File size: 5,778 Bytes
ad16788 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 |
#!/usr/bin/env python3
import configargparse
from distutils.util import strtobool
import logging
import os
import random
import sys
import numpy as np
from espnet.asr.pytorch_backend.asr import enhance
# NOTE: you need this func to generate our sphinx doc
def get_parser():
parser = configargparse.ArgumentParser(
description="Enhance noisy speech for speech recognition",
config_file_parser_class=configargparse.YAMLConfigFileParser,
formatter_class=configargparse.ArgumentDefaultsHelpFormatter,
)
# general configuration
parser.add("--config", is_config_file=True, help="config file path")
parser.add(
"--config2",
is_config_file=True,
help="second config file path that overwrites the settings in `--config`.",
)
parser.add(
"--config3",
is_config_file=True,
help="third config file path that overwrites the settings "
"in `--config` and `--config2`.",
)
parser.add_argument("--ngpu", default=0, type=int, help="Number of GPUs")
parser.add_argument(
"--backend",
default="chainer",
type=str,
choices=["chainer", "pytorch"],
help="Backend library",
)
parser.add_argument("--debugmode", default=1, type=int, help="Debugmode")
parser.add_argument("--seed", default=1, type=int, help="Random seed")
parser.add_argument("--verbose", "-V", default=1, type=int, help="Verbose option")
parser.add_argument(
"--batchsize",
default=1,
type=int,
help="Batch size for beam search (0: means no batch processing)",
)
parser.add_argument(
"--preprocess-conf",
type=str,
default=None,
help="The configuration file for the pre-processing",
)
# task related
parser.add_argument(
"--recog-json", type=str, help="Filename of recognition data (json)"
)
# model (parameter) related
parser.add_argument(
"--model", type=str, required=True, help="Model file parameters to read"
)
parser.add_argument(
"--model-conf", type=str, default=None, help="Model config file"
)
# Outputs configuration
parser.add_argument(
"--enh-wspecifier",
type=str,
default=None,
help="Specify the output way for enhanced speech."
"e.g. ark,scp:outdir,wav.scp",
)
parser.add_argument(
"--enh-filetype",
type=str,
default="sound",
choices=["mat", "hdf5", "sound.hdf5", "sound"],
help="Specify the file format for enhanced speech. "
'"mat" is the matrix format in kaldi',
)
parser.add_argument("--fs", type=int, default=16000, help="The sample frequency")
parser.add_argument(
"--keep-length",
type=strtobool,
default=True,
help="Adjust the output length to match " "with the input for enhanced speech",
)
parser.add_argument(
"--image-dir", type=str, default=None, help="The directory saving the images."
)
parser.add_argument(
"--num-images",
type=int,
default=20,
help="The number of images files to be saved. "
"If negative, all samples are to be saved.",
)
# IStft
parser.add_argument(
"--apply-istft",
type=strtobool,
default=True,
help="Apply istft to the output from the network",
)
parser.add_argument(
"--istft-win-length",
type=int,
default=512,
help="The window length for istft. "
"This option is ignored "
"if stft is found in the preprocess-conf",
)
parser.add_argument(
"--istft-n-shift",
type=str,
default=256,
help="The window type for istft. "
"This option is ignored "
"if stft is found in the preprocess-conf",
)
parser.add_argument(
"--istft-window",
type=str,
default="hann",
help="The window type for istft. "
"This option is ignored "
"if stft is found in the preprocess-conf",
)
return parser
def main(args):
parser = get_parser()
args = parser.parse_args(args)
# logging info
if args.verbose == 1:
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
)
elif args.verbose == 2:
logging.basicConfig(
level=logging.DEBUG,
format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
)
else:
logging.basicConfig(
level=logging.WARN,
format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s",
)
logging.warning("Skip DEBUG/INFO messages")
# check CUDA_VISIBLE_DEVICES
if args.ngpu > 0:
cvd = os.environ.get("CUDA_VISIBLE_DEVICES")
if cvd is None:
logging.warning("CUDA_VISIBLE_DEVICES is not set.")
elif args.ngpu != len(cvd.split(",")):
logging.error("#gpus is not matched with CUDA_VISIBLE_DEVICES.")
sys.exit(1)
# TODO(kamo): support of multiple GPUs
if args.ngpu > 1:
logging.error("The program only supports ngpu=1.")
sys.exit(1)
# display PYTHONPATH
logging.info("python path = " + os.environ.get("PYTHONPATH", "(None)"))
# seed setting
random.seed(args.seed)
np.random.seed(args.seed)
logging.info("set random seed = %d" % args.seed)
# recog
logging.info("backend = " + args.backend)
if args.backend == "pytorch":
enhance(args)
else:
raise ValueError("Only pytorch is supported.")
if __name__ == "__main__":
main(sys.argv[1:])
|