Spaces:
Running
on
Zero
Running
on
Zero
File size: 3,348 Bytes
568e264 |
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 |
# Copyright (c) 2023 Binbin Zhang ([email protected])
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import argparse
from wenet.cli.paraformer_model import load_model as load_paraformer
from wenet.cli.model import load_model
def get_args():
parser = argparse.ArgumentParser(description='')
parser.add_argument('audio_file', help='audio file to transcribe')
parser.add_argument('-l',
'--language',
choices=[
'chinese',
'english',
],
default='chinese',
help='language type')
parser.add_argument('-m',
'--model_dir',
default=None,
help='specify your own model dir')
parser.add_argument('-g',
'--gpu',
type=int,
default='-1',
help='gpu id to decode, default is cpu.')
parser.add_argument('--device',
type=str,
default='cpu',
choices=["cpu", "npu", "cuda"],
help='accelerator to use')
parser.add_argument('-t',
'--show_tokens_info',
action='store_true',
help='whether to output token(word) level information'
', such times/confidence')
parser.add_argument('--align',
action='store_true',
help='force align the input audio and transcript')
parser.add_argument('--label', type=str, help='the input label to align')
parser.add_argument('--paraformer',
action='store_true',
help='whether to use the best chinese model')
parser.add_argument('--beam', type=int, default=5, help="beam size")
parser.add_argument('--context_path',
type=str,
default=None,
help='context list file')
parser.add_argument('--context_score',
type=float,
default=6.0,
help='context score')
args = parser.parse_args()
return args
def main():
args = get_args()
if args.paraformer:
model = load_paraformer(args.model_dir, args.gpu, args.device)
else:
model = load_model(args.language, args.model_dir, args.gpu, args.beam,
args.context_path, args.context_score, args.device)
if args.align:
result = model.align(args.audio_file, args.label)
else:
result = model.transcribe(args.audio_file, args.show_tokens_info)
print(result)
if __name__ == "__main__":
main()
|