Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
def convert(jsonf, dic, refs, hyps, num_spkrs=1):
n_ref = len(refs)
n_hyp = len(hyps)
assert n_ref == n_hyp
assert n_ref == num_spkrs
# logging info
logfmt = '%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s'
logging.basicConfig(level=logging.INFO, format=logfmt)
logging.info(get_commandline_args())
logging.info("reading %s", jsonf)
with codecs.open(jsonf, 'r', encoding="utf-8") as f:
j = json.load(f)
logging.info("reading %s", dic)
with codecs.open(dic, 'r', encoding="utf-8") as f:
dictionary = f.readlines()
char_list = [entry.split(' ')[0] for entry in dictionary]
char_list.insert(0, '')
char_list.append('')
for ns in range(num_spkrs):
hyp_file = codecs.open(hyps[ns], 'w', encoding="utf-8")
ref_file = codecs.open(refs[ns], 'w', encoding="utf-8")
def main():
parser = get_parser()
args = parser.parse_args()
logfmt = "%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s"
if args.verbose > 0:
logging.basicConfig(level=logging.INFO, format=logfmt)
else:
logging.basicConfig(level=logging.WARN, format=logfmt)
logging.info(get_commandline_args())
if len(args.reffiles) != len(args.enhfiles):
raise RuntimeError(
'The number of ref files are different '
'from the enh files: {} != {}'.format(len(args.reffiles),
len(args.enhfiles)))
if len(args.enhfiles) == 1:
args.permutation = False
# Read text files and created a mapping of key2filepath
reffiles_dict = OrderedDict() # Dict[str, Dict[str, str]]
for ref in args.reffiles:
d = OrderedDict()
with open(ref, 'r') as f:
for line in f:
key, path = line.split(None, 1)
d[key] = path.rstrip()
help='Verbose option')
parser.add_argument('--normalize', choices=[1, 16, 24, 32], type=int,
default=None,
help='Give the bit depth of the PCM, '
'then normalizes data to scale in [-1,1]')
parser.add_argument('rspecifier', type=str, help='WAV scp file')
parser.add_argument('wspecifier', type=str, help='Segments file')
args = parser.parse_args()
# set logger
logfmt = "%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s"
if args.verbose > 0:
logging.basicConfig(level=logging.INFO, format=logfmt)
else:
logging.basicConfig(level=logging.WARN, format=logfmt)
logging.info(get_commandline_args())
with kaldiio.ReadHelper(args.rspecifier) as reader, \
codecs.open(args.wspecifier, "w", encoding="utf-8") as f:
for utt_id, (rate, array) in reader:
assert rate == args.fs
array = array.astype(numpy.float32)
if args.normalize is not None and args.normalize != 1:
array = array / (1 << (args.normalize - 1))
array_trim, idx = librosa.effects.trim(
y=array,
top_db=args.threshold,
frame_length=args.win_length,
hop_length=args.shift_length
)
start, end = idx / args.fs
def main():
parser = get_parser()
args = parser.parse_args()
# logging info
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s")
logging.info(get_commandline_args())
# check directory
if not os.path.exists(args.outdir):
os.makedirs(args.outdir)
for idx, (utt_id, lmspc) in enumerate(
file_reader_helper(args.rspecifier, args.filetype), 1):
if args.n_mels is not None:
spc = logmelspc_to_linearspc(
lmspc,
fs=args.fs,
n_mels=args.n_mels,
n_fft=args.n_fft,
fmin=args.fmin,
fmax=args.fmax)
else:
def main():
args = get_parser().parse_args()
# logging info
logfmt = "%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s"
if args.verbose > 0:
logging.basicConfig(level=logging.INFO, format=logfmt)
else:
logging.basicConfig(level=logging.WARN, format=logfmt)
logging.info(get_commandline_args())
if ':' in args.stats_rspecifier_or_rxfilename:
is_rspcifier = True
if args.stats_filetype == 'npy':
stats_filetype = 'hdf5'
else:
stats_filetype = args.stats_filetype
stats_dict = dict(file_reader_helper(
args.stats_rspecifier_or_rxfilename, stats_filetype))
else:
is_rspcifier = False
if args.stats_filetype == 'mat':
stats = kaldiio.load_mat(args.stats_rspecifier_or_rxfilename)
else:
stats = numpy.load(args.stats_rspecifier_or_rxfilename)
from espnet.utils.cli_utils import get_commandline_args
is_python2 = sys.version_info[0] == 2
if __name__ == '__main__':
parser = argparse.ArgumentParser(
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('jsons', type=str, nargs='+',
help='json files')
args = parser.parse_args()
# logging info
logfmt = '%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s'
logging.basicConfig(level=logging.INFO, format=logfmt)
logging.info(get_commandline_args())
# make intersection set for utterance keys
num_keys = 0
js = {}
for i, x in enumerate(args.jsons):
with codecs.open(x, encoding="utf-8") as f:
j = json.load(f)
ks = j['utts'].keys()
logging.debug(x + ': has ' + str(len(ks)) + ' utterances')
num_keys += len(ks)
if i > 0:
for k in ks:
js[k + '.' + str(i)] = j['utts'][k]
else:
js = j['utts']
def convert(jsonf, refs, hyps, num_spkrs=1):
n_ref = len(refs)
n_hyp = len(hyps)
assert n_ref == n_hyp
assert n_ref == num_spkrs
# logging info
logfmt = '%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s'
logging.basicConfig(level=logging.INFO, format=logfmt)
logging.info(get_commandline_args())
logging.info("reading %s", jsonf)
with codecs.open(jsonf, 'r', encoding="utf-8") as f:
j = json.load(f)
for ns in range(num_spkrs):
hyp_file = codecs.open(hyps[ns], 'w', encoding="utf-8")
ref_file = codecs.open(refs[ns], 'w', encoding="utf-8")
for x in j['utts']:
# recognition hypothesis
if num_spkrs == 1:
seq = j['utts'][x]['output'][0]['rec_text'].replace('', '')
else:
seq = j['utts'][x]['output'][ns][0]['rec_text'].replace('', '')
# In the recognition hypothesis, the symbol is usually attached in the last part of the sentence
def main():
parser = get_parser()
args = parser.parse_args()
logfmt = "%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s"
if args.verbose > 0:
logging.basicConfig(level=logging.INFO, format=logfmt)
else:
logging.basicConfig(level=logging.WARN, format=logfmt)
logging.info(get_commandline_args())
with kaldiio.ReadHelper(args.rspecifier,
segments=args.segments) as reader, \
file_writer_helper(args.wspecifier,
filetype=args.filetype,
write_num_frames=args.write_num_frames,
compress=args.compress,
compression_method=args.compression_method
) as writer:
for utt_id, (_, array) in reader:
array = array.astype(numpy.float32)
if args.normalize is not None and args.normalize != 1:
array = array / (1 << (args.normalize - 1))
spc = spectrogram(
x=array,
n_fft=args.n_fft,
def main():
logfmt = '%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s'
logging.basicConfig(level=logging.INFO, format=logfmt)
logging.info(get_commandline_args())
parser = argparse.ArgumentParser(
description='Create waves list from "wav.scp"',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('scp')
parser.add_argument('outdir')
parser.add_argument('--name', default='wav',
help='Specify the prefix word of output file name '
'such as "wav.scp"')
parser.add_argument('--segments', default=None)
parser.add_argument('--fs', type=humanfriendly_or_none, default=None,
help='If the sampling rate specified, '
'Change the sampling rate.')
parser.add_argument('--audio-format', default='wav')
group = parser.add_mutually_exclusive_group()
group.add_argument('--ref-channels', default=None, type=str2int_tuple)