{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c4f101d6-ca08-4236-9034-c86968fc7830",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Found cached dataset librispeech_asr (/home/jupyter/.cache/huggingface/datasets/librispeech_asr/all/2.1.0/cff5df6e7955c80a67f80e27e7e655de71c689e2d2364bece785b972acb37fe7)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c87424fb2904b1894800b3ec3de48ec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loading cached shuffled indices for dataset at /home/jupyter/.cache/huggingface/datasets/librispeech_asr/all/2.1.0/cff5df6e7955c80a67f80e27e7e655de71c689e2d2364bece785b972acb37fe7/cache-c0ea02798a773cf7.arrow\n"
     ]
    }
   ],
   "source": [
    "# load the dataset\n",
    "from datasets import load_dataset, DatasetDict\n",
    "import torch \n",
    "\n",
    "# fix a random seed to make sure different run give the same dataset\n",
    "seed = 17\n",
    "\n",
    "#  load the librispeech dataset and shuffle it \n",
    "dataset = load_dataset(\"librispeech_asr\",\"all\")['train.clean.100']\n",
    "dataset = dataset.shuffle(seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c5153afc-4ce3-45a6-9480-a33c636c7706",
   "metadata": {},
   "outputs": [],
   "source": [
    "# select enough element to get elements to get to 20 minutes\n",
    "i_10mn = 0\n",
    "duration = 0 \n",
    "while duration < 600: # until we reach 1200s \n",
    "    duration += dataset[i_10mn]['audio']['array'].shape[0]/dataset[i_10mn]['audio']['sampling_rate']\n",
    "    i_10mn += 1\n",
    "\n",
    "librispeech_10mn = dataset.select(range(i_10mn))    \n",
    "\n",
    "# select elements on the following to get 1h\n",
    "i_1h = i_10mn\n",
    "duration=0\n",
    "while duration < 3600: # until we reach 1200s \n",
    "    duration += dataset[i_1h]['audio']['array'].shape[0]/dataset[i_1h]['audio']['sampling_rate']\n",
    "    i_1h += 1\n",
    "    \n",
    "librispeech_1h = dataset.select(range(i_10mn,i_1h))\n",
    "\n",
    "# select elements to get to 2h\n",
    "i_2h = i_1h\n",
    "duration = 0\n",
    "while duration < 7200: # until we reach 1200s \n",
    "    duration += dataset[i_2h]['audio']['array'].shape[0]/dataset[i_2h]['audio']['sampling_rate']\n",
    "    i_2h += 1\n",
    "    \n",
    "librispeech_2h = dataset.select(range(i_1h,i_2h))\n",
    "\n",
    "# put them together\n",
    "librispeech_tiny = DatasetDict({'10mn':librispeech_10mn,\n",
    "                                '1h': librispeech_1h,\n",
    "                                '2h': librispeech_2h\n",
    "                               }\n",
    "                              )\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5cc28e11-803c-4760-9005-516157eb8945",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatasetDict({\n",
       "    10mn: Dataset({\n",
       "        features: ['file', 'audio', 'text', 'speaker_id', 'chapter_id', 'id'],\n",
       "        num_rows: 50\n",
       "    })\n",
       "    1h: Dataset({\n",
       "        features: ['file', 'audio', 'text', 'speaker_id', 'chapter_id', 'id'],\n",
       "        num_rows: 284\n",
       "    })\n",
       "    2h: Dataset({\n",
       "        features: ['file', 'audio', 'text', 'speaker_id', 'chapter_id', 'id'],\n",
       "        num_rows: 576\n",
       "    })\n",
       "})"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "librispeech_tiny"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e08cddc6-364c-45cf-ac7b-d2a4d0ec8e65",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Pushing split 10mn to the Hub.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2135c4ef2af4b72803dddaff88cb092",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d42ce27a96f448cc924703bfe0b30065",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Pushing dataset shards to the dataset hub:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Pushing split 1h to the Hub.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57a8cb106ff14b1fb186c3e86122f497",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a3c8ea385d1a46108db9a51144e82f66",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Pushing dataset shards to the dataset hub:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Pushing split 2h to the Hub.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33a0b8ce30d04b5c87e72efbee8e40d1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1ac0a669b0e54dffae68dce597d7486b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Pushing dataset shards to the dataset hub:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "librispeech_tiny.push_to_hub('Isma/librispeech_tiny')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d938122a-3c6b-4585-951a-6e9aac77f804",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50, 334, 910, 7208.405187500002)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#i_10mn,i_1h, i_2h, duration "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "df7b7a37-1b07-4454-bb2e-f381b938ed41",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "71ae1e94-4d66-41db-9dd2-f54eaed61187",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['path', 'array', 'sampling_rate'])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#dataset[0]['audio'].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7a934fa2-8273-496b-b20c-9431cc965181",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "65d22ed9-f920-48eb-8ea4-670111c5dd47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, 127)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "i_1h = i_10mn\n",
    "i_1h = 10\n",
    "i_1h, i_10mn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "99bfbd13-93ca-4d64-bfe5-e6b883b24cef",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "environment": {
   "kernel": "python3",
   "name": "common-cu110.m102",
   "type": "gcloud",
   "uri": "gcr.io/deeplearning-platform-release/base-cu110:m102"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}