initial version of whisper normalizer

kurianbenoy · Mar 21, 2023 · f65fe64 · f65fe64
1 parent c47942d
commit f65fe64
Show file tree

Hide file tree

Showing 4 changed files with 107 additions and 53 deletions.
diff --git a/nbs/00_basic.ipynb b/nbs/00_basic.ipynb
@@ -131,7 +131,7 @@
       "text/markdown": [
        "---\n",
        "\n",
-       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/base.py#L62){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/basic.py#L62){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### BasicTextNormalizer\n",
        "\n",
@@ -143,7 +143,7 @@
       "text/plain": [
        "---\n",
        "\n",
-       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/base.py#L62){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/basic.py#L62){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### BasicTextNormalizer\n",
        "\n",
@@ -174,37 +174,28 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'എന റ കമ പ യ ട ടറ ന എന റ ഭ ഷ'"
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "normalizer = BasicTextNormalizer()\n"
+    "normalizer = BasicTextNormalizer()\n",
+    "normalizer(\"എന്റെ കമ്പ്യൂട്ടറിനു് എന്റെ ഭാഷ\")"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "ename": "JSONDecodeError",
-     "evalue": "Expecting value: line 1 column 1 (char 0)",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mJSONDecodeError\u001b[0m                           Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[7], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[39m#| hide\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mnbdev\u001b[39;00m; nbdev\u001b[39m.\u001b[39;49mnbdev_export()\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/fastcore/script.py:110\u001b[0m, in \u001b[0;36mcall_parse.<locals>._f\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    107\u001b[0m \u001b[39m@wraps\u001b[39m(func)\n\u001b[1;32m    108\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_f\u001b[39m(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs):\n\u001b[1;32m    109\u001b[0m     mod \u001b[39m=\u001b[39m inspect\u001b[39m.\u001b[39mgetmodule(inspect\u001b[39m.\u001b[39mcurrentframe()\u001b[39m.\u001b[39mf_back)\n\u001b[0;32m--> 110\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m mod: \u001b[39mreturn\u001b[39;00m func(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    111\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m SCRIPT_INFO\u001b[39m.\u001b[39mfunc \u001b[39mand\u001b[39;00m mod\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m\u001b[39m==\u001b[39m\u001b[39m\"\u001b[39m\u001b[39m__main__\u001b[39m\u001b[39m\"\u001b[39m: SCRIPT_INFO\u001b[39m.\u001b[39mfunc \u001b[39m=\u001b[39m func\u001b[39m.\u001b[39m\u001b[39m__name__\u001b[39m\n\u001b[1;32m    112\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mlen\u001b[39m(sys\u001b[39m.\u001b[39margv)\u001b[39m>\u001b[39m\u001b[39m1\u001b[39m \u001b[39mand\u001b[39;00m sys\u001b[39m.\u001b[39margv[\u001b[39m1\u001b[39m]\u001b[39m==\u001b[39m\u001b[39m'\u001b[39m\u001b[39m'\u001b[39m: sys\u001b[39m.\u001b[39margv\u001b[39m.\u001b[39mpop(\u001b[39m1\u001b[39m)\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/nbdev/doclinks.py:138\u001b[0m, in \u001b[0;36mnbdev_export\u001b[0;34m(path, **kwargs)\u001b[0m\n\u001b[1;32m    136\u001b[0m \u001b[39mif\u001b[39;00m os\u001b[39m.\u001b[39menviron\u001b[39m.\u001b[39mget(\u001b[39m'\u001b[39m\u001b[39mIN_TEST\u001b[39m\u001b[39m'\u001b[39m,\u001b[39m0\u001b[39m): \u001b[39mreturn\u001b[39;00m\n\u001b[1;32m    137\u001b[0m files \u001b[39m=\u001b[39m nbglob(path\u001b[39m=\u001b[39mpath, as_path\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\u001b[39m.\u001b[39msorted(\u001b[39m'\u001b[39m\u001b[39mname\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[0;32m--> 138\u001b[0m \u001b[39mfor\u001b[39;00m f \u001b[39min\u001b[39;00m files: nb_export(f)\n\u001b[1;32m    139\u001b[0m add_init(get_config()\u001b[39m.\u001b[39mlib_path)\n\u001b[1;32m    140\u001b[0m _build_modidx()\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/nbdev/export.py:48\u001b[0m, in \u001b[0;36mnb_export\u001b[0;34m(nbname, lib_path, procs, debug, mod_maker, name)\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[39mif\u001b[39;00m lib_path \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: lib_path \u001b[39m=\u001b[39m get_config()\u001b[39m.\u001b[39mlib_path\n\u001b[1;32m     47\u001b[0m exp \u001b[39m=\u001b[39m ExportModuleProc()\n\u001b[0;32m---> 48\u001b[0m nb \u001b[39m=\u001b[39m NBProcessor(nbname, [exp]\u001b[39m+\u001b[39;49mL(procs), debug\u001b[39m=\u001b[39;49mdebug)\n\u001b[1;32m     49\u001b[0m nb\u001b[39m.\u001b[39mprocess()\n\u001b[1;32m     50\u001b[0m \u001b[39mfor\u001b[39;00m mod,cells \u001b[39min\u001b[39;00m exp\u001b[39m.\u001b[39mmodules\u001b[39m.\u001b[39mitems():\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/nbdev/process.py:92\u001b[0m, in \u001b[0;36mNBProcessor.__init__\u001b[0;34m(self, path, procs, nb, debug, rm_directives, process)\u001b[0m\n\u001b[1;32m     91\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__init__\u001b[39m(\u001b[39mself\u001b[39m, path\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, procs\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, nb\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, debug\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m, rm_directives\u001b[39m=\u001b[39m\u001b[39mTrue\u001b[39;00m, process\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m):\n\u001b[0;32m---> 92\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mnb \u001b[39m=\u001b[39m read_nb(path) \u001b[39mif\u001b[39;00m nb \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39melse\u001b[39;00m nb\n\u001b[1;32m     93\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mlang \u001b[39m=\u001b[39m nb_lang(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mnb)\n\u001b[1;32m     94\u001b[0m     \u001b[39mfor\u001b[39;00m cell \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mnb\u001b[39m.\u001b[39mcells: cell\u001b[39m.\u001b[39mdirectives_ \u001b[39m=\u001b[39m extract_directives(cell, remove\u001b[39m=\u001b[39mrm_directives, lang\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mlang)\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/execnb/nbio.py:57\u001b[0m, in \u001b[0;36mread_nb\u001b[0;34m(path)\u001b[0m\n\u001b[1;32m     55\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mread_nb\u001b[39m(path):\n\u001b[1;32m     56\u001b[0m     \u001b[39m\"\u001b[39m\u001b[39mReturn notebook at `path`\u001b[39m\u001b[39m\"\u001b[39m\n\u001b[0;32m---> 57\u001b[0m     res \u001b[39m=\u001b[39m dict2nb(_read_json(path, encoding\u001b[39m=\u001b[39;49m\u001b[39m'\u001b[39;49m\u001b[39mutf-8\u001b[39;49m\u001b[39m'\u001b[39;49m))\n\u001b[1;32m     58\u001b[0m     res[\u001b[39m'\u001b[39m\u001b[39mpath_\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m \u001b[39mstr\u001b[39m(path)\n\u001b[1;32m     59\u001b[0m     \u001b[39mreturn\u001b[39;00m res\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/site-packages/execnb/nbio.py:16\u001b[0m, in \u001b[0;36m_read_json\u001b[0;34m(self, encoding, errors)\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_read_json\u001b[39m(\u001b[39mself\u001b[39m, encoding\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, errors\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[0;32m---> 16\u001b[0m     \u001b[39mreturn\u001b[39;00m loads(Path(\u001b[39mself\u001b[39;49m)\u001b[39m.\u001b[39;49mread_text(encoding\u001b[39m=\u001b[39;49mencoding, errors\u001b[39m=\u001b[39;49merrors))\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/json/__init__.py:346\u001b[0m, in \u001b[0;36mloads\u001b[0;34m(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)\u001b[0m\n\u001b[1;32m    341\u001b[0m     s \u001b[39m=\u001b[39m s\u001b[39m.\u001b[39mdecode(detect_encoding(s), \u001b[39m'\u001b[39m\u001b[39msurrogatepass\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[1;32m    343\u001b[0m \u001b[39mif\u001b[39;00m (\u001b[39mcls\u001b[39m \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m object_hook \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m\n\u001b[1;32m    344\u001b[0m         parse_int \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m parse_float \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m\n\u001b[1;32m    345\u001b[0m         parse_constant \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m object_pairs_hook \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m \u001b[39mnot\u001b[39;00m kw):\n\u001b[0;32m--> 346\u001b[0m     \u001b[39mreturn\u001b[39;00m _default_decoder\u001b[39m.\u001b[39;49mdecode(s)\n\u001b[1;32m    347\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mcls\u001b[39m \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m    348\u001b[0m     \u001b[39mcls\u001b[39m \u001b[39m=\u001b[39m JSONDecoder\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/json/decoder.py:337\u001b[0m, in \u001b[0;36mJSONDecoder.decode\u001b[0;34m(self, s, _w)\u001b[0m\n\u001b[1;32m    332\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mdecode\u001b[39m(\u001b[39mself\u001b[39m, s, _w\u001b[39m=\u001b[39mWHITESPACE\u001b[39m.\u001b[39mmatch):\n\u001b[1;32m    333\u001b[0m     \u001b[39m\"\"\"Return the Python representation of ``s`` (a ``str`` instance\u001b[39;00m\n\u001b[1;32m    334\u001b[0m \u001b[39m    containing a JSON document).\u001b[39;00m\n\u001b[1;32m    335\u001b[0m \n\u001b[1;32m    336\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 337\u001b[0m     obj, end \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mraw_decode(s, idx\u001b[39m=\u001b[39;49m_w(s, \u001b[39m0\u001b[39;49m)\u001b[39m.\u001b[39;49mend())\n\u001b[1;32m    338\u001b[0m     end \u001b[39m=\u001b[39m _w(s, end)\u001b[39m.\u001b[39mend()\n\u001b[1;32m    339\u001b[0m     \u001b[39mif\u001b[39;00m end \u001b[39m!=\u001b[39m \u001b[39mlen\u001b[39m(s):\n",
-      "File \u001b[0;32m~/mambaforge/lib/python3.10/json/decoder.py:355\u001b[0m, in \u001b[0;36mJSONDecoder.raw_decode\u001b[0;34m(self, s, idx)\u001b[0m\n\u001b[1;32m    353\u001b[0m     obj, end \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mscan_once(s, idx)\n\u001b[1;32m    354\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mStopIteration\u001b[39;00m \u001b[39mas\u001b[39;00m err:\n\u001b[0;32m--> 355\u001b[0m     \u001b[39mraise\u001b[39;00m JSONDecodeError(\u001b[39m\"\u001b[39m\u001b[39mExpecting value\u001b[39m\u001b[39m\"\u001b[39m, s, err\u001b[39m.\u001b[39mvalue) \u001b[39mfrom\u001b[39;00m \u001b[39mNone\u001b[39m\n\u001b[1;32m    356\u001b[0m \u001b[39mreturn\u001b[39;00m obj, end\n",
-      "\u001b[0;31mJSONDecodeError\u001b[0m: Expecting value: line 1 column 1 (char 0)"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "#| hide\n",
     "import nbdev; nbdev.nbdev_export()"

diff --git a/nbs/01_english.ipynb b/nbs/01_english.ipynb
@@ -32,26 +32,15 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "ename": "ModuleNotFoundError",
-     "evalue": "No module named 'whisper_normalizer'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[3], line 10\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mtyping\u001b[39;00m \u001b[39mimport\u001b[39;00m Iterator, List, Match, Optional, Union\n\u001b[1;32m      8\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mmore_itertools\u001b[39;00m \u001b[39mimport\u001b[39;00m windowed\n\u001b[0;32m---> 10\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mwhisper_normalizer\u001b[39;00m\u001b[39m.\u001b[39;00m\u001b[39mbasic\u001b[39;00m \u001b[39mimport\u001b[39;00m remove_symbols_and_diacritics\n\u001b[1;32m     13\u001b[0m \u001b[39mclass\u001b[39;00m \u001b[39mEnglishNumberNormalizer\u001b[39;00m:\n\u001b[1;32m     14\u001b[0m     \u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m     15\u001b[0m \u001b[39m    Convert any spelled-out numbers into arabic numbers, while handling:\u001b[39;00m\n\u001b[1;32m     16\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     21\u001b[0m \u001b[39m    - interpret successive single-digit numbers as nominal: `one oh one` -> `101`\u001b[39;00m\n\u001b[1;32m     22\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n",
-      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'whisper_normalizer'"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# | export\n",
     "import json\n",
     "import os\n",
     "import re\n",
     "from fractions import Fraction\n",
     "from typing import Iterator, List, Match, Optional, Union\n",
+    "import urllib\n",
     "\n",
     "from more_itertools import windowed\n",
     "\n",
@@ -493,9 +482,16 @@
     "        s = \" \".join(word for word in self.process_words(s.split()) if word is not None)\n",
     "        s = self.postprocess(s)\n",
     "\n",
-    "        return s\n",
-    "\n",
-    "\n",
+    "        return s\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| export\n",
     "class EnglishSpellingNormalizer:\n",
     "    \"\"\"\n",
     "    Applies British-American spelling mappings as listed in [1].\n",
@@ -504,14 +500,35 @@
     "    \"\"\"\n",
     "\n",
     "    def __init__(self):\n",
-    "        mapping_path = os.path.join(os.path.dirname(__file__), \"english.json\")\n",
-    "        self.mapping = json.load(open(mapping_path))\n",
+    "        response = urllib.request.urlopen(\"https://gist.githubusercontent.com/kurianbenoy/715c4528be9859ff64338f69416795c7/raw/936c565f059b81d007e2c52beb733b3a01937d90/openai_whisper.json\")\n",
+    "        self.mapping = json.loads(response.read())\n",
     "\n",
     "    def __call__(self, s: str):\n",
     "        return \" \".join(self.mapping.get(word, word) for word in s.split())\n",
     "\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'accessorize'"
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "n = EnglishSpellingNormalizer()\n",
+    "n(\"accessorise\")"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -604,8 +621,54 @@
     "\n",
     "        s = re.sub(r\"\\s+\", \" \", s)  # replace any successive whitespaces with a space\n",
     "\n",
-    "        return s\n"
+    "        return s"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/markdown": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/english.py#L473){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### EnglishTextNormalizer\n",
+       "\n",
+       ">      EnglishTextNormalizer ()\n",
+       "\n",
+       "Initialize self.  See help(type(self)) for accurate signature."
+      ],
+      "text/plain": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kurianbenoy/whisper_normalizer/blob/main/whisper_normalizer/english.py#L473){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### EnglishTextNormalizer\n",
+       "\n",
+       ">      EnglishTextNormalizer ()\n",
+       "\n",
+       "Initialize self.  See help(type(self)) for accurate signature."
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "show_doc(EnglishTextNormalizer)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {

diff --git a/whisper_normalizer/basic.py b/whisper_normalizer/basic.py
@@ -29,6 +29,7 @@
     "Ł": "L",
 }
 
+
 def remove_symbols_and_diacritics(s: str, keep=""):
     """
     Replace any other markers, symbols, and punctuations with a space,
@@ -57,7 +58,6 @@ def remove_symbols(s: str):
         for c in unicodedata.normalize("NFKC", s)
     )
 
-
 # %% ../nbs/00_basic.ipynb 4
 class BasicTextNormalizer:
     def __init__(self, remove_diacritics: bool = False, split_letters: bool = False):

diff --git a/whisper_normalizer/english.py b/whisper_normalizer/english.py
@@ -9,6 +9,7 @@
 import re
 from fractions import Fraction
 from typing import Iterator, List, Match, Optional, Union
+import urllib
 
 from more_itertools import windowed
 
@@ -452,7 +453,7 @@ def __call__(self, s: str):
 
         return s
 
-
+# %% ../nbs/01_english.ipynb 4
 class EnglishSpellingNormalizer:
     """
     Applies British-American spelling mappings as listed in [1].
@@ -461,15 +462,15 @@ class EnglishSpellingNormalizer:
     """
 
     def __init__(self):
-        mapping_path = os.path.join(os.path.dirname(__file__), "english.json")
-        self.mapping = json.load(open(mapping_path))
+        response = urllib.request.urlopen(
+            "https://gist.githubusercontent.com/kurianbenoy/715c4528be9859ff64338f69416795c7/raw/936c565f059b81d007e2c52beb733b3a01937d90/openai_whisper.json"
+        )
+        self.mapping = json.loads(response.read())
 
     def __call__(self, s: str):
         return " ".join(self.mapping.get(word, word) for word in s.split())
 
-
-
-# %% ../nbs/01_english.ipynb 4
+# %% ../nbs/01_english.ipynb 6
 class EnglishTextNormalizer:
     def __init__(self):
         self.ignore_patterns = r"\b(hmm|mm|mhm|mmm|uh|um)\b"
@@ -556,4 +557,3 @@ def __call__(self, s: str):
         s = re.sub(r"\s+", " ", s)  # replace any successive whitespaces with a space
 
         return s
-