{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a9612b23",
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://github.com/SpeechColab/GigaSpeech\n",
    "\n",
    "# {\n",
    "#   \"dataset\": \"GigaSpeech\",\n",
    "#   \"language\": \"EN\",\n",
    "#   \"version\": \"v1.0.0\",\n",
    "#   ... ...\n",
    "#   \"audios\": [\n",
    "#     {\n",
    "#       \"title\": \"The Architect of Hollywood\",\n",
    "#       \"url\": \"https://99percentinvisible.org/episode/the-architect-of-hollywood/download\",\n",
    "#       \"path\": \"audio/podcast/P0001/POD0000000025.opus\",\n",
    "#       ... ...\n",
    "#       \"segments\": [\n",
    "#         {\n",
    "#           \"sid\": \"POD0000000025_S0000103\",\n",
    "#           \"speaker\": \"N/A\",\n",
    "#           \"begin_time\": 780.31,\n",
    "#           \"end_time\": 783.13,\n",
    "#           \"text_tn\": \"FOUR O'CLOCK TOMORROW AFTERNOON <COMMA> SAID WILLIAMS <PERIOD>\",\n",
    "#           \"subsets\": [\n",
    "#             \"{XL}\",\n",
    "#             \"{L}\"\n",
    "#           ]\n",
    "#         },\n",
    "#         ... ...\n",
    "#       ],\n",
    "#       ... ...\n",
    "#     },\n",
    "#     ... ...\n",
    "#   ]\n",
    "# }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "352975be",
   "metadata": {},
   "outputs": [],
   "source": [
    "# segment\n",
    "\n",
    "\"original_audio_id\": Optional[str],\n",
    "\"original_audio_offset_s\": float,\n",
    "\n",
    "\"uid\": str,\n",
    "\"duration_s\": float,\n",
    "\"uri\": str, \n",
    "\"transcript\": {\n",
    "    \"tokens\": [  # space separated token\n",
    "        {\n",
    "            \"value\": str,\n",
    "            \"type\": \"text|tag\",\n",
    "            \"speaker_id\": Optional[str],\n",
    "            \"start_s\": Optional[float],\n",
    "            \"end_s\": Optional[float],\n",
    "            ...\n",
    "        }\n",
    "    ],\n",
    "    \"tokens_normalized\": [\n",
    "        # same as \"tokens\"\n",
    "    ],\n",
    "}\n",
    "\"metadata\": Any,\n",
    "    \n",
    "    \n",
    "# [\n",
    "#     {\n",
    "#         \"type\": str,  # quality\n",
    "#         \"value\": Optional[Any],  # 45\n",
    "#         \"start_s\": Optional[float],\n",
    "#         \"end_s\": Optional[float],\n",
    "#         ...\n",
    "#     }\n",
    "# ],\n",
    "# accent?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ad94e21f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9a975839",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "51127622",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e1ccdd7e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b29094c",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}