File size: 11,880 Bytes
2f6b10b | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 | import collections
import json
import os
import time
import requests
from montreal_forced_aligner.models import MODEL_TYPES, ModelManager, ModelRelease
mfa_model_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
UPDATE = False
with open(os.path.join(mfa_model_root, "scripts", "token"), "r") as f:
token = f.read()
CURRENT_VERSION = "v3.3.0"
tag_template = "{model_type}-{model_name}-v{version}"
manager = ModelManager(token=token)
manager.refresh_remote()
model_type_names = {
"acoustic": "Acoustic models",
"dictionary": "Pronunciation dictionaries",
"g2p": "G2P models",
"language_model": "Language models",
"ivector": "Ivector extractors",
"corpus": "Corpora",
"tokenizer": "Tokenizers",
}
print(manager.remote_models)
base_dict_template = "https://github.com/MontrealCorpusTools/mfa-models/tree/main/dictionary/{language}/{phone_set}/{version}/{model_name}.dict"
acoustic_mfas = set()
for model_type, model_class in MODEL_TYPES.items():
model_directory = os.path.join(mfa_model_root, model_type)
staging_directory = os.path.join(model_directory, "staging")
languages = os.listdir(model_directory)
for lang in languages:
if lang == "staging":
continue
lang_dir = os.path.join(model_directory, lang)
if not os.path.isdir(lang_dir):
continue
if model_type in {"ivector", "tokenizer"}:
versions = os.listdir(lang_dir)
for v in versions:
version_dir = os.path.join(lang_dir, v)
if v == "v2.0.0":
continue
if not os.path.isdir(version_dir):
continue
if not os.listdir(version_dir):
continue
with open(os.path.join(version_dir, "meta.json"), "r", encoding="utf8") as f:
meta = json.load(f)
model_name = meta["name"]
version = meta["version"]
with open(os.path.join(version_dir, "README.md"), "r", encoding="utf8") as f:
readme = f.read()
tag = tag_template.format(
model_type=model_type, model_name=model_name, version=version
)
if "mfa" in tag and model_type == "acoustic":
acoustic_mfas.add(lang)
elif "mfa" in tag and lang not in acoustic_mfas and model_type != "ivector":
continue
if ("mfa" in tag or "arpa" in tag) and model_type == "dictionary":
dict_url = base_dict_template.format(
language=lang,
phone_set=phone_set,
version=v,
model_name=model_name,
)
readme = readme.replace(
"\n\n## Installation",
f"\n- The dictionary downloadable from this release has trained pronunciation and silence probabilities. The base dictionary is available [here]({dict_url})\n\n##Installation",
)
if "../../../../corpus/" in readme:
readme = readme.replace(
"../../../../corpus/",
"https://github.com/MontrealCorpusTools/mfa-models/tree/main/corpus/",
)
elif "../../../corpus/" in readme:
readme = readme.replace(
"../../../corpus/",
"https://github.com/MontrealCorpusTools/mfa-models/tree/main/corpus/",
)
existing_releases = manager.remote_models[model_type]
if model_name in existing_releases:
continue
existing = existing_releases[model_name]
if existing.version.replace("v", "") == version:
if UPDATE:
print("UPDATING", existing.release_link)
r = requests.patch(existing.release_link, json={"body": readme})
time.sleep(5)
continue
release = ModelRelease(model_name, tag, version, "", "")
if model_type == "dictionary":
ext = ".dict"
content_type = "text/tab-separated-values"
else:
ext = ".zip"
content_type = "application/zip"
model_path = os.path.join(staging_directory, model_name + ext)
print(tag, len(readme))
print(tag)
r = requests.post(
manager.base_url,
json={
"tag_name": tag,
"name": f"{model_name} v{version}",
"body": readme,
"target_commitish": "main",
"draft": False,
"prerelease": False,
"generate_release_notes": False,
},
headers={
"Accept": "application/vnd.github.v3+json",
"Authorization": f"token {token}",
},
)
d = r.json()
time.sleep(5)
print(d)
if "errors" in d:
continue
with open(model_path, "rb") as f:
data = f.read()
r2 = requests.post(
d["upload_url"].replace("{?name,label}", ""),
data=data,
params={"name": os.path.basename(model_path)},
headers={
"Content-Type": "application/zip",
"Accept": "application/vnd.github.v3+json",
"Authorization": f"token {token}",
},
)
print(r2.json())
print(meta)
print(tag)
time.sleep(5)
else:
for phone_set in os.listdir(lang_dir):
phone_set_dir = os.path.join(lang_dir, phone_set)
if not os.path.isdir(phone_set_dir):
continue
versions = os.listdir(phone_set_dir)
for v in versions:
version_dir = os.path.join(phone_set_dir, v)
if v != CURRENT_VERSION:
continue
if not os.path.isdir(version_dir):
continue
if not os.listdir(version_dir):
continue
with open(os.path.join(version_dir, "meta.json"), "r", encoding="utf8") as f:
meta = json.load(f)
model_name = meta["name"]
version = meta["version"]
with open(os.path.join(version_dir, "README.md"), "r", encoding="utf8") as f:
readme = f.read()
tag = tag_template.format(
model_type=model_type, model_name=model_name, version=version
)
if "mfa" in tag and model_type == "acoustic":
acoustic_mfas.add(lang)
elif "mfa" in tag and lang not in acoustic_mfas and model_type != "ivector":
continue
if ("mfa" in tag or "arpa" in tag) and model_type == "dictionary":
dict_url = base_dict_template.format(
language=lang,
phone_set=phone_set,
version=v,
model_name=model_name,
)
readme = readme.replace(
"\n\n## Installation",
f"\n- The dictionary downloadable from this release has trained pronunciation and silence probabilities. The base dictionary is available [here]({dict_url})\n\n##Installation",
)
if "../../../../corpus/" in readme:
readme = readme.replace(
"../../../../corpus/",
"https://github.com/MontrealCorpusTools/mfa-models/tree/main/corpus/",
)
elif "../../../corpus/" in readme:
readme = readme.replace(
"../../../corpus/",
"https://github.com/MontrealCorpusTools/mfa-models/tree/main/corpus/",
)
existing_releases = manager.remote_models[model_type]
if model_name in existing_releases:
existing = existing_releases[model_name]
found_existing = False
for existing_version, model in existing.items():
if existing_version.replace("v", "") == version:
if UPDATE:
print("UPDATING", existing.release_link)
r = requests.patch(
existing.release_link, json={"body": readme}
)
time.sleep(5)
found_existing = True
if found_existing:
continue
release = ModelRelease(model_name, tag, version, "", "")
if model_type == "dictionary":
ext = ".dict"
content_type = "text/tab-separated-values"
else:
ext = ".zip"
content_type = "application/zip"
model_path = os.path.join(staging_directory, model_name + ext)
print(tag, len(readme))
print(tag)
r = requests.post(
manager.base_url,
json={
"tag_name": tag,
"name": f"{model_name} v{version}",
"body": readme,
"target_commitish": "main",
"draft": False,
"prerelease": False,
"generate_release_notes": False,
},
headers={
"Accept": "application/vnd.github.v3+json",
"Authorization": f"token {token}",
},
)
d = r.json()
time.sleep(5)
print(d)
if "errors" in d:
continue
with open(model_path, "rb") as f:
data = f.read()
r2 = requests.post(
d["upload_url"].replace("{?name,label}", ""),
data=data,
params={"name": os.path.basename(model_path)},
headers={
"Content-Type": "application/zip",
"Accept": "application/vnd.github.v3+json",
"Authorization": f"token {token}",
},
)
print(r2.json())
print(meta)
print(tag)
time.sleep(5)
|