{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "438927ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import AutoModelForCTC, Wav2Vec2Processor\n",
    "from datasets import load_dataset, load_metric, Audio\n",
    "import numpy as np\n",
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "27a57965",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = AutoModelForCTC.from_pretrained(\".\").to('cuda')\n",
    "processor = Wav2Vec2Processor.from_pretrained(\".\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1d4324df",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "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)",
      "Input \u001b[0;32mIn [3]\u001b[0m, in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m model \u001b[38;5;241m=\u001b[39m AutoModelForCTC\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mvitouphy/xls-r-300m-km\u001b[39m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39mto(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcuda\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m----> 2\u001b[0m processor \u001b[38;5;241m=\u001b[39m \u001b[43mWav2Vec2Processor\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_pretrained\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mvitouphy/xls-r-300m-km\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/site-packages/transformers/models/wav2vec2/processing_wav2vec2.py:117\u001b[0m, in \u001b[0;36mWav2Vec2Processor.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[1;32m    112\u001b[0m \u001b[38;5;66;03m# load generic `AutoTokenizer`\u001b[39;00m\n\u001b[1;32m    113\u001b[0m \u001b[38;5;66;03m# need fallback here for backward compatibility in case processor is\u001b[39;00m\n\u001b[1;32m    114\u001b[0m \u001b[38;5;66;03m# loaded from just a tokenizer file that does not have a `tokenizer_class` attribute\u001b[39;00m\n\u001b[1;32m    115\u001b[0m \u001b[38;5;66;03m# behavior should be deprecated in major future release\u001b[39;00m\n\u001b[1;32m    116\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 117\u001b[0m     tokenizer \u001b[38;5;241m=\u001b[39m \u001b[43mAutoTokenizer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_pretrained\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpretrained_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    118\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mOSError\u001b[39;00m:\n\u001b[1;32m    119\u001b[0m     warnings\u001b[38;5;241m.\u001b[39mwarn(\n\u001b[1;32m    120\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mLoading a tokenizer inside \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m from a config that does not\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    121\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m include a `tokenizer_class` attribute is deprecated and will be \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    125\u001b[0m         \u001b[38;5;167;01mFutureWarning\u001b[39;00m,\n\u001b[1;32m    126\u001b[0m     )\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/site-packages/transformers/models/auto/tokenization_auto.py:514\u001b[0m, in \u001b[0;36mAutoTokenizer.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, *inputs, **kwargs)\u001b[0m\n\u001b[1;32m    510\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m tokenizer_class \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    511\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m    512\u001b[0m             \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mTokenizer class \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mtokenizer_class_candidate\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m does not exist or is not currently imported.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    513\u001b[0m         )\n\u001b[0;32m--> 514\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtokenizer_class\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_pretrained\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpretrained_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    516\u001b[0m \u001b[38;5;66;03m# Otherwise we have to be creative.\u001b[39;00m\n\u001b[1;32m    517\u001b[0m \u001b[38;5;66;03m# if model is an encoder decoder, the encoder tokenizer class is used by default\u001b[39;00m\n\u001b[1;32m    518\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(config, EncoderDecoderConfig):\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/site-packages/transformers/tokenization_utils_base.py:1773\u001b[0m, in \u001b[0;36mPreTrainedTokenizerBase.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, *init_inputs, **kwargs)\u001b[0m\n\u001b[1;32m   1770\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1771\u001b[0m         logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mloading file \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfile_path\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m from cache at \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mresolved_vocab_files[file_id]\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m-> 1773\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_from_pretrained\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1774\u001b[0m \u001b[43m    \u001b[49m\u001b[43mresolved_vocab_files\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1775\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpretrained_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1776\u001b[0m \u001b[43m    \u001b[49m\u001b[43minit_configuration\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1777\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43minit_inputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1778\u001b[0m \u001b[43m    \u001b[49m\u001b[43muse_auth_token\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muse_auth_token\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1779\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcache_dir\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcache_dir\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1780\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1781\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/site-packages/transformers/tokenization_utils_base.py:1908\u001b[0m, in \u001b[0;36mPreTrainedTokenizerBase._from_pretrained\u001b[0;34m(cls, resolved_vocab_files, pretrained_model_name_or_path, init_configuration, use_auth_token, cache_dir, *init_inputs, **kwargs)\u001b[0m\n\u001b[1;32m   1906\u001b[0m \u001b[38;5;66;03m# Instantiate tokenizer.\u001b[39;00m\n\u001b[1;32m   1907\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1908\u001b[0m     tokenizer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43minit_inputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43minit_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1909\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mOSError\u001b[39;00m:\n\u001b[1;32m   1910\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mOSError\u001b[39;00m(\n\u001b[1;32m   1911\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mUnable to load vocabulary from file. \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1912\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPlease check that the provided vocabulary is accessible and not corrupted.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1913\u001b[0m     )\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/site-packages/transformers/models/wav2vec2/tokenization_wav2vec2.py:142\u001b[0m, in \u001b[0;36mWav2Vec2CTCTokenizer.__init__\u001b[0;34m(self, vocab_file, bos_token, eos_token, unk_token, pad_token, word_delimiter_token, do_lower_case, **kwargs)\u001b[0m\n\u001b[1;32m    139\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdo_lower_case \u001b[38;5;241m=\u001b[39m do_lower_case\n\u001b[1;32m    141\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mopen\u001b[39m(vocab_file, encoding\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mutf-8\u001b[39m\u001b[38;5;124m\"\u001b[39m) \u001b[38;5;28;01mas\u001b[39;00m vocab_handle:\n\u001b[0;32m--> 142\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mencoder \u001b[38;5;241m=\u001b[39m \u001b[43mjson\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mload\u001b[49m\u001b[43m(\u001b[49m\u001b[43mvocab_handle\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    143\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdecoder \u001b[38;5;241m=\u001b[39m {v: k \u001b[38;5;28;01mfor\u001b[39;00m k, v \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mencoder\u001b[38;5;241m.\u001b[39mitems()}\n\u001b[1;32m    145\u001b[0m \u001b[38;5;66;03m# make sure that tokens made of several\u001b[39;00m\n\u001b[1;32m    146\u001b[0m \u001b[38;5;66;03m# characters are not split at tokenization\u001b[39;00m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/json/__init__.py:293\u001b[0m, in \u001b[0;36mload\u001b[0;34m(fp, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)\u001b[0m\n\u001b[1;32m    274\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mload\u001b[39m(fp, \u001b[38;5;241m*\u001b[39m, \u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, object_hook\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, parse_float\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[1;32m    275\u001b[0m         parse_int\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, parse_constant\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, object_pairs_hook\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkw):\n\u001b[1;32m    276\u001b[0m     \u001b[38;5;124;03m\"\"\"Deserialize ``fp`` (a ``.read()``-supporting file-like object containing\u001b[39;00m\n\u001b[1;32m    277\u001b[0m \u001b[38;5;124;03m    a JSON document) to a Python object.\u001b[39;00m\n\u001b[1;32m    278\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    291\u001b[0m \u001b[38;5;124;03m    kwarg; otherwise ``JSONDecoder`` is used.\u001b[39;00m\n\u001b[1;32m    292\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 293\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mloads\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    294\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mcls\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mobject_hook\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mobject_hook\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    295\u001b[0m \u001b[43m        \u001b[49m\u001b[43mparse_float\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparse_float\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparse_int\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparse_int\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    296\u001b[0m \u001b[43m        \u001b[49m\u001b[43mparse_constant\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparse_constant\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mobject_pairs_hook\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mobject_pairs_hook\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkw\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/json/__init__.py:357\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    352\u001b[0m     \u001b[38;5;28;01mdel\u001b[39;00m kw[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mencoding\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[1;32m    354\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (\u001b[38;5;28mcls\u001b[39m \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m object_hook \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[1;32m    355\u001b[0m         parse_int \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m parse_float \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[1;32m    356\u001b[0m         parse_constant \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m object_pairs_hook \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m kw):\n\u001b[0;32m--> 357\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_default_decoder\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdecode\u001b[49m\u001b[43m(\u001b[49m\u001b[43ms\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    358\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mcls\u001b[39m \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    359\u001b[0m     \u001b[38;5;28mcls\u001b[39m \u001b[38;5;241m=\u001b[39m JSONDecoder\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/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[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecode\u001b[39m(\u001b[38;5;28mself\u001b[39m, s, _w\u001b[38;5;241m=\u001b[39mWHITESPACE\u001b[38;5;241m.\u001b[39mmatch):\n\u001b[1;32m    333\u001b[0m     \u001b[38;5;124;03m\"\"\"Return the Python representation of ``s`` (a ``str`` instance\u001b[39;00m\n\u001b[1;32m    334\u001b[0m \u001b[38;5;124;03m    containing a JSON document).\u001b[39;00m\n\u001b[1;32m    335\u001b[0m \n\u001b[1;32m    336\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 337\u001b[0m     obj, end \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mraw_decode\u001b[49m\u001b[43m(\u001b[49m\u001b[43ms\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43midx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m_w\u001b[49m\u001b[43m(\u001b[49m\u001b[43ms\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mend\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    338\u001b[0m     end \u001b[38;5;241m=\u001b[39m _w(s, end)\u001b[38;5;241m.\u001b[39mend()\n\u001b[1;32m    339\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m end \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mlen\u001b[39m(s):\n",
      "File \u001b[0;32m/opt/conda/lib/python3.8/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[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mscan_once(s, idx)\n\u001b[1;32m    354\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mStopIteration\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 355\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m JSONDecodeError(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mExpecting value\u001b[39m\u001b[38;5;124m\"\u001b[39m, s, err\u001b[38;5;241m.\u001b[39mvalue) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;28mNone\u001b[39m\n\u001b[1;32m    356\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m obj, end\n",
      "\u001b[0;31mJSONDecodeError\u001b[0m: Expecting value: line 1 column 1 (char 0)"
     ]
    }
   ],
   "source": [
    "model = AutoModelForCTC.from_pretrained(\"vitouphy/xls-r-300m-km\").to('cuda')\n",
    "processor = Wav2Vec2Processor.from_pretrained(\"vitouphy/xls-r-300m-km\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3d61ff3b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using custom data configuration default-fbad308ab5a03eb2\n",
      "Reusing dataset csv (/workspace/.cache/huggingface/datasets/csv/default-fbad308ab5a03eb2/0.0.0/6b9057d9e23d9d8a2f05b985917a0da84d70c5dae3d22ddd8a3f22fb01c69d9e)\n"
     ]
    }
   ],
   "source": [
    "common_voice_test = load_dataset('csv', data_files='km_kh_male/line_index_test.csv', split = 'train')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a03f3af4",
   "metadata": {},
   "outputs": [],
   "source": [
    "common_voice_test  = (common_voice_test\n",
    "                      .remove_columns([\"Unnamed: 0\", \"drop\"])\n",
    "                      .rename_column('text', 'sentence'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9c88048b",
   "metadata": {},
   "outputs": [],
   "source": [
    "common_voice_test  = common_voice_test.cast_column(\"path\", Audio(sampling_rate=16_000)).rename_column('path', 'audio')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f3bfc930",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'audio': {'path': '/workspace/xls-r-300m-km/km_kh_male/wavs/khm_1443_3799144408.wav',\n",
       "  'array': array([-1.0600963e-06,  1.2359066e-06, -1.4001107e-06, ...,\n",
       "         -3.1423504e-05,  4.4914182e-06,  0.0000000e+00], dtype=float32),\n",
       "  'sampling_rate': 16000},\n",
       " 'sentence': 'ស៊ី ដាច់ ម៉ូតូ នៅ ពេល ដែល ប្រើ ឱ្យ ឌុប សម្ភារៈ គ្រឿង សង្ហារឹម យក ទៅ ឱ្យ ម៉ូយ នៅ ម្ដុំ វត្ដ សំរោងអណ្ដែត'}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "common_voice_test[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "122a898b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_dataset(batch):\n",
    "    audio = batch[\"audio\"]\n",
    "    \n",
    "    # batched output is \"un-batched\"\n",
    "    batch[\"input_values\"] = processor(np.array(audio[\"array\"]), sampling_rate=audio[\"sampling_rate\"]).input_values[0]\n",
    "    batch[\"input_length\"] = len(batch[\"input_values\"])\n",
    "    \n",
    "    with processor.as_target_processor():\n",
    "        batch[\"labels\"] = processor(batch[\"sentence\"]).input_ids\n",
    "    return batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "153e7f45",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0dd47d98a4e448c9f786ce464348946",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0ex [00:00, ?ex/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "common_voice_test = common_voice_test.map(prepare_dataset, remove_columns=common_voice_test.column_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8947d307",
   "metadata": {},
   "outputs": [],
   "source": [
    "i = 25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3d6b46ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "It is strongly recommended to pass the ``sampling_rate`` argument to this function. Failing to do so can result in silent errors that might be hard to debug.\n"
     ]
    }
   ],
   "source": [
    "input_dict = processor(common_voice_test[i][\"input_values\"], return_tensors=\"pt\", padding=True)\n",
    "logits = model(input_dict.input_values.to(\"cuda\")).logits\n",
    "pred_ids = torch.argmax(logits, dim=-1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d1550ddc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:\n",
      "ក្រុង ប៉យប៉ែត និង ខ្វាះទឹក ស្អាត ប្រើ ចាប់ ពិសព្ ប្តាល ក្រោយ ទៅ\n",
      "\n",
      "Reference:\n",
      "ក្រុង ប៉ោយប៉ែត នឹង ខ្វះ ទឹក ស្អាត ប្រើ ចាប់ ពី សប្តាហ៍ ក្រោយ ទៅ\n"
     ]
    }
   ],
   "source": [
    "print(\"Prediction:\")\n",
    "pred_ids = pred_ids[pred_ids != processor.tokenizer.pad_token_id]\n",
    "print(processor.decode(pred_ids))\n",
    "\n",
    "print(\"\\nReference:\")\n",
    "print(processor.decode(common_voice_test['labels'][i]))\n",
    "# print(common_voice_test_transcription[0][\"sentence\"].lower())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5bbf1c82",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "71b6f502",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}