] for f in audio_files]
idxs = [
idx for idx in range(len(audio_files))
if audio_lengths[idx] > audio_length_threshold
]
if len(audio_files) != len(idxs):
logging.waning(
f"some files are filtered by audio length threshold "
f"({len(audio_files)} -> {len(idxs)}).")
audio_files = [audio_files[idx] for idx in idxs]
# assert the number of files
assert len(
audio_files) != 0, f"Not found any audio files in ${root_dir}."
self.audio_files = audio_files
self.audio_load_fn = audio_load_fn
self.return_filename = return_filename
self.allow_cache = allow_cache
if allow_cache:
# NOTE(kan-bayashi): Manager is need to share memory in dataloader with num_workers > 0
self.manager = Manager()
self.caches = self.manager.list()
self.caches += [() for _ in range(len(audio_files))]
def __init__(self,
root_dir,
audio_query="*.h5",
mel_query="*.h5",
audio_load_fn=lambda x: read_hdf5(x, "wave"),
mel_load_fn=lambda x: read_hdf5(x, "feats"),
audio_length_threshold=None,
mel_length_threshold=None,
return_utt_id=False,
return_spk_id=False,
allow_cache=False,
):
"""Initialize dataset.
Args:
root_dir (str): Root directory including dumped files.
audio_query (str): Query to find audio files in root_dir.
mel_query (str): Query to find feature files in root_dir.
audio_load_fn (func): Function to load audio file.
mel_load_fn (func): Function to load feature file.
audio_length_threshold (int): Threshold to remove short audio files.
mel_length_threshold (int): Threshold to remove short feature files.
return_utt_id (bool): Whether to return the utterance id with arrays.
allow_cache (bool): Whether to allow cache of the loaded files.
"""
# find all of audio and mel files
audio_files = sorted(find_files(root_dir, audio_query))
mel_files = sorted(find_files(root_dir, mel_query))
# filter by threshold
if audio_length_threshold is not None:
audio_lengths = [audio_load_fn(f).shape[0] for f in audio_files]
idxs = [idx for idx in range(len(audio_files)) if audio_lengths[idx] > audio_length_threshold]
if len(audio_files) != len(idxs):
logging.warning(f"Some files are filtered by audio length threshold "
f"({len(audio_files)} -> {len(idxs)}).")
audio_files = [audio_files[idx] for idx in idxs]
mel_files = [mel_files[idx] for idx in idxs]
if mel_length_threshold is not None:
mel_lengths = [mel_load_fn(f).shape[0] for f in mel_files]
idxs = [idx for idx in range(len(mel_files)) if mel_lengths[idx] > mel_length_threshold]
if len(mel_files) != len(idxs):
logging.warning(f"Some files are filtered by mel length threshold "
f"({len(mel_files)} -> {len(idxs)}).")
audio_files = [audio_files[idx] for idx in idxs]
mel_files = [mel_files[idx] for idx in idxs]
# assert the number of files
assert len(audio_files) != 0, f"Not found any audio files in ${root_dir}."
assert len(audio_files) == len(mel_files), \
f"Number of audio and mel files are different ({len(audio_files)} vs {len(mel_files)})."
self.audio_files = audio_files
self.audio_load_fn = audio_load_fn
self.mel_load_fn = mel_load_fn
self.mel_files = mel_files
if ".npy" in audio_query:
self.utt_ids = [os.path.basename(f).replace("-wave.npy", "") for f in audio_files]
else:
self.utt_ids = [os.path.splitext(os.path.basename(f))[0] for f in audio_files]
# only meaningful for vctk
self.return_utt_id = return_utt_id
self.spk_ids = list(set([self.get_spk_id(s) for s in self.utt_ids]))
self.spk_ids.sort()
#self.spk_id_encoder = LabelBinarizer().fit(self.spk_ids)
self.spk_id_encoder = LabelEncoder().fit(self.spk_ids)
self.return_spk_id = return_spk_id
self.allow_cache = allow_cache
if allow_cache:
# NOTE(kan-bayashi): Manager is need to share memory in dataloader with num_workers > 0
self.manager = Manager()
self.caches = self.manager.list()
self.caches += [() for _ in range(len(audio_files))]
