{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "genius_hq\n",
      "youtube_music\n",
      "imslp\n",
      "discogs\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "# combine all audio production features\n",
    "combined_df = pd.DataFrame()\n",
    "for source in [\"genius_hq\", \"youtube_music\", \"imslp\", \"discogs\"]:\n",
    "    print(source)\n",
    "    filepath = f\"/home/christian/code/christian/metadata/{source}_audio_production_features_v2.csv\"\n",
    "    df = pd.read_csv(filepath, index_col=0)\n",
    "    if df.empty:\n",
    "        continue\n",
    "    if combined_df.empty:\n",
    "        combined_df = df\n",
    "    else:\n",
    "        combined_df = pd.concat([combined_df, df])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define the mean and std for each feature (hard code for reproducibility)\n",
    "feature_stats = {\n",
    "    \"spectral_centroid\": {\"mean\": 3100.0, \"std\": 900.0},\n",
    "    \"bass\": {\"mean\": 0.3, \"std\": 0.1},\n",
    "    \"mid\": {\"mean\": 0.6, \"std\": 0.1}, \n",
    "    \"high\": {\"mean\": 0.6, \"std\": 0.3},\n",
    "    \"stereo_width\": {\"mean\": 0.2, \"std\": 0.1},\n",
    "    \"spectral_flatness\": {\"mean\": 0.08, \"std\": 0.05},\n",
    "    \"crest_factor\": {\"mean\": 1.8, \"std\": 0.5},\n",
    "    \"silence_percentage\": {\"mean\": 1.5, \"std\": 3.0},\n",
    "    \"loudness\": {\"mean\": -12.0, \"std\": 5.0}\n",
    "}\n",
    "\n",
    "# normalize each feature using the predefined means and stds\n",
    "for feature, stats in feature_stats.items():\n",
    "    combined_df[f\"{feature}_normalized\"] = (combined_df[feature] - stats[\"mean\"]) / stats[\"std\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "spectral_centroid\n",
      "3136.4829300455526\n",
      "897.9225185300384\n",
      "\n",
      "bass\n",
      "0.28948773623689245\n",
      "0.11040897459049791\n",
      "\n",
      "mid\n",
      "0.6299259507228461\n",
      "0.15238813034394916\n",
      "\n",
      "high\n",
      "0.6010692517751821\n",
      "0.32611605278276656\n",
      "\n",
      "stereo_width\n",
      "0.1708744406520693\n",
      "0.1057837962040145\n",
      "\n",
      "spectral_flatness\n",
      "0.08299266052756754\n",
      "0.04755443548658767\n",
      "\n",
      "crest_factor\n",
      "1.8272148970512998\n",
      "0.4506846975985857\n",
      "\n",
      "silence_percentage\n",
      "1.4253822087225672\n",
      "3.0571802788473197\n",
      "\n",
      "loudness\n",
      "-12.363741194642557\n",
      "4.879313367440992\n",
      "\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>spectral_centroid</th>\n",
       "      <th>bass</th>\n",
       "      <th>mid</th>\n",
       "      <th>high</th>\n",
       "      <th>crest_factor</th>\n",
       "      <th>stereo_width</th>\n",
       "      <th>spectral_flatness</th>\n",
       "      <th>silence_percentage</th>\n",
       "      <th>loudness</th>\n",
       "      <th>spectral_centroid_normalized</th>\n",
       "      <th>bass_normalized</th>\n",
       "      <th>mid_normalized</th>\n",
       "      <th>high_normalized</th>\n",
       "      <th>stereo_width_normalized</th>\n",
       "      <th>spectral_flatness_normalized</th>\n",
       "      <th>crest_factor_normalized</th>\n",
       "      <th>silence_percentage_normalized</th>\n",
       "      <th>loudness_normalized</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "      <td>1.571600e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.136483e+03</td>\n",
       "      <td>2.894877e-01</td>\n",
       "      <td>6.299260e-01</td>\n",
       "      <td>6.010693e-01</td>\n",
       "      <td>1.827215e+00</td>\n",
       "      <td>1.708744e-01</td>\n",
       "      <td>8.299266e-02</td>\n",
       "      <td>1.425382e+00</td>\n",
       "      <td>-1.236374e+01</td>\n",
       "      <td>8.374323e-16</td>\n",
       "      <td>-2.148192e-15</td>\n",
       "      <td>-3.660053e-15</td>\n",
       "      <td>1.133429e-15</td>\n",
       "      <td>-4.774652e-16</td>\n",
       "      <td>4.281845e-14</td>\n",
       "      <td>8.093634e-15</td>\n",
       "      <td>-3.010249e-16</td>\n",
       "      <td>-1.651125e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8.979225e+02</td>\n",
       "      <td>1.104090e-01</td>\n",
       "      <td>1.523881e-01</td>\n",
       "      <td>3.261161e-01</td>\n",
       "      <td>4.506847e-01</td>\n",
       "      <td>1.057838e-01</td>\n",
       "      <td>4.755444e-02</td>\n",
       "      <td>3.057180e+00</td>\n",
       "      <td>4.879313e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-1.842068e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>5.573746e-05</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-8.000000e+01</td>\n",
       "      <td>-3.493044e+00</td>\n",
       "      <td>-2.621958e+00</td>\n",
       "      <td>-4.133694e+00</td>\n",
       "      <td>-1.843115e+00</td>\n",
       "      <td>-1.615318e+00</td>\n",
       "      <td>-1.744042e+00</td>\n",
       "      <td>-4.492697e+01</td>\n",
       "      <td>-4.662408e-01</td>\n",
       "      <td>-1.386184e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.497555e+03</td>\n",
       "      <td>2.217019e-01</td>\n",
       "      <td>5.340773e-01</td>\n",
       "      <td>3.509817e-01</td>\n",
       "      <td>1.525360e+00</td>\n",
       "      <td>9.905377e-02</td>\n",
       "      <td>4.815602e-02</td>\n",
       "      <td>8.413967e-02</td>\n",
       "      <td>-1.473758e+01</td>\n",
       "      <td>-7.115622e-01</td>\n",
       "      <td>-6.139520e-01</td>\n",
       "      <td>-6.289772e-01</td>\n",
       "      <td>-7.668668e-01</td>\n",
       "      <td>-6.789383e-01</td>\n",
       "      <td>-7.325635e-01</td>\n",
       "      <td>-6.697704e-01</td>\n",
       "      <td>-4.387188e-01</td>\n",
       "      <td>-4.865108e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.109054e+03</td>\n",
       "      <td>3.001570e-01</td>\n",
       "      <td>6.374277e-01</td>\n",
       "      <td>5.917474e-01</td>\n",
       "      <td>1.747335e+00</td>\n",
       "      <td>1.564386e-01</td>\n",
       "      <td>7.868735e-02</td>\n",
       "      <td>7.085917e-01</td>\n",
       "      <td>-1.119873e+01</td>\n",
       "      <td>-3.054715e-02</td>\n",
       "      <td>9.663358e-02</td>\n",
       "      <td>4.922783e-02</td>\n",
       "      <td>-2.858436e-02</td>\n",
       "      <td>-1.364652e-01</td>\n",
       "      <td>-9.053434e-02</td>\n",
       "      <td>-1.772409e-01</td>\n",
       "      <td>-2.344613e-01</td>\n",
       "      <td>2.387655e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3.678957e+03</td>\n",
       "      <td>3.635084e-01</td>\n",
       "      <td>7.311115e-01</td>\n",
       "      <td>8.155239e-01</td>\n",
       "      <td>2.055750e+00</td>\n",
       "      <td>2.230073e-01</td>\n",
       "      <td>1.117401e-01</td>\n",
       "      <td>1.855895e+00</td>\n",
       "      <td>-8.989421e+00</td>\n",
       "      <td>6.041438e-01</td>\n",
       "      <td>6.704226e-01</td>\n",
       "      <td>6.639986e-01</td>\n",
       "      <td>6.576022e-01</td>\n",
       "      <td>4.928247e-01</td>\n",
       "      <td>6.045165e-01</td>\n",
       "      <td>5.070837e-01</td>\n",
       "      <td>1.408203e-01</td>\n",
       "      <td>6.915564e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.762906e+04</td>\n",
       "      <td>7.324730e+00</td>\n",
       "      <td>6.153669e+00</td>\n",
       "      <td>5.984927e+00</td>\n",
       "      <td>1.269876e+01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>9.699697e-01</td>\n",
       "      <td>1.000000e+02</td>\n",
       "      <td>5.952439e+01</td>\n",
       "      <td>1.614012e+01</td>\n",
       "      <td>6.371984e+01</td>\n",
       "      <td>3.624786e+01</td>\n",
       "      <td>1.650902e+01</td>\n",
       "      <td>7.837926e+00</td>\n",
       "      <td>1.865183e+01</td>\n",
       "      <td>2.412229e+01</td>\n",
       "      <td>3.224364e+01</td>\n",
       "      <td>1.473325e+01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       spectral_centroid          bass           mid          high  \\\n",
       "count       1.571600e+07  1.571600e+07  1.571600e+07  1.571600e+07   \n",
       "mean        3.136483e+03  2.894877e-01  6.299260e-01  6.010693e-01   \n",
       "std         8.979225e+02  1.104090e-01  1.523881e-01  3.261161e-01   \n",
       "min         0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "25%         2.497555e+03  2.217019e-01  5.340773e-01  3.509817e-01   \n",
       "50%         3.109054e+03  3.001570e-01  6.374277e-01  5.917474e-01   \n",
       "75%         3.678957e+03  3.635084e-01  7.311115e-01  8.155239e-01   \n",
       "max         1.762906e+04  7.324730e+00  6.153669e+00  5.984927e+00   \n",
       "\n",
       "       crest_factor  stereo_width  spectral_flatness  silence_percentage  \\\n",
       "count  1.571600e+07  1.571600e+07       1.571600e+07        1.571600e+07   \n",
       "mean   1.827215e+00  1.708744e-01       8.299266e-02        1.425382e+00   \n",
       "std    4.506847e-01  1.057838e-01       4.755444e-02        3.057180e+00   \n",
       "min   -1.842068e+01  0.000000e+00       5.573746e-05        0.000000e+00   \n",
       "25%    1.525360e+00  9.905377e-02       4.815602e-02        8.413967e-02   \n",
       "50%    1.747335e+00  1.564386e-01       7.868735e-02        7.085917e-01   \n",
       "75%    2.055750e+00  2.230073e-01       1.117401e-01        1.855895e+00   \n",
       "max    1.269876e+01  1.000000e+00       9.699697e-01        1.000000e+02   \n",
       "\n",
       "           loudness  spectral_centroid_normalized  bass_normalized  \\\n",
       "count  1.571600e+07                  1.571600e+07     1.571600e+07   \n",
       "mean  -1.236374e+01                  8.374323e-16    -2.148192e-15   \n",
       "std    4.879313e+00                  1.000000e+00     1.000000e+00   \n",
       "min   -8.000000e+01                 -3.493044e+00    -2.621958e+00   \n",
       "25%   -1.473758e+01                 -7.115622e-01    -6.139520e-01   \n",
       "50%   -1.119873e+01                 -3.054715e-02     9.663358e-02   \n",
       "75%   -8.989421e+00                  6.041438e-01     6.704226e-01   \n",
       "max    5.952439e+01                  1.614012e+01     6.371984e+01   \n",
       "\n",
       "       mid_normalized  high_normalized  stereo_width_normalized  \\\n",
       "count    1.571600e+07     1.571600e+07             1.571600e+07   \n",
       "mean    -3.660053e-15     1.133429e-15            -4.774652e-16   \n",
       "std      1.000000e+00     1.000000e+00             1.000000e+00   \n",
       "min     -4.133694e+00    -1.843115e+00            -1.615318e+00   \n",
       "25%     -6.289772e-01    -7.668668e-01            -6.789383e-01   \n",
       "50%      4.922783e-02    -2.858436e-02            -1.364652e-01   \n",
       "75%      6.639986e-01     6.576022e-01             4.928247e-01   \n",
       "max      3.624786e+01     1.650902e+01             7.837926e+00   \n",
       "\n",
       "       spectral_flatness_normalized  crest_factor_normalized  \\\n",
       "count                  1.571600e+07             1.571600e+07   \n",
       "mean                   4.281845e-14             8.093634e-15   \n",
       "std                    1.000000e+00             1.000000e+00   \n",
       "min                   -1.744042e+00            -4.492697e+01   \n",
       "25%                   -7.325635e-01            -6.697704e-01   \n",
       "50%                   -9.053434e-02            -1.772409e-01   \n",
       "75%                    6.045165e-01             5.070837e-01   \n",
       "max                    1.865183e+01             2.412229e+01   \n",
       "\n",
       "       silence_percentage_normalized  loudness_normalized  \n",
       "count                   1.571600e+07         1.571600e+07  \n",
       "mean                   -3.010249e-16        -1.651125e-15  \n",
       "std                     1.000000e+00         1.000000e+00  \n",
       "min                    -4.662408e-01        -1.386184e+01  \n",
       "25%                    -4.387188e-01        -4.865108e-01  \n",
       "50%                    -2.344613e-01         2.387655e-01  \n",
       "75%                     1.408203e-01         6.915564e-01  \n",
       "max                     3.224364e+01         1.473325e+01  "
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cols = [\n",
    "    \"spectral_centroid\",\n",
    "    \"bass\", \n",
    "    \"mid\", \n",
    "    \"high\", \n",
    "    \"stereo_width\", \n",
    "    \"spectral_flatness\", \n",
    "    \"crest_factor\",\n",
    "    \"silence_percentage\",\n",
    "    \"loudness\",\n",
    "]\n",
    "\n",
    "for col in cols:\n",
    "    print(col)\n",
    "    # create new column with standard normalized values\n",
    "    print(combined_df[col].mean())\n",
    "    print(combined_df[col].std())\n",
    "    combined_df[f\"{col}_normalized\"] = (combined_df[col] - combined_df[col].mean()) / combined_df[col].std()\n",
    "    print()\n",
    "\n",
    "combined_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x2000 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For a single feature:\n",
    "def get_feature_bounds(df, feature, multiplier=1.5):\n",
    "    Q1 = df[feature].quantile(0.25)\n",
    "    Q3 = df[feature].quantile(0.75)\n",
    "    IQR = Q3 - Q1\n",
    "    lower = Q1 - multiplier * IQR\n",
    "    upper = Q3 + multiplier * IQR\n",
    "    return lower, upper\n",
    "\n",
    "def get_feature_bounds_std(df, feature, multiplier=2.0):\n",
    "    std = df[feature].std()\n",
    "    mean = df[feature].mean()\n",
    "    lower = mean - multiplier * std\n",
    "    upper = mean + multiplier * std\n",
    "    return lower, upper\n",
    "\n",
    "# make a multiplot histogram of all the normalized columns\n",
    "num_bins = 1000\n",
    "fig, axs = plt.subplots(nrows=len(cols), ncols=1, figsize=(10, 20), sharex=False, sharey=False)\n",
    "for i, col in enumerate(cols):\n",
    "    axs[i].hist(combined_df[f\"{col}\"], bins=num_bins)\n",
    "    lower, upper = get_feature_bounds(combined_df, f\"{col}\", multiplier=1.5)\n",
    "    axs[i].axvline(x=lower, color='r', linestyle='--', label=f'Lower Bound: {lower:.2f}')\n",
    "    axs[i].axvline(x=upper, color='r', linestyle='--', label=f'Upper Bound: {upper:.2f}')\n",
    "    axs[i].set_title(col)\n",
    "\n",
    "    lower, upper = get_feature_bounds_std(combined_df, f\"{col}\", multiplier=2.0)\n",
    "    #print(f\"Values outside bounds: {df[~df[f'{col}'].between(lower, upper)][f'{col}'].count()} ({df[~df[f'{col}'].between(lower, upper)][f'{col}'].count() / len(df):.2%})\")\n",
    "    axs[i].axvline(x=lower, color='g', linestyle='--', label=f'Lower Bound: {lower:.2f}')\n",
    "    axs[i].axvline(x=upper, color='g', linestyle='--', label=f'Upper Bound: {upper:.2f}')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "def describe_spectrum(bass_energy, mid_energy, high_energy):\n",
    "    # Calculate ratios\n",
    "    if mid_energy == 0:  # Avoid division by zero\n",
    "        mid_energy = 1e-10\n",
    "    if high_energy == 0:  # Avoid division by zero\n",
    "        high_energy = 1e-10\n",
    "\n",
    "    bass_mid_ratio = bass_energy / mid_energy\n",
    "    bass_high_ratio = bass_energy / high_energy\n",
    "    mid_high_ratio = mid_energy / high_energy\n",
    "\n",
    "    # Descriptors based on the ratios\n",
    "    descriptors = []\n",
    "\n",
    "    # Bass dominant\n",
    "    if bass_mid_ratio > 1.5 and bass_high_ratio > 1.5:\n",
    "        descriptors.extend([\"bassy\", \"boomy\", \"thumpy\"])\n",
    "\n",
    "    # Mid dominant\n",
    "    if bass_mid_ratio < 0.67 and mid_high_ratio > 1.5:\n",
    "        descriptors.extend([\"warm\", \"full-bodied\", \"nasal\", \"boxy\"])\n",
    "\n",
    "    # High dominant\n",
    "    if bass_high_ratio < 0.67 and mid_high_ratio < 0.67:\n",
    "        descriptors.extend([\"bright\", \"sibilant\", \"tinny\"])\n",
    "\n",
    "    # Balanced audio\n",
    "    if 0.9 <= bass_mid_ratio <= 1.1 and 0.9 <= bass_high_ratio <= 1.1 and 0.9 <= mid_high_ratio <= 1.1:\n",
    "        descriptors.extend([\"balanced\", \"even\", \"neutral\"])\n",
    "\n",
    "    # Muddy (overlapping with mid dominant but more specific to lack of clarity)\n",
    "    if bass_mid_ratio > 0.8 and mid_high_ratio < 1.2:\n",
    "        descriptors.append(\"muddy\")\n",
    "\n",
    "    # Detailed or clear, considering mid-high clarity\n",
    "    if mid_high_ratio > 1.2:\n",
    "        descriptors.append(\"detailed\")\n",
    "\n",
    "    # Forward (applicable if mids are clearly dominant over highs and slightly over bass)\n",
    "    if mid_high_ratio > 1.2 and bass_mid_ratio < 1.2:\n",
    "        descriptors.append(\"forward\")\n",
    "\n",
    "    return descriptors\n",
    "\n",
    "# given a row of audio features, return a list of descriptor features\n",
    "def get_descriptor_features(row):\n",
    "    tags = []\n",
    "\n",
    "    stereo_width = row[\"stereo_width\"]\n",
    "    stereo_width_normalized = row[\"stereo_width_normalized\"] \n",
    "    crest_factor = row[\"crest_factor\"]\n",
    "    crest_factor_normalized = row[\"crest_factor_normalized\"]\n",
    "    spectral_flatness = row[\"spectral_flatness\"]\n",
    "    spectral_flatness_normalized = row[\"spectral_flatness_normalized\"]\n",
    "    spectral_centroid = row[\"spectral_centroid\"]\n",
    "    spectral_centroid_normalized = row[\"spectral_centroid_normalized\"]\n",
    "    bass = row[\"bass\"]\n",
    "    mid = row[\"mid\"]\n",
    "    high = row[\"high\"]\n",
    "    loudness = row[\"loudness\"]\n",
    "\n",
    "    # -------- spectral flatness --------\n",
    "    if not np.isnan(spectral_flatness) and not np.isnan(spectral_flatness_normalized):\n",
    "        if spectral_flatness < 0.05:\n",
    "            tags.append(\"rolled-off\")\n",
    "        elif spectral_flatness > 0.2:\n",
    "            tags.append(\"noisy\")\n",
    "        tags.append(f\"sf:{round(spectral_flatness_normalized)}\")\n",
    "\n",
    "    # -------- spectrum analysis --------\n",
    "    if not np.isnan(bass) and not np.isnan(mid) and not np.isnan(high):\n",
    "        tags.extend(describe_spectrum(bass, mid, high))\n",
    "\n",
    "    # -------- spectral centroid --------\n",
    "    if not np.isnan(spectral_centroid) and not np.isnan(spectral_centroid_normalized):\n",
    "        if spectral_centroid < 1500:\n",
    "            tags.append(\"very warm\")\n",
    "            tags.append(\"very dark\")\n",
    "        elif spectral_centroid >= 1500 and spectral_centroid < 2000:\n",
    "            tags.append(\"warm\")\n",
    "            tags.append(\"dark\")\n",
    "        elif spectral_centroid >= 2000 and spectral_centroid < 3000:\n",
    "            pass\n",
    "        elif spectral_centroid >= 3750 and spectral_centroid < 4500:\n",
    "            tags.append(\"bright\")\n",
    "        else:\n",
    "            tags.append(\"bright\")\n",
    "            tags.append(\"very bright\")\n",
    "            tags.append(\"sharp\")\n",
    "        tags.append(f\"sc:{round(spectral_centroid_normalized)}\")\n",
    "\n",
    "    # -------- stereo width --------\n",
    "    if not np.isnan(stereo_width) and not np.isnan(stereo_width_normalized):\n",
    "        if stereo_width < 0.05:\n",
    "            tags.append(\"mono\")\n",
    "        elif stereo_width >= 0.05 and stereo_width < 0.1:\n",
    "            tags.append(\"narrow\")\n",
    "        elif stereo_width >= 0.1 and stereo_width < 0.3:\n",
    "            tags.append(\"stereo\")\n",
    "        else:\n",
    "            tags.append(\"stereo\")\n",
    "            tags.append(\"wide stereo\")\n",
    "            tags.append(\"wide\")\n",
    "        tags.append(f\"sw:{round(stereo_width_normalized)}\")\n",
    "\n",
    "    # -------- crest factor --------\n",
    "    if not np.isnan(crest_factor) and not np.isnan(crest_factor_normalized):\n",
    "        if crest_factor < 1.0:\n",
    "            tags.append(\"slammed\")\n",
    "            tags.append(\"maximized loudness\")\n",
    "            tags.append(\"very compressed\")\n",
    "        elif crest_factor >= 1.0 and crest_factor < 1.5:\n",
    "            tags.append(\"compressed\")\n",
    "        elif crest_factor >= 1.5 and crest_factor < 2.5:\n",
    "            tags.append(\"dynamic\")\n",
    "        else:\n",
    "            tags.append(\"very dynamic\")\n",
    "            tags.append(\"dynamic\")\n",
    "        tags.append(f\"cf:{round(crest_factor_normalized)}\")\n",
    "\n",
    "    return tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load metas and make into metas map\n",
    "from suno_utils.utils.text import read_jsonl\n",
    "metas_filepath = \"/home/christian/code/christian/metadata/genius_hq_metas.jsonl\"\n",
    "metas = read_jsonl(metas_filepath)\n",
    "metas_map = {meta[\"id\"]: meta for meta in metas}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort combined_df by spectral centroid\n",
    "combined_df = combined_df.sort_values(by=\"spectral_centroid\", ascending=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "spectral_centroid                3652.870361\n",
      "bass                                0.246464\n",
      "mid                                 0.757266\n",
      "high                                0.842247\n",
      "crest_factor                        1.478852\n",
      "stereo_width                        0.255651\n",
      "spectral_flatness                   0.096444\n",
      "silence_percentage                  3.985341\n",
      "loudness                           -9.986890\n",
      "spectral_centroid_normalized        0.614300\n",
      "bass_normalized                    -0.535362\n",
      "mid_normalized                      1.572659\n",
      "high_normalized                     0.807490\n",
      "stereo_width_normalized             0.556511\n",
      "spectral_flatness_normalized        0.328871\n",
      "crest_factor_normalized            -0.642296\n",
      "silence_percentage_normalized       0.828447\n",
      "loudness_normalized                 0.402622\n",
      "Name: 278f0149-f642-4e6f-a4c1-6e92e0faf5be, dtype: float64\n",
      "278f0149-f642-4e6f-a4c1-6e92e0faf5be\n",
      "['sf:0', 'bright', 'very bright', 'sharp', 'sc:1', 'stereo', 'sw:1', 'compressed', 'cf:-1']\n",
      "download: s3://suno-data/datasets/harvest/genius_hq/audio/Qx6jBKzdfeU.webm to tmp/audio.wav\n",
      "/home/christian/code/christian/notebooks/tmp/audio.wav\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "idx = 6000 #np.random.randint(0, len(combined_df))\n",
    "\n",
    "meta_id = combined_df.iloc[idx].name\n",
    "print(combined_df.iloc[idx])\n",
    "print(meta_id)\n",
    "print(get_descriptor_features(combined_df.iloc[idx]))\n",
    "\n",
    "s3_filepath = metas_map[meta_id][\"audio_filepath\"]\n",
    "out_filepath = f\"/home/christian/code/christian/notebooks/tmp/audio.wav\"\n",
    "os.system(f\"aws s3 cp {s3_filepath} {out_filepath}\")\n",
    "print(out_filepath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run this over all rows and collect all tags\n",
    "from tqdm import tqdm\n",
    "df_new = combined_df.copy()\n",
    "#df_new = df_new.iloc[:100]\n",
    "# itererate over a subset of rows\n",
    "results = [get_descriptor_features(row) for _, row in tqdm(df_new.iterrows())]\n",
    "print(results)\n",
    "df_new[\"tags\"] = results\n",
    "df_new.head()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count the occurrences of each tag\n",
    "tag_counts = df_new[\"tags\"].explode().value_counts()\n",
    "print(tag_counts)\n",
    "\n",
    "# get the top 10 tags\n",
    "top_tags = tag_counts.head(10)\n",
    "print(top_tags)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "suno_env",
   "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}