Spaces:
Sleeping
Sleeping
File size: 2,184 Bytes
5ce1fe8 941850e 5ce1fe8 941850e 5ce1fe8 159e07d 5ce1fe8 159e07d 5ce1fe8 159e07d 5ce1fe8 159e07d 5ce1fe8 159e07d 5ce1fe8 159e07d 5ce1fe8 |
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 |
#!/usr/bin/python3
# -*- coding: utf-8 -*-
import argparse
import os
from pathlib import Path
import sys
pwd = os.path.abspath(os.path.dirname(__file__))
sys.path.append(os.path.join(pwd, "../../"))
import huggingface_hub
import sherpa
from project_settings import project_path
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument(
"--repo_id",
default="csukuangfj/wenet-chinese-model",
# default="csukuangfj/wenet-english-model",
type=str
)
parser.add_argument("--model_filename", default="final.zip", type=str)
parser.add_argument("--tokens_filename", default="units.txt", type=str)
parser.add_argument(
"--pretrained_model_dir",
default=(project_path / "pretrained_models").as_posix(),
type=str
)
args = parser.parse_args()
return args
def main():
args = get_args()
pretrained_model_dir = Path(args.pretrained_model_dir)
pretrained_model_dir.mkdir(exist_ok=True)
model_dir = pretrained_model_dir / "huggingface" / args.repo_id
model_dir.mkdir(exist_ok=True)
print("download model")
model_filename = huggingface_hub.hf_hub_download(
repo_id=args.repo_id,
filename=args.model_filename,
subfolder=".",
local_dir=model_dir.as_posix(),
)
print(model_filename)
print("download tokens")
tokens_filename = huggingface_hub.hf_hub_download(
repo_id=args.repo_id,
filename=args.tokens_filename,
subfolder=".",
local_dir=model_dir.as_posix(),
)
print(tokens_filename)
feat_config = sherpa.FeatureConfig(normalize_samples=False)
feat_config.fbank_opts.frame_opts.samp_freq = 16000
feat_config.fbank_opts.mel_opts.num_bins = 80
feat_config.fbank_opts.frame_opts.dither = 0
config = sherpa.OfflineRecognizerConfig(
nn_model=model_filename,
tokens=tokens_filename,
use_gpu=False,
feat_config=feat_config,
decoding_method="greedy_search",
num_active_paths=2,
)
recognizer = sherpa.OfflineRecognizer(config)
print(recognizer)
return
if __name__ == "__main__":
main()
|