File size: 3,682 Bytes
b81c7d0 64d7832 b81c7d0 b3439d5 93cb4db b3439d5 b81c7d0 b3439d5 0a30a52 b3439d5 b81c7d0 |
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 |
#!/usr/bin/env python3
import os
import re
from pathlib import Path
from typing import List
BASE_URL = "https://huggingface.co/csukuangfj/sherpa-onnx-apk/resolve/main/"
from dataclasses import dataclass
@dataclass
class APK:
major: int
minor: int
patch: int
arch: str
short_name: str
def __init__(self, s):
# sherpa-onnx-1.9.22-x86-slid-whisper_tiny.apk
# sherpa-onnx-1.9.22-arm64-v8a-slid-whisper_tiny.apk
s = str(s)[len("slid/") :]
split = s.split("-")
self.major, self.minor, self.patch = list(map(int, split[2].split(".")))
self.arch = split[3]
self.short_name = split[5]
if "arm" in s:
self.arch += "-" + split[4]
self.short_name = split[6]
if "armeabi" in self.arch:
self.arch = "y" + self.arch
if "arm64" in self.arch:
self.arch = "z" + self.arch
def sort_by_apk(x):
x = APK(x)
return (x.major, x.minor, x.patch, x.arch, x.short_name)
def generate_url(files: List[str]) -> List[str]:
ans = []
base = BASE_URL
for f in files:
ans.append(base + str(f))
return ans
def get_all_files(d: str, suffix: str) -> List[str]:
ans = sorted(Path(d).glob(suffix), key=sort_by_apk, reverse=True)
return list(map(lambda x: BASE_URL + str(x), ans))
def to_file(filename: str, files: List[str]):
content = r"""
<h1> APKs for spoken language identification </h1>
This page lists the <strong>spoken language identification</strong> APKs for <a href="http://github.com/k2-fsa/sherpa-onnx">sherpa-onnx</a>,
one of the deployment frameworks of <a href="https://github.com/k2-fsa">the Next-gen Kaldi project</a>.
<br/>
The name of an APK has the following rule:
<ul>
<li> sherpa-onnx-{version}-{arch}-slid-{model}.apk
</ul>
where
<ul>
<li> version: It specifies the current version, e.g., 1.9.21
<li> arch: The architecture targeted by this APK, e.g., arm64-v8a, armeabi-v7a, x86_64, x86
<li> model: The name of the model used in the APK
</ul>
<br/><br/>
You can download all supported models from
<a href="https://github.com/k2-fsa/sherpa-onnx/releases/tag/asr-models">https://github.com/k2-fsa/sherpa-onnx/releases/tag/asr-models</a>
<br/>
<br/>
Please see
<a href="https://k2-fsa.github.io/sherpa/onnx/spolken-language-identification/">https://k2-fsa.github.io/sherpa/onnx/spolken-language-identification/</a>
for more information.
<br/>
<br/>
<strong>Note about the license</strong> The code of Next-gen Kaldi is using
<a href="https://www.apache.org/licenses/LICENSE-2.0">Apache-2.0 license</a>. However,
we support models from different frameworks. Please check the license of your selected model.
<br/>
<br/>
<div/>
"""
if "-cn" not in filename:
content += """
For Chinese users, please <a href="./apk-cn.html">visit this address</a>,
which replaces <a href="huggingface.co">huggingface.co</a> with <a href="hf-mirror.com">hf-mirror.com</a>
<br/>
<br/>
中国用户, 请访问<a href="./apk-cn.html">这个地址</a>
<br/>
<br/>
"""
with open(filename, "w") as f:
print(content, file=f)
for x in files:
name = x.rsplit("/", maxsplit=1)[-1]
print(f'<a href="{x}" />{name}<br/>', file=f)
def main():
apk = get_all_files("slid", suffix="*.apk")
to_file("./apk-slid.html", apk)
# for Chinese users
apk2 = []
for a in apk:
a = a.replace("huggingface.co", "hf-mirror.com")
a = a.replace("resolve", "blob")
apk2.append(a)
to_file("./apk-slid-cn.html", apk2)
if __name__ == "__main__":
main()
|