{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading speech data catalog\n",
      "Reading speech data survey taxonomy\n",
      "Index(['Dataset name', 'Dataset ID', 'Access type', 'Access link',\n",
      "       'Available online', 'License', 'Publisher', 'Repository', 'Languages',\n",
      "       'Creation year', 'ISLRN', 'ISBN', 'LR catalog ID',\n",
      "       'Reference publication', 'Contact point', 'Latest version',\n",
      "       'Last update year', 'Sponsor', 'Price - non-commercial usage',\n",
      "       'Price - commercial usage', 'Purpose and split',\n",
      "       'Size audio total [hours]', 'Size audio transcribed [hours]',\n",
      "       'Size [GB]', 'Speakers', 'Audio recordings', 'Audio segmentation',\n",
      "       'Tokens', 'Unique tokens', 'Automatic QA', 'Manual QA',\n",
      "       'Manual QA scope', 'Transcription coverage', 'Transcription protocol',\n",
      "       'Denormalized transcriptions', 'Transcription and annotation format',\n",
      "       'Domain', 'Speech type', 'Audio collection process',\n",
      "       'Speech recordings source', 'Acoustic environment', 'Audio device',\n",
      "       'Device model', 'Audio format', 'Audio codec', 'Audio channels',\n",
      "       'Sampling rate [Hz]', 'Bits per sample', 'Speaker info', 'Age info',\n",
      "       'Age balance', 'Age distribution notes', 'Gender info',\n",
      "       'Gender balance', 'Gender distribution notes', 'Nativity info',\n",
      "       'Accent info', 'Accent representative', 'Accent distribution notes',\n",
      "       'Education info', 'Occupation info', 'Health info',\n",
      "       'Time alignement annotation', 'Named entities annotation',\n",
      "       'Part of speech annotation', 'Speaker diarization annotation',\n",
      "       'Comment'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from utils import download_tsv_from_google_sheet, load_catalog, load_taxonomy\n",
    "\n",
    "df_cat = load_catalog()\n",
    "df_tax = load_taxonomy()\n",
    "\n",
    "df_cat_available_free = df_cat[(df_cat['Available online'] == 'yes') & (df_cat['Price - non-commercial usage'] == 'free')]\n",
    "df_cat_available_paid = df_cat[(df_cat['Available online'] == 'yes') & (df_cat['Price - non-commercial usage'] != 'free')]\n",
    "\n",
    "print(df_cat.columns)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Number of datasets</th>\n",
       "      <th>Total transcribed [hours]</th>\n",
       "      <th>Percent of total</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Speech type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>read</th>\n",
       "      <td>25</td>\n",
       "      <td>3362.1</td>\n",
       "      <td>56.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conversational</th>\n",
       "      <td>13</td>\n",
       "      <td>1184.0</td>\n",
       "      <td>19.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>various</th>\n",
       "      <td>4</td>\n",
       "      <td>1134.0</td>\n",
       "      <td>18.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>public speech</th>\n",
       "      <td>8</td>\n",
       "      <td>275.0</td>\n",
       "      <td>4.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>no info</th>\n",
       "      <td>3</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Number of datasets  Total transcribed [hours]  \\\n",
       "Speech type                                                     \n",
       "read                            25                     3362.1   \n",
       "conversational                  13                     1184.0   \n",
       "various                          4                     1134.0   \n",
       "public speech                    8                      275.0   \n",
       "no info                          3                       31.0   \n",
       "\n",
       "                Percent of total  \n",
       "Speech type                       \n",
       "read                       56.17  \n",
       "conversational             19.78  \n",
       "various                    18.94  \n",
       "public speech               4.59  \n",
       "no info                     0.52  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from utils import datasets_count_and_total_size\n",
    "col_groupby = ['Speech type']\n",
    "df_datasets_per_speech_type = datasets_count_and_total_size(df_cat, col_groupby)\n",
    "df_datasets_per_speech_type\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Number of datasets</th>\n",
       "      <th>Total transcribed [hours]</th>\n",
       "      <th>Percent of total</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Part of speech annotation</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>no</th>\n",
       "      <td>13</td>\n",
       "      <td>3172</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           Number of datasets  Total transcribed [hours]  \\\n",
       "Part of speech annotation                                                  \n",
       "no                                         13                       3172   \n",
       "\n",
       "                           Percent of total  \n",
       "Part of speech annotation                    \n",
       "no                                    100.0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_datasets_per_meta_paid = datasets_count_and_total_size(df_cat_available_paid, 'Part of speech annotation')\n",
    "df_datasets_per_meta_paid\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             Number of datasets  Total transcribed [hours]  Percent of total\n",
      "Gender info                                                                 \n",
      "yes                          19                     4874.1             81.42\n",
      "no info                      23                      889.0             14.85\n",
      "no                           11                      223.0              3.73\n",
      "          Number of datasets  Total transcribed [hours]  Percent of total\n",
      "Age info                                                                 \n",
      "no info                   33                     4043.0             67.54\n",
      "yes                        8                     1581.0             26.41\n",
      "no                        12                      362.1              6.05\n",
      "             Number of datasets  Total transcribed [hours]  Percent of total\n",
      "Accent info                                                                 \n",
      "no                           49                     4276.1             71.43\n",
      "yes                           4                     1710.0             28.57\n",
      "               Number of datasets  Total transcribed [hours]  Percent of total\n",
      "Nativity info                                                                 \n",
      "no                             33                     3254.0             54.36\n",
      "yes                            12                     2648.1             44.24\n",
      "no info                         8                       84.0              1.40\n",
      "                            Number of datasets  Total transcribed [hours]  \\\n",
      "Time alignement annotation                                                  \n",
      "no                                          48                     4852.1   \n",
      "yes                                          5                     1134.0   \n",
      "\n",
      "                            Percent of total  \n",
      "Time alignement annotation                    \n",
      "no                                     81.06  \n",
      "yes                                    18.94  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n",
      "/home/michal/Development/hugging-face/michaljunczyk/pl-asr-speech-data-survey-analysis/utils.py:48: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_cat[col_sum] = num_values\n"
     ]
    }
   ],
   "source": [
    "from utils import metadata_coverage\n",
    "df_meta_all_flat, df_meta_all_pivot = metadata_coverage(df_cat, df_cat_available_free, df_cat_available_paid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "# compare coverage of free and paid datasets for all metadata\n",
    "\n",
    "sns.set(style=\"whitegrid\")\n",
    "plt.figure(figsize=(10, 5))\n",
    "ax = sns.barplot(x=\"Percent of total\", y=df_meta_all_flat.index, hue=\"Type\", data=df_meta_all_flat)\n",
    "plt.title('Coverage of metadata in free and paid datasets')\n",
    "plt.xlabel('Percent of total')\n",
    "plt.ylabel('Metadata')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"whitegrid\")\n",
    "plt.figure(figsize=(10, 5))\n",
    "ax = sns.barplot(x=\"Number of datasets\", y=df_meta_all_flat.index, hue=\"Type\", data=df_meta_all_flat)\n",
    "plt.title('Number of datasets with metadata info')\n",
    "plt.xlabel('Number of datasets')\n",
    "plt.ylabel('Metadata')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"whitegrid\")\n",
    "plt.figure(figsize=(10, 5))\n",
    "ax = sns.barplot(x=\"Total transcribed [hours]\", y=df_meta_all_flat.index, hue=\"Type\", data=df_meta_all_flat)\n",
    "plt.title('Transcribed speech material with metadata information')\n",
    "plt.xlabel('Amount of transcribed speech')\n",
    "plt.ylabel('Metadata')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "streamlit",
   "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": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}