CleanedGroupProjectPred.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4a852031",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/lib/python3.10/site-packages/pydub/utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work\n",
      "  warn(\"Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work\", RuntimeWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import colors\n",
    "import scipy.io.wavfile as wavfile\n",
    "from scipy.signal import spectrogram\n",
    "from scipy.fftpack import fft\n",
    "from pydub import AudioSegment\n",
    "import os\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import ParameterGrid\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.svm import SVC\n",
    "import librosa\n",
    "from scipy.stats import zscore\n",
    "from warnings import simplefilter\n",
    "simplefilter(action=\"ignore\", category=pd.errors.PerformanceWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e1bf3495",
   "metadata": {},
   "outputs": [],
   "source": [
    "dev_path = 'dsl_data/development.csv'\n",
    "eval_path = 'dsl_data/evaluation.csv'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1a0d7879",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_dataframe(path):\n",
    "    df = pd.read_csv(path)\n",
    "    # Dataframe exploration\n",
    "    print('***** DATAFRAME INFO *****\\n')\n",
    "    print(df.info())\n",
    "    print('\\n\\n***** HEAD *****\\n')\n",
    "    print(df.head())\n",
    "    print('\\n\\n ***** POSSIBLE CLASSIFICATION OUTCOMES *****\\n')\n",
    "    #print(df.groupby(['action', 'object']).size())\n",
    "    sampleRates = list()\n",
    "    audios = list()\n",
    "    for i in range(0, df.shape[0]):\n",
    "        audio_path = df['path'].loc[i]\n",
    "        sampleRates.append(get_sample_rate(audio_path))\n",
    "        audios.append(get_audio_data(audio_path))\n",
    "    df['SampleRates'] = sampleRates\n",
    "    df['AudioData'] = audios\n",
    "    print('\\n\\n ***** SAMPLE RATES *****')\n",
    "    print(df.groupby(['SampleRates']).size())\n",
    "    \n",
    "    print('\\n\\n ***** Self-reported fluency level *****')\n",
    "    print(df.groupby(['Self-reported fluency level ']).size())\n",
    "    \n",
    "    print('\\n\\n ***** First Language spoken *****')\n",
    "    print(df.groupby(['First Language spoken']).size())\n",
    "    \n",
    "    print('\\n\\n ***** Current language used for work/school *****')\n",
    "    print(df.groupby(['Current language used for work/school']).size())\n",
    "    \n",
    "    print('\\n\\n ***** Gender *****')\n",
    "    print(df.groupby(['gender']).size())\n",
    "    print('\\n\\n ***** Age range *****')\n",
    "    print(df.groupby(['ageRange']).size())\n",
    "    return df\n",
    "\n",
    "def get_sample_rate(filepath):\n",
    "    sr, audio = wavfile.read(filepath)\n",
    "    return sr\n",
    "\n",
    "def get_audio_data(filepath):\n",
    "    sr, audio = wavfile.read(filepath)\n",
    "    return audio\n",
    "\n",
    "\n",
    "def edit_sample_rates(df, goal_sr):\n",
    "    for index, row in df.iterrows():\n",
    "    # Load the audio file\n",
    "        audio = AudioSegment.from_file(row['path'])\n",
    "    # Check if the sample rate is different from the target sample rate\n",
    "        if audio.frame_rate != goal_sr:\n",
    "            # Convert the sample rate to the target sample rate\n",
    "            audio = audio.set_frame_rate(goal_sr)\n",
    "            # get file format\n",
    "            file_format = os.path.splitext(row['path'])[1][1:]\n",
    "            # Save the converted audio\n",
    "            audio.export(row['path'], format=file_format)\n",
    "    sampleRates = list()\n",
    "    for i in range(0, df.shape[0]):\n",
    "        audio_path = df['path'].loc[i]\n",
    "        sampleRates.append(get_sample_rate(audio_path))\n",
    "    df['SampleRates'] = sampleRates\n",
    "    print(df.groupby(['SampleRates']).size())\n",
    "\n",
    "# definition of functions to represent each audio \n",
    "\n",
    "def plot_spectrogram(audio_data, sr):\n",
    "    frequencies, times, spec = spectrogram(audio_data, fs=sr)\n",
    "    plt.pcolormesh(times, frequencies, 10*np.log10(spec)) #changed the scale to dB\n",
    "    plt.colorbar()\n",
    "    plt.xlabel(\"Time (s)\")\n",
    "    plt.ylabel(\"Frequency (Hz)\")\n",
    "    plt.show()\n",
    "    return spec, len(audio)\n",
    "\n",
    "\n",
    "def plot_time_dimension(audio_data, sr):\n",
    "    audioArray = np.array(audio_data)\n",
    "    duration = len(audio) / sr\n",
    "    # Create the time axis in seconds\n",
    "    time = np.linspace(0, duration, len(audio))\n",
    "    plt.plot(time, audioArray)\n",
    "    plt.xlabel(\"Time (s)\")\n",
    "    plt.ylabel(\"Amplitude\")\n",
    "    plt.show()\n",
    "    \n",
    "def plot_freq_dimension(audio_data, sr):\n",
    "    audioArray = np.array(audio_data)\n",
    "    freqArray = np.abs(fft(audioArray))\n",
    "    decibels = 20 * np.log10(freqArray)\n",
    "    plt.figure(figsize=(10,4))\n",
    "    plt.plot(decibels)\n",
    "    plt.xlabel(\"Frequency (Hz)\")\n",
    "    plt.ylabel(\"Magnitude (dB)\")\n",
    "    plt.show()\n",
    "    \n",
    "\n",
    "def encode_categorical(df):\n",
    "    #ordinal encoding of categorical features\n",
    "    catFeatures = ['Self-reported fluency level ', 'First Language spoken', 'Current language used for work/school', 'gender', 'ageRange']\n",
    "    dicts = list()\n",
    "    dic_fluency = {'native' : 0, 'advanced' : 1, 'intermediate':2, 'basic': 3}\n",
    "    dicts.append(dic_fluency)\n",
    "    dic_first_lan = {'English (United States)': 0, 'English (Canada)': 1, 'French (Canada)' : 2, 'Spanish (Venezuela)':3, 'Telugu':4 }\n",
    "    dicts.append(dic_first_lan)\n",
    "    dic_current = {'English (United States)': 0, 'English (Canada)': 1, 'English (Australia)' : 2, 'Spanish (Venezuela)':3 }\n",
    "    dicts.append(dic_current)\n",
    "    dic_gender = {'female': 0, 'male': 1 }\n",
    "    dicts.append(dic_gender)\n",
    "    dic_age = {'22-40':0, '41-65':1, '65+': 3}\n",
    "    dicts.append(dic_age)\n",
    "    for i in range(0, len(catFeatures)):\n",
    "        dic = dicts[i]\n",
    "        feat = catFeatures[i]\n",
    "        df[feat] = df[feat].replace(dic)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c5ac19ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** DATAFRAME INFO *****\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 9854 entries, 0 to 9853\n",
      "Data columns (total 10 columns):\n",
      " #   Column                                 Non-Null Count  Dtype \n",
      "---  ------                                 --------------  ----- \n",
      " 0   Id                                     9854 non-null   int64 \n",
      " 1   path                                   9854 non-null   object\n",
      " 2   speakerId                              9854 non-null   object\n",
      " 3   action                                 9854 non-null   object\n",
      " 4   object                                 9854 non-null   object\n",
      " 5   Self-reported fluency level            9854 non-null   object\n",
      " 6   First Language spoken                  9854 non-null   object\n",
      " 7   Current language used for work/school  9854 non-null   object\n",
      " 8   gender                                 9854 non-null   object\n",
      " 9   ageRange                               9854 non-null   object\n",
      "dtypes: int64(1), object(9)\n",
      "memory usage: 770.0+ KB\n",
      "None\n",
      "\n",
      "\n",
      "***** HEAD *****\n",
      "\n",
      "   Id                                               path         speakerId  \\\n",
      "0   0  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0a312...  2BqVo8kVB2Skwgyb   \n",
      "1   1  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0ee42...  2BqVo8kVB2Skwgyb   \n",
      "2   2  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/1d9f3...  2BqVo8kVB2Skwgyb   \n",
      "3   3  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/269fc...  2BqVo8kVB2Skwgyb   \n",
      "4   4  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/5bbda...  2BqVo8kVB2Skwgyb   \n",
      "\n",
      "            action  object Self-reported fluency level   \\\n",
      "0  change language    none                     advanced   \n",
      "1         activate   music                     advanced   \n",
      "2       deactivate  lights                     advanced   \n",
      "3         increase  volume                     advanced   \n",
      "4         increase  volume                     advanced   \n",
      "\n",
      "     First Language spoken Current language used for work/school  gender  \\\n",
      "0  English (United States)               English (United States)  female   \n",
      "1  English (United States)               English (United States)  female   \n",
      "2  English (United States)               English (United States)  female   \n",
      "3  English (United States)               English (United States)  female   \n",
      "4  English (United States)               English (United States)  female   \n",
      "\n",
      "  ageRange  \n",
      "0    22-40  \n",
      "1    22-40  \n",
      "2    22-40  \n",
      "3    22-40  \n",
      "4    22-40  \n",
      "\n",
      "\n",
      " ***** POSSIBLE CLASSIFICATION OUTCOMES *****\n",
      "\n",
      "\n",
      "\n",
      " ***** SAMPLE RATES *****\n",
      "SampleRates\n",
      "16000    9854\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Self-reported fluency level *****\n",
      "Self-reported fluency level \n",
      "advanced         346\n",
      "basic             56\n",
      "intermediate       8\n",
      "native          9444\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** First Language spoken *****\n",
      "First Language spoken\n",
      "English (Canada)              8\n",
      "English (United States)    9642\n",
      "French (Canada)             148\n",
      "Spanish (Venezuela)          35\n",
      "Telugu                       21\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Current language used for work/school *****\n",
      "Current language used for work/school\n",
      "English (Australia)          65\n",
      "English (Canada)            156\n",
      "English (United States)    9598\n",
      "Spanish (Venezuela)          35\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Gender *****\n",
      "gender\n",
      "female    4589\n",
      "male      5265\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Age range *****\n",
      "ageRange\n",
      "22-40    7496\n",
      "41-65    2115\n",
      "65+       243\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "df = load_dataframe(dev_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d6da9ca5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SampleRates\n",
      "16000    9854\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "target_rate = 16000\n",
    "edit_sample_rates(df, target_rate) #bring all sr's to 16kHz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "28325fb7",
   "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>Id</th>\n",
       "      <th>path</th>\n",
       "      <th>speakerId</th>\n",
       "      <th>action</th>\n",
       "      <th>object</th>\n",
       "      <th>Self-reported fluency level</th>\n",
       "      <th>First Language spoken</th>\n",
       "      <th>Current language used for work/school</th>\n",
       "      <th>gender</th>\n",
       "      <th>ageRange</th>\n",
       "      <th>SampleRates</th>\n",
       "      <th>AudioData</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0a312...</td>\n",
       "      <td>2BqVo8kVB2Skwgyb</td>\n",
       "      <td>change language</td>\n",
       "      <td>none</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0ee42...</td>\n",
       "      <td>2BqVo8kVB2Skwgyb</td>\n",
       "      <td>activate</td>\n",
       "      <td>music</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/1d9f3...</td>\n",
       "      <td>2BqVo8kVB2Skwgyb</td>\n",
       "      <td>deactivate</td>\n",
       "      <td>lights</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/269fc...</td>\n",
       "      <td>2BqVo8kVB2Skwgyb</td>\n",
       "      <td>increase</td>\n",
       "      <td>volume</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/5bbda...</td>\n",
       "      <td>2BqVo8kVB2Skwgyb</td>\n",
       "      <td>increase</td>\n",
       "      <td>volume</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9849</th>\n",
       "      <td>9849</td>\n",
       "      <td>dsl_data/audio/speakers/vnljypgejkINbBAY/4fb3d...</td>\n",
       "      <td>vnljypgejkINbBAY</td>\n",
       "      <td>decrease</td>\n",
       "      <td>volume</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[315, 321, 328, 335, 340, 342, 344, 351, 351, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9850</th>\n",
       "      <td>9850</td>\n",
       "      <td>dsl_data/audio/speakers/vnljypgejkINbBAY/59e6a...</td>\n",
       "      <td>vnljypgejkINbBAY</td>\n",
       "      <td>deactivate</td>\n",
       "      <td>lights</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[320, 322, 327, 332, 334, 334, 336, 338, 343, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9851</th>\n",
       "      <td>9851</td>\n",
       "      <td>dsl_data/audio/speakers/vnljypgejkINbBAY/5c81c...</td>\n",
       "      <td>vnljypgejkINbBAY</td>\n",
       "      <td>deactivate</td>\n",
       "      <td>lights</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[49, 52, 55, 59, 61, 65, 70, 73, 76, 81, 85, 8...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9852</th>\n",
       "      <td>9852</td>\n",
       "      <td>dsl_data/audio/speakers/vnljypgejkINbBAY/5ef42...</td>\n",
       "      <td>vnljypgejkINbBAY</td>\n",
       "      <td>deactivate</td>\n",
       "      <td>lights</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[333, 339, 342, 344, 346, 349, 353, 355, 357, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9853</th>\n",
       "      <td>9853</td>\n",
       "      <td>dsl_data/audio/speakers/vnljypgejkINbBAY/61b7a...</td>\n",
       "      <td>vnljypgejkINbBAY</td>\n",
       "      <td>increase</td>\n",
       "      <td>volume</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>[325, 331, 333, 335, 340, 342, 346, 351, 353, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9854 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Id                                               path  \\\n",
       "0        0  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0a312...   \n",
       "1        1  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/0ee42...   \n",
       "2        2  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/1d9f3...   \n",
       "3        3  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/269fc...   \n",
       "4        4  dsl_data/audio/speakers/2BqVo8kVB2Skwgyb/5bbda...   \n",
       "...    ...                                                ...   \n",
       "9849  9849  dsl_data/audio/speakers/vnljypgejkINbBAY/4fb3d...   \n",
       "9850  9850  dsl_data/audio/speakers/vnljypgejkINbBAY/59e6a...   \n",
       "9851  9851  dsl_data/audio/speakers/vnljypgejkINbBAY/5c81c...   \n",
       "9852  9852  dsl_data/audio/speakers/vnljypgejkINbBAY/5ef42...   \n",
       "9853  9853  dsl_data/audio/speakers/vnljypgejkINbBAY/61b7a...   \n",
       "\n",
       "             speakerId           action  object  Self-reported fluency level   \\\n",
       "0     2BqVo8kVB2Skwgyb  change language    none                             1   \n",
       "1     2BqVo8kVB2Skwgyb         activate   music                             1   \n",
       "2     2BqVo8kVB2Skwgyb       deactivate  lights                             1   \n",
       "3     2BqVo8kVB2Skwgyb         increase  volume                             1   \n",
       "4     2BqVo8kVB2Skwgyb         increase  volume                             1   \n",
       "...                ...              ...     ...                           ...   \n",
       "9849  vnljypgejkINbBAY         decrease  volume                             0   \n",
       "9850  vnljypgejkINbBAY       deactivate  lights                             0   \n",
       "9851  vnljypgejkINbBAY       deactivate  lights                             0   \n",
       "9852  vnljypgejkINbBAY       deactivate  lights                             0   \n",
       "9853  vnljypgejkINbBAY         increase  volume                             0   \n",
       "\n",
       "      First Language spoken  Current language used for work/school  gender  \\\n",
       "0                         0                                      0       0   \n",
       "1                         0                                      0       0   \n",
       "2                         0                                      0       0   \n",
       "3                         0                                      0       0   \n",
       "4                         0                                      0       0   \n",
       "...                     ...                                    ...     ...   \n",
       "9849                      0                                      0       1   \n",
       "9850                      0                                      0       1   \n",
       "9851                      0                                      0       1   \n",
       "9852                      0                                      0       1   \n",
       "9853                      0                                      0       1   \n",
       "\n",
       "      ageRange  SampleRates                                          AudioData  \n",
       "0            0        16000  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "1            0        16000  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "2            0        16000  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "3            0        16000  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "4            0        16000  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "...        ...          ...                                                ...  \n",
       "9849         0        16000  [315, 321, 328, 335, 340, 342, 344, 351, 351, ...  \n",
       "9850         0        16000  [320, 322, 327, 332, 334, 334, 336, 338, 343, ...  \n",
       "9851         0        16000  [49, 52, 55, 59, 61, 65, 70, 73, 76, 81, 85, 8...  \n",
       "9852         0        16000  [333, 339, 342, 344, 346, 349, 353, 355, 357, ...  \n",
       "9853         0        16000  [325, 331, 333, 335, 340, 342, 346, 351, 353, ...  \n",
       "\n",
       "[9854 rows x 12 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encode_categorical(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d320c1a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_all_audios(df):\n",
    "    fig, ax = plt.subplots(1,1)\n",
    "    for ind in range(0, len(df)):\n",
    "        audio = df['AudioData'].loc[ind]\n",
    "        sr = df['SampleRates'].loc[ind]\n",
    "        audioArray = np.array(audio)\n",
    "        duration = len(audio) / sr\n",
    "        time = np.linspace(0, duration, len(audio))\n",
    "        ax.plot(time, audioArray)\n",
    "    ax.set_ylabel(\"Amplitude\")\n",
    "    ax.set_xlabel(\"Time [s]\")\n",
    "    plt.show()\n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "095c6464",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_all_audios(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "da8b57ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "# By normalising the amplitude of the audios it is possible \n",
    "#to lower the impact of different recording conditions:\n",
    "\n",
    "def normalise_audios(df):\n",
    "    normalised = list()\n",
    "    for idx, rec in df.iterrows():\n",
    "        normalised.append(zscore(rec['AudioData']))\n",
    "    df['AudioData'] = normalised\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4d4d6221",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = normalise_audios(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7e81e9a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_all_audios(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "299ad240",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_duration_histogram(df):\n",
    "    recList = []\n",
    "    lenList = []\n",
    "\n",
    "    for idx, row in df.iterrows():\n",
    "        rec = row['AudioData']\n",
    "        sr = row['SampleRates']\n",
    "        recList.append(rec)\n",
    "        lenList.append(len(rec)/sr)\n",
    "\n",
    "    plt.xlabel('Seconds')\n",
    "    plt.ylabel('Frequency')\n",
    "    hist_array = plt.hist(np.array(lenList), bins='auto', color='#0504aa', alpha=0.7, rwidth=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8b498848",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGwCAYAAABPSaTdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA2O0lEQVR4nO3dfVxUdd7/8fcod94BisHAimjmHaWWWjhlXZuSqNSCcrXpWmJa7brYqnTj+lvL1L2yqMytJW33oWBXa23uVW6ZaYh3qXhHVmYumeuKLQy4GqC03Ajn90cXczUBKsPADMfX8/GYx8Nzzvd8z+fLmXnM23O+M2MxDMMQAACASbXzdAEAAAAtibADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMzcfTBXiD2tpaFRQUqEuXLrJYLJ4uBwAAXAbDMHTu3DlFRESoXbvGr98QdiQVFBQoMjLS02UAAAAXnDp1Sj169Gh0O2FHUpcuXSR998cKDAz0cDUAAOBylJWVKTIy0vE+3hjCjuS4dRUYGEjYAQCgjbnUFBQmKAMAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7JhMYuIuT5cAAIBXIewAAABTI+wAAABTI+wAAABTI+wAAABTI+wAAABTI+wAAABT82jY6dWrlywWS71HSkqKJKmiokIpKSkKCQlR586dlZSUpKKiIqc+8vPzFR8fr44dOyo0NFSPPfaYLly44InhAAAAL+TRsHPgwAEVFhY6HllZWZKku+++W5I0d+5cvffee1q3bp127NihgoICTZw40bF/TU2N4uPjVVVVpT179mjNmjXKzMzUk08+6ZHxAAAA72MxDMPwdBF15syZow0bNujYsWMqKyvTVVddpbVr1+o///M/JUl/+9vfNHDgQOXk5GjEiBH64IMPdOedd6qgoEBhYWGSpJUrV2revHk6ffq0/Pz8GjxOZWWlKisrHctlZWWKjIxUaWmpAgMDW36gLSgxcZfWrx/p6TIAAGhxZWVlCgoKuuT7t9fM2amqqtLrr7+u6dOny2KxKDc3V9XV1YqNjXW0GTBggHr27KmcnBxJUk5OjgYNGuQIOpIUFxensrIyHTlypNFjLV26VEFBQY5HZGRkyw0MAAB4lNeEnfXr16ukpETTpk2TJNntdvn5+Sk4ONipXVhYmOx2u6PN94NO3fa6bY2ZP3++SktLHY9Tp065byAAAMCr+Hi6gDqrVq3SuHHjFBER0eLH8vf3l7+/f4sfBwAAeJ5XXNk5efKktmzZogceeMCxzmq1qqqqSiUlJU5ti4qKZLVaHW1++OmsuuW6NgAA4MrmFWEnIyNDoaGhio+Pd6wbNmyYfH19lZ2d7ViXl5en/Px82Ww2SZLNZtPhw4dVXFzsaJOVlaXAwEBFR0e33gAAAIDX8vhtrNraWmVkZCg5OVk+Pv9XTlBQkGbMmKHU1FR169ZNgYGBevjhh2Wz2TRixAhJ0pgxYxQdHa377rtPaWlpstvtWrBggVJSUrhNBQAAJHlB2NmyZYvy8/M1ffr0ettefPFFtWvXTklJSaqsrFRcXJxeeeUVx/b27dtrw4YNmjlzpmw2mzp16qTk5GQtXry4NYfgMYmJuySJj5oDAHARHg87Y8aMUWNf9RMQEKD09HSlp6c3un9UVJQ2btzYUuUBAIA2zivm7AAAALQUwg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1ws4VoO4HQwEAuBIRdgAAgKkRdgAAgKkRdgAAgKkRdgAAgKkRdgAAgKkRdkwoMXEXn8ACAOB/EXbaKMIMAACXh7ADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMjbADAABMzeNh55///KfuvfdehYSEqEOHDho0aJAOHjzo2G4Yhp588kmFh4erQ4cOio2N1bFjx5z6OHv2rKZMmaLAwEAFBwdrxowZOn/+fGsPBQAAeCGPhp1vvvlGt9xyi3x9ffXBBx/oiy++0AsvvKCuXbs62qSlpemll17SypUrtW/fPnXq1ElxcXGqqKhwtJkyZYqOHDmirKwsbdiwQTt37tRDDz3kiSEBAAAv4+PJgz/77LOKjIxURkaGY13v3r0d/zYMQ8uXL9eCBQuUkJAgSXrttdcUFham9evXa9KkSTp69Kg2bdqkAwcOaPjw4ZKkl19+WePHj9fzzz+viIiIesetrKxUZWWlY7msrKylhggAADzMo1d23n33XQ0fPlx33323QkNDdcMNN+iPf/yjY/uJEydkt9sVGxvrWBcUFKSYmBjl5ORIknJychQcHOwIOpIUGxurdu3aad++fQ0ed+nSpQoKCnI8IiMjW2iEAADA0zwadv7+979rxYoV6tu3rzZv3qyZM2fqV7/6ldasWSNJstvtkqSwsDCn/cLCwhzb7Ha7QkNDnbb7+PioW7dujjY/NH/+fJWWljoep06dcvfQAACAl/Dobaza2loNHz5cTz/9tCTphhtu0Oeff66VK1cqOTm5xY7r7+8vf3//FusfAAB4D49e2QkPD1d0dLTTuoEDByo/P1+SZLVaJUlFRUVObYqKihzbrFariouLnbZfuHBBZ8+edbQBAABXLo+GnVtuuUV5eXlO67788ktFRUVJ+m6ystVqVXZ2tmN7WVmZ9u3bJ5vNJkmy2WwqKSlRbm6uo83WrVtVW1urmJiYVhgFAADwZh69jTV37lzdfPPNevrpp/XTn/5U+/fv1x/+8Af94Q9/kCRZLBbNmTNHv/3tb9W3b1/17t1bTzzxhCIiIpSYmCjpuytBY8eO1YMPPqiVK1equrpas2bN0qRJkxr8JBYAALiyeDTs3HjjjXrnnXc0f/58LV68WL1799by5cs1ZcoUR5vHH39c5eXleuihh1RSUqKRI0dq06ZNCggIcLT505/+pFmzZmn06NFq166dkpKS9NJLL3liSAAAwMt4NOxI0p133qk777yz0e0Wi0WLFy/W4sWLG23TrVs3rV27tiXKAwAAbZzHfy4CAACgJRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2rhCJibuUmLjL02UAANDqCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDsAAMDUCDtXGH4MFABwpSHsAAAAUyPsAAAAU/No2HnqqadksVicHgMGDHBsr6ioUEpKikJCQtS5c2clJSWpqKjIqY/8/HzFx8erY8eOCg0N1WOPPaYLFy609lAAAICX8vF0Addee622bNniWPbx+b+S5s6dq/fff1/r1q1TUFCQZs2apYkTJ2r37t2SpJqaGsXHx8tqtWrPnj0qLCzU1KlT5evrq6effrrVxwIAALyPx8OOj4+PrFZrvfWlpaVatWqV1q5dq1GjRkmSMjIyNHDgQO3du1cjRozQhx9+qC+++EJbtmxRWFiYrr/+ei1ZskTz5s3TU089JT8/v9YeDgAA8DIen7Nz7NgxRURE6Oqrr9aUKVOUn58vScrNzVV1dbViY2MdbQcMGKCePXsqJydHkpSTk6NBgwYpLCzM0SYuLk5lZWU6cuRIo8esrKxUWVmZ0wMAAJiTR6/sxMTEKDMzU/3791dhYaEWLVqkW2+9VZ9//rnsdrv8/PwUHBzstE9YWJjsdrskyW63OwWduu112xqzdOlSLVq0yL2D8TA+Ug4AQMM8GnbGjRvn+PfgwYMVExOjqKgovfXWW+rQoUOLHXf+/PlKTU11LJeVlSkyMrLFjgcAADzH47exvi84OFj9+vXTV199JavVqqqqKpWUlDi1KSoqcszxsVqt9T6dVbfc0DygOv7+/goMDHR6AAAAc/KqsHP+/HkdP35c4eHhGjZsmHx9fZWdne3YnpeXp/z8fNlsNkmSzWbT4cOHVVxc7GiTlZWlwMBARUdHt3r9AADA+3j0Ntajjz6qu+66S1FRUSooKNDChQvVvn17TZ48WUFBQZoxY4ZSU1PVrVs3BQYG6uGHH5bNZtOIESMkSWPGjFF0dLTuu+8+paWlyW63a8GCBUpJSZG/v78nhwYAALyER8PO119/rcmTJ+vMmTO66qqrNHLkSO3du1dXXXWVJOnFF19Uu3btlJSUpMrKSsXFxemVV15x7N++fXtt2LBBM2fOlM1mU6dOnZScnKzFixd7akgAAMDLeDTsvPnmmxfdHhAQoPT0dKWnpzfaJioqShs3bnR3aQAAwCS8as4OAACAuxF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2AACAqRF2rmCJibs8XQIAAC2OsAMAAEzNpbDz97//3d11AAAAtAiXws4111yj22+/Xa+//roqKircXRNaUWLiLm5nAQBMzaWw8/HHH2vw4MFKTU2V1WrVz3/+c+3fv9/dtQEAADSbS2Hn+uuv1+9+9zsVFBRo9erVKiws1MiRI3Xddddp2bJlOn36tLvrBAAAcEmzJij7+Pho4sSJWrdunZ599ll99dVXevTRRxUZGampU6eqsLDQXXUCAAC4pFlh5+DBg/rlL3+p8PBwLVu2TI8++qiOHz+urKwsFRQUKCEhwV11AgAAuMTHlZ2WLVumjIwM5eXlafz48Xrttdc0fvx4tWv3XXbq3bu3MjMz1atXL3fWCgAA0GQuhZ0VK1Zo+vTpmjZtmsLDwxtsExoaqlWrVjWrOAAAgOZyKewcO3bskm38/PyUnJzsSvcAAABu49KcnYyMDK1bt67e+nXr1mnNmjXNLgoAAMBdXAo7S5cuVffu3eutDw0N1dNPP93sogAAANzFpbCTn5+v3r1711sfFRWl/Pz8ZhcFAADgLi6FndDQUH322Wf11n/66acKCQlpdlEAAADu4lLYmTx5sn71q19p27ZtqqmpUU1NjbZu3arZs2dr0qRJ7q4RAADAZS59GmvJkiX6xz/+odGjR8vH57suamtrNXXqVObsAAAAr+JS2PHz89Of//xnLVmyRJ9++qk6dOigQYMGKSoqyt31AQAANItLYadOv3791K9fP3fVAgAA4HYuhZ2amhplZmYqOztbxcXFqq2tddq+detWtxQHAADQXC5NUJ49e7Zmz56tmpoaXXfddRoyZIjTwxXPPPOMLBaL5syZ41hXUVGhlJQUhYSEqHPnzkpKSlJRUZHTfvn5+YqPj1fHjh0VGhqqxx57TBcuXHCpBm+XmLjL0yUAANDmuHRl580339Rbb72l8ePHu6WIAwcO6NVXX9XgwYOd1s+dO1fvv/++1q1bp6CgIM2aNUsTJ07U7t27JX13hSk+Pl5Wq1V79uxRYWGhpk6dKl9fXyZKAwAASS5e2fHz89M111zjlgLOnz+vKVOm6I9//KO6du3qWF9aWqpVq1Zp2bJlGjVqlIYNG6aMjAzt2bNHe/fulSR9+OGH+uKLL/T666/r+uuv17hx47RkyRKlp6erqqrKLfUBAIC2zaWw88gjj+h3v/udDMNodgEpKSmKj49XbGys0/rc3FxVV1c7rR8wYIB69uypnJwcSVJOTo4GDRqksLAwR5u4uDiVlZXpyJEjjR6zsrJSZWVlTg8AAGBOLt3G2rVrl7Zt26YPPvhA1157rXx9fZ22v/3225fVz5tvvqmPP/5YBw4cqLfNbrfLz89PwcHBTuvDwsJkt9sdbb4fdOq2121rzNKlS7Vo0aLLqhEAALRtLoWd4OBgTZgwoVkHPnXqlGbPnq2srCwFBAQ0q6+mmj9/vlJTUx3LZWVlioyMbNUaAABA63Ap7GRkZDT7wLm5uSouLtbQoUMd62pqarRz5079/ve/1+bNm1VVVaWSkhKnqztFRUWyWq2SJKvVqv379zv1W/dprbo2DfH395e/v3+zxwAAALyfS3N2JOnChQvasmWLXn31VZ07d06SVFBQoPPnz1/W/qNHj9bhw4f1ySefOB7Dhw/XlClTHP/29fVVdna2Y5+8vDzl5+fLZrNJkmw2mw4fPqzi4mJHm6ysLAUGBio6OtrVoQEAABNx6crOyZMnNXbsWOXn56uyslJ33HGHunTpomeffVaVlZVauXLlJfvo0qWLrrvuOqd1nTp1UkhIiGP9jBkzlJqaqm7duikwMFAPP/ywbDabRowYIUkaM2aMoqOjdd999yktLU12u10LFixQSkoKV24AAICkZnyp4PDhw/XNN9+oQ4cOjvUTJkxwuhLTXC+++KLuvPNOJSUl6bbbbpPVanWa/Ny+fXtt2LBB7du3l81m07333qupU6dq8eLFbqsBAAC0bS5d2fnoo4+0Z88e+fn5Oa3v1auX/vnPf7pczPbt252WAwIClJ6ervT09Eb3iYqK0saNG10+JgAAMDeXruzU1taqpqam3vqvv/5aXbp0aXZRAAAA7uJS2BkzZoyWL1/uWLZYLDp//rwWLlzotp+QAAAAcAeXbmO98MILiouLU3R0tCoqKvSzn/1Mx44dU/fu3fXGG2+4u0YAAACXuRR2evTooU8//VRvvvmmPvvsM50/f14zZszQlClTnCYsAwAAeJpLYUeSfHx8dO+997qzFgAAALdzKey89tprF90+depUl4oBAABwN5fCzuzZs52Wq6ur9e2338rPz08dO3Yk7AAAAK/h0qexvvnmG6fH+fPnlZeXp5EjRzJBGQAAeBWXfxvrh/r27atnnnmm3lUfAAAAT3Jb2JG+m7RcUFDgzi4BAACaxaU5O++++67TsmEYKiws1O9//3vdcsstbikMAADAHVwKO4mJiU7LFotFV111lUaNGqUXXnjBHXUBAAC4hUthp7a21t11AAAAtAi3ztkBAADwNi5d2UlNTb3stsuWLXPlEAAAAG7hUtg5dOiQDh06pOrqavXv31+S9OWXX6p9+/YaOnSoo53FYnFPlQAAAC5yKezcdddd6tKli9asWaOuXbtK+u6LBu+//37deuuteuSRR9xaJAAAgKtcmrPzwgsvaOnSpY6gI0ldu3bVb3/7Wz6NBQAAvIpLYaesrEynT5+ut/706dM6d+5cs4sCAABwF5fCzoQJE3T//ffr7bff1tdff62vv/5a//M//6MZM2Zo4sSJ7q4RAADAZS7N2Vm5cqUeffRR/exnP1N1dfV3Hfn4aMaMGXruuefcWiAAAEBzuBR2OnbsqFdeeUXPPfecjh8/Lknq06ePOnXq5NbiAAAAmqtZXypYWFiowsJC9e3bV506dZJhGO6qCwAAwC1cCjtnzpzR6NGj1a9fP40fP16FhYWSpBkzZvCxcwAA4FVcCjtz586Vr6+v8vPz1bFjR8f6e+65R5s2bXJbcQAAAM3l0pydDz/8UJs3b1aPHj2c1vft21cnT550S2FofYmJuyRJ69eP9HAlAAC4j0tXdsrLy52u6NQ5e/as/P39m10UAACAu7gUdm699Va99tprjmWLxaLa2lqlpaXp9ttvd1txAAAAzeXSbay0tDSNHj1aBw8eVFVVlR5//HEdOXJEZ8+e1e7du91dIwAAgMtcurJz3XXX6csvv9TIkSOVkJCg8vJyTZw4UYcOHVKfPn3cXSMAAIDLmnxlp7q6WmPHjtXKlSv1m9/8piVqAgAAcJsmX9nx9fXVZ5991hK1AAAAuJ1Lt7HuvfderVq1yt21AAAAuJ1LE5QvXLig1atXa8uWLRo2bFi938RatmyZW4oDAABoriaFnb///e/q1auXPv/8cw0dOlSS9OWXXzq1sVgs7qsOAACgmZoUdvr27avCwkJt27ZN0nc/D/HSSy8pLCysRYoDAABoribN2fnhr5p/8MEHKi8vd2tBAAAA7uTSBOU6Pww/AAAA3qZJYcdisdSbk9OcOTorVqzQ4MGDFRgYqMDAQNlsNn3wwQeO7RUVFUpJSVFISIg6d+6spKQkFRUVOfWRn5+v+Ph4dezYUaGhoXrsscd04cIFl2sCAADm0qQ5O4ZhaNq0aY4f+6yoqNAvfvGLep/Gevvtty+rvx49euiZZ55R3759ZRiG1qxZo4SEBB06dEjXXnut5s6dq/fff1/r1q1TUFCQZs2apYkTJzp+kqKmpkbx8fGyWq3as2ePCgsLNXXqVPn6+urpp59uytAAAIBJNSnsJCcnOy3fe++9zTr4XXfd5bT8X//1X1qxYoX27t2rHj16aNWqVVq7dq1GjRolScrIyNDAgQO1d+9ejRgxQh9++KG++OILbdmyRWFhYbr++uu1ZMkSzZs3T0899ZT8/PwaPG5lZaUqKysdy2VlZc0aBwAA8F5NCjsZGRktVYdqamq0bt06lZeXy2azKTc3V9XV1YqNjXW0GTBggHr27KmcnByNGDFCOTk5GjRokNOnweLi4jRz5kwdOXJEN9xwQ4PHWrp0qRYtWtRiYwEAAN9JTNyl9etHerSGZk1QdofDhw+rc+fO8vf31y9+8Qu98847io6Olt1ul5+fn4KDg53ah4WFyW63S5Lsdnu9j73XLde1acj8+fNVWlrqeJw6dcq9gwIAAF7DpW9Qdqf+/fvrk08+UWlpqf7yl78oOTlZO3bsaNFj+vv7O+YdtTWJibs8XQIAAG2Kx8OOn5+frrnmGknSsGHDdODAAf3ud7/TPffco6qqKpWUlDhd3SkqKpLVapUkWa1W7d+/36m/uk9r1bUBAABXNo/fxvqh2tpaVVZWatiwYfL19VV2drZjW15envLz82Wz2SRJNptNhw8fVnFxsaNNVlaWAgMDFR0d3eq1AwAA7+PRKzvz58/XuHHj1LNnT507d05r167V9u3btXnzZgUFBWnGjBlKTU1Vt27dFBgYqIcfflg2m00jRoyQJI0ZM0bR0dG67777lJaWJrvdrgULFiglJaXN3qYCAADu5dGwU1xcrKlTp6qwsFBBQUEaPHiwNm/erDvuuEOS9OKLL6pdu3ZKSkpSZWWl4uLi9Morrzj2b9++vTZs2KCZM2fKZrOpU6dOSk5O1uLFiz01JAAA4GU8GnZWrVp10e0BAQFKT09Xenp6o22ioqK0ceNGd5cGAABMwuvm7AAAALgTYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYQcAAJgaYccEEhN3eboEAAC8FmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHAACYGmEHkvh9LQCAeRF2AACAqRF2AACAqRF2AACAqRF20KDExF3M4wEAmAJhBwAAmBphBwAAmBphBwAAmBphBwAAmBphBwAAmJpHw87SpUt14403qkuXLgoNDVViYqLy8vKc2lRUVCglJUUhISHq3LmzkpKSVFRU5NQmPz9f8fHx6tixo0JDQ/XYY4/pwoULrTkUAADgpTwadnbs2KGUlBTt3btXWVlZqq6u1pgxY1ReXu5oM3fuXL333ntat26dduzYoYKCAk2cONGxvaamRvHx8aqqqtKePXu0Zs0aZWZm6sknn/TEkAAAgJfx8eTBN23a5LScmZmp0NBQ5ebm6rbbblNpaalWrVqltWvXatSoUZKkjIwMDRw4UHv37tWIESP04Ycf6osvvtCWLVsUFham66+/XkuWLNG8efP01FNPyc/Pr95xKysrVVlZ6VguKytr2YECAACP8ao5O6WlpZKkbt26SZJyc3NVXV2t2NhYR5sBAwaoZ8+eysnJkSTl5ORo0KBBCgsLc7SJi4tTWVmZjhw50uBxli5dqqCgIMcjMjKypYYEAAA8zGvCTm1trebMmaNbbrlF1113nSTJbrfLz89PwcHBTm3DwsJkt9sdbb4fdOq2121ryPz581VaWup4nDp1ys2jAQAA3sKjt7G+LyUlRZ9//rl27Wr5nyjw9/eXv79/ix8HAAB4nldc2Zk1a5Y2bNigbdu2qUePHo71VqtVVVVVKikpcWpfVFQkq9XqaPPDT2fVLde1AQAAVy6Phh3DMDRr1iy988472rp1q3r37u20fdiwYfL19VV2drZjXV5envLz82Wz2SRJNptNhw8fVnFxsaNNVlaWAgMDFR0d3ToDAQAAXsujt7FSUlK0du1a/fWvf1WXLl0cc2yCgoLUoUMHBQUFacaMGUpNTVW3bt0UGBiohx9+WDabTSNGjJAkjRkzRtHR0brvvvuUlpYmu92uBQsWKCUlhVtVAADAs2FnxYoVkqQf//jHTuszMjI0bdo0SdKLL76odu3aKSkpSZWVlYqLi9Mrr7ziaNu+fXtt2LBBM2fOlM1mU6dOnZScnKzFixe31jAAAIAX82jYMQzjkm0CAgKUnp6u9PT0RttERUVp48aN7iwNAACYhFdMUAYAAGgphB0AAGBqhB1cVGJiy3/vEQAALYmwAwAATI2w4+W4sgIAQPMQdgAAgKkRdgAAgKl5zQ+Bwpk33b6qq2X9+pEergQAgKbjyg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wg5ckpi4S4mJuzxdBgAAl0TYAQAApkbYAQAApkbYQZNx+woA0JYQdgAAgKkRdgAAgKl5NOzs3LlTd911lyIiImSxWLR+/Xqn7YZh6Mknn1R4eLg6dOig2NhYHTt2zKnN2bNnNWXKFAUGBio4OFgzZszQ+fPnW3EUAADAm3k07JSXl2vIkCFKT09vcHtaWppeeuklrVy5Uvv27VOnTp0UFxeniooKR5spU6boyJEjysrK0oYNG7Rz50499NBDrTUEAADg5Xw8efBx48Zp3LhxDW4zDEPLly/XggULlJCQIEl67bXXFBYWpvXr12vSpEk6evSoNm3apAMHDmj48OGSpJdfflnjx4/X888/r4iIiFYbCwAA8E5eO2fnxIkTstvtio2NdawLCgpSTEyMcnJyJEk5OTkKDg52BB1Jio2NVbt27bRv375G+66srFRZWZnTw5vxBX4AALjOa8OO3W6XJIWFhTmtDwsLc2yz2+0KDQ112u7j46Nu3bo52jRk6dKlCgoKcjwiIyPdXD0AAPAWXht2WtL8+fNVWlrqeJw6dcrTJQEAgBbitWHHarVKkoqKipzWFxUVObZZrVYVFxc7bb9w4YLOnj3raNMQf39/BQYGOj3gOm6zAQC8mdeGnd69e8tqtSo7O9uxrqysTPv27ZPNZpMk2Ww2lZSUKDc319Fm69atqq2tVUxMTKvXDAAAvI9HP411/vx5ffXVV47lEydO6JNPPlG3bt3Us2dPzZkzR7/97W/Vt29f9e7dW0888YQiIiKUmJgoSRo4cKDGjh2rBx98UCtXrlR1dbVmzZqlSZMm8UksAAAgycNh5+DBg7r99tsdy6mpqZKk5ORkZWZm6vHHH1d5ebkeeughlZSUaOTIkdq0aZMCAgIc+/zpT3/SrFmzNHr0aLVr105JSUl66aWXWn0s7sQtIQAA3MejYefHP/6xDMNodLvFYtHixYu1ePHiRtt069ZNa9eubYnycBkIZgAAb+e1c3YAAADcgbADAABMjbADAABMjbADAABMjbCDFsHEZQCAtyDsAAAAUyPsoB6uygAAzISwAwAATI2wAwAATM2j36AM8+EWGADA23BlBwAAmBphBwAAmBphBwAAmBphBwAAmBphBwAAmBphxwskJu7iU0wAALQQwg4AADA1wg4AADA1wg4AADA1wg4AADA1wg4AADA1wo6H8SksAABaFmEHAACYGmEHAACYGmEHrYIvTgQAeIqPpwtA20FYAQC0RVzZQasiMAEAWhthB63uYre0CEMAAHcj7KDFME8HAOANCDsAAMDUmKDsIVzxAACgdXBlBwAAmBphBwAAmBphBx7HLT0AQEsi7AAAAFNjgjK8wqW+d2f9+pGtWQ4AwEQIO16E2zkAALgft7E8gFBzcU35MkL+lgCASzFN2ElPT1evXr0UEBCgmJgY7d+/39Ml4X+5I5AQagAArjLFbaw///nPSk1N1cqVKxUTE6Ply5crLi5OeXl5Cg0N9XR5aERTA8z35+98f98f9sP8HgDA95niys6yZcv04IMP6v7771d0dLRWrlypjh07avXq1Z4uzXFLhisTnsPfHwCubG3+yk5VVZVyc3M1f/58x7p27dopNjZWOTk5De5TWVmpyspKx3JpaakkqayszO31VVeXO/5d1//317Wk+PjNTVrfXGVlZW4dW2P9Xeo4roz7jTdskqTJk+s/Z+q2NaSu/cXaXGy/Hx77+/1cbt+u7ne5/TXX9/tzd99oOs4BWlt1dXmLvL9K//e+ahjGRdu1+bDzr3/9SzU1NQoLC3NaHxYWpr/97W8N7rN06VItWrSo3vrIyMgWqbFOUFCLdu9x7h5fY/21xN/xYn1ezvGaU9P3922oH1eP766a3OFSY0Tr4hygtbX0c+7cuXMKushB2nzYccX8+fOVmprqWK6trdXZs2cVEhIii8Xi1LasrEyRkZE6deqUAgMDW7vUVnOljFO6csbKOM3lShmndOWMlXE2n2EYOnfunCIiIi7ars2Hne7du6t9+/YqKipyWl9UVCSr1drgPv7+/vL393daFxwcfNHjBAYGmvrJWOdKGad05YyVcZrLlTJO6coZK+Nsnotd0anT5ico+/n5adiwYcrOznasq62tVXZ2tmw27ksDAHCla/NXdiQpNTVVycnJGj58uG666SYtX75c5eXluv/++z1dGgAA8DBThJ177rlHp0+f1pNPPim73a7rr79emzZtqjdp2RX+/v5auHBhvdteZnOljFO6csbKOM3lShmndOWMlXG2Hotxqc9rAQAAtGFtfs4OAADAxRB2AACAqRF2AACAqRF2AACAqRF2JKWnp6tXr14KCAhQTEyM9u/ff9H269at04ABAxQQEKBBgwZp48aNrVSpa5YuXaobb7xRXbp0UWhoqBITE5WXl3fRfTIzM2WxWJweAQEBrVSx65566ql6dQ8YMOCi+7S18ylJvXr1qjdOi8WilJSUBtu3lfO5c+dO3XXXXYqIiJDFYtH69eudthuGoSeffFLh4eHq0KGDYmNjdezYsUv229TXeGu42Firq6s1b948DRo0SJ06dVJERISmTp2qgoKCi/bpyvO/pV3qnE6bNq1ezWPHjr1kv952Ti81zoZerxaLRc8991yjfXrj+byc95OKigqlpKQoJCREnTt3VlJSUr0v/v0hV1/bl+uKDzt//vOflZqaqoULF+rjjz/WkCFDFBcXp+Li4gbb79mzR5MnT9aMGTN06NAhJSYmKjExUZ9//nkrV375duzYoZSUFO3du1dZWVmqrq7WmDFjVF5+8R/tDAwMVGFhoeNx8uTJVqq4ea699lqnunftavwXz9vi+ZSkAwcOOI0xKytLknT33Xc3uk9bOJ/l5eUaMmSI0tPTG9yelpaml156SStXrtS+ffvUqVMnxcXFqaKiotE+m/oaby0XG+u3336rjz/+WE888YQ+/vhjvf3228rLy9NPfvKTS/bblOd/a7jUOZWksWPHOtX8xhtvXLRPbzynlxrn98dXWFio1atXy2KxKCkp6aL9etv5vJz3k7lz5+q9997TunXrtGPHDhUUFGjixIkX7deV13aTGFe4m266yUhJSXEs19TUGBEREcbSpUsbbP/Tn/7UiI+Pd1oXExNj/PznP2/ROt2puLjYkGTs2LGj0TYZGRlGUFBQ6xXlJgsXLjSGDBly2e3NcD4NwzBmz55t9OnTx6itrW1we1s8n5KMd955x7FcW1trWK1W47nnnnOsKykpMfz9/Y033nij0X6a+hr3hB+OtSH79+83JBknT55stE1Tn/+traFxJicnGwkJCU3qx9vP6eWcz4SEBGPUqFEXbePt59Mw6r+flJSUGL6+vsa6descbY4ePWpIMnJychrsw9XXdlNc0Vd2qqqqlJubq9jYWMe6du3aKTY2Vjk5OQ3uk5OT49RekuLi4hpt741KS0slSd26dbtou/PnzysqKkqRkZFKSEjQkSNHWqO8Zjt27JgiIiJ09dVXa8qUKcrPz2+0rRnOZ1VVlV5//XVNnz693g/Zfl9bPZ91Tpw4Ibvd7nS+goKCFBMT0+j5cuU17q1KS0tlsVgu+Tt+TXn+e4vt27crNDRU/fv318yZM3XmzJlG25rhnBYVFen999/XjBkzLtnW28/nD99PcnNzVV1d7XR+BgwYoJ49ezZ6flx5bTfVFR12/vWvf6mmpqbeNy2HhYXJbrc3uI/dbm9Se29TW1urOXPm6JZbbtF1113XaLv+/ftr9erV+utf/6rXX39dtbW1uvnmm/X111+3YrVNFxMTo8zMTG3atEkrVqzQiRMndOutt+rcuXMNtm/r51OS1q9fr5KSEk2bNq3RNm31fH5f3Tlpyvly5TXujSoqKjRv3jxNnjz5oj+k2NTnvzcYO3asXnvtNWVnZ+vZZ5/Vjh07NG7cONXU1DTY3gzndM2aNerSpcslb+14+/ls6P3EbrfLz8+vXii/1PtqXZvL3aepTPFzEbh8KSkp+vzzzy9539dmszn9kOrNN9+sgQMH6tVXX9WSJUtaukyXjRs3zvHvwYMHKyYmRlFRUXrrrbcu639RbdGqVas0btw4RURENNqmrZ5PfDdZ+ac//akMw9CKFSsu2rYtPv8nTZrk+PegQYM0ePBg9enTR9u3b9fo0aM9WFnLWb16taZMmXLJDwl4+/m83PcTb3BFX9np3r272rdvX2+WeFFRkaxWa4P7WK3WJrX3JrNmzdKGDRu0bds29ejRo0n7+vr66oYbbtBXX33VQtW1jODgYPXr16/Rutvy+ZSkkydPasuWLXrggQeatF9bPJ9156Qp58uV17g3qQs6J0+eVFZW1kWv6jTkUs9/b3T11Vere/fujdbc1s/pRx99pLy8vCa/ZiXvOp+NvZ9YrVZVVVWppKTEqf2l3lfr2lzuPk11RYcdPz8/DRs2TNnZ2Y51tbW1ys7Odvpf8PfZbDan9pKUlZXVaHtvYBiGZs2apXfeeUdbt25V7969m9xHTU2NDh8+rPDw8BaosOWcP39ex48fb7Tutng+vy8jI0OhoaGKj49v0n5t8Xz27t1bVqvV6XyVlZVp3759jZ4vV17j3qIu6Bw7dkxbtmxRSEhIk/u41PPfG3399dc6c+ZMozW35XMqfXcldtiwYRoyZEiT9/WG83mp95Nhw4bJ19fX6fzk5eUpPz+/0fPjymvblcKvaG+++abh7+9vZGZmGl988YXx0EMPGcHBwYbdbjcMwzDuu+8+49e//rWj/e7duw0fHx/j+eefN44ePWosXLjQ8PX1NQ4fPuypIVzSzJkzjaCgIGP79u1GYWGh4/Htt9862vxwnIsWLTI2b95sHD9+3MjNzTUmTZpkBAQEGEeOHPHEEC7bI488Ymzfvt04ceKEsXv3biM2Ntbo3r27UVxcbBiGOc5nnZqaGqNnz57GvHnz6m1rq+fz3LlzxqFDh4xDhw4Zkoxly5YZhw4dcnwC6ZlnnjGCg4ONv/71r8Znn31mJCQkGL179zb+/e9/O/oYNWqU8fLLLzuWL/Ua95SLjbWqqsr4yU9+YvTo0cP45JNPnF63lZWVjj5+ONZLPf894WLjPHfunPHoo48aOTk5xokTJ4wtW7YYQ4cONfr27WtUVFQ4+mgL5/RSz13DMIzS0lKjY8eOxooVKxrsoy2cz8t5P/nFL35h9OzZ09i6datx8OBBw2azGTabzamf/v37G2+//bZj+XJe281xxYcdwzCMl19+2ejZs6fh5+dn3HTTTcbevXsd2/7jP/7DSE5Odmr/1ltvGf369TP8/PyMa6+91nj//fdbueKmkdTgIyMjw9Hmh+OcM2eO428SFhZmjB8/3vj4449bv/gmuueee4zw8HDDz8/P+NGPfmTcc889xldffeXYbobzWWfz5s2GJCMvL6/etrZ6Prdt29bgc7VuLLW1tcYTTzxhhIWFGf7+/sbo0aPrjT8qKspYuHCh07qLvcY95WJjPXHiRKOv223btjn6+OFYL/X894SLjfPbb781xowZY1x11VWGr6+vERUVZTz44IP1QktbOKeXeu4ahmG8+uqrRocOHYySkpIG+2gL5/Ny3k/+/e9/G7/85S+Nrl27Gh07djQmTJhgFBYW1uvn+/tczmu7OSz/e1AAAABTuqLn7AAAAPMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7AAAAFMj7ADAZZo2bZoSExM9XQaAJiLsAGh1p0+f1syZM9WzZ0/5+/vLarUqLi5Ou3fv9nRpAEzIx9MFALjyJCUlqaqqSmvWrNHVV1+toqIiZWdn68yZM54uDYAJcWUHQKsqKSnRRx99pGeffVa33367oqKidNNNN2n+/Pn6yU9+4mjzwAMP6KqrrlJgYKBGjRqlTz/91Kmf9957TzfeeKMCAgLUvXt3TZgwwbHtm2++0dSpU9W1a1d17NhR48aN07FjxxzbMzMzFRwcrM2bN2vgwIHq3Lmzxo4dq8LCQkebmpoapaamKjg4WCEhIXr88cf1w58S/Mtf/qJBgwapQ4cOCgkJUWxsrMrLy1vizwagGQg7AFpV586d1blzZ61fv16VlZUNtrn77rtVXFysDz74QLm5uRo6dKhGjx6ts2fPSpLef/99TZgwQePHj9ehQ4eUnZ2tm266ybH/tGnTdPDgQb377rvKycmRYRgaP368qqurHW2+/fZbPf/88/rv//5v7dy5U/n5+Xr00Ucd21944QVlZmZq9erV2rVrl86ePat33nnHsb2wsFCTJ0/W9OnTdfToUW3fvl0TJ06sF4gAeAG3/X46AFymv/zlL0bXrl2NgIAA4+abbzbmz59vfPrpp4ZhGMZHH31kBAYGGhUVFU779OnTx3j11VcNwzAMm81mTJkypcG+v/zyS0OSsXv3bse6f/3rX0aHDh2Mt956yzAMw8jIyDAkGV999ZWjTXp6uhEWFuZYDg8PN9LS0hzL1dXVRo8ePYyEhATDMAwjNzfXkGT84x//aMZfAkBr4MoOgFaXlJSkgoICvfvuuxo7dqy2b9+uoUOHKjMzU59++qnOnz+vkJAQx1Wgzp0768SJEzp+/Lgk6ZNPPtHo0aMb7Pvo0aPy8fFRTEyMY11ISIj69++vo0ePOtZ17NhRffr0cSyHh4eruLhYklRaWqrCwkKnPnx8fDR8+HDH8pAhQzR69GgNGjRId999t/74xz/qm2++cc8fCIBbMUEZgEcEBATojjvu0B133KEnnnhCDzzwgBYuXKhf/vKXCg8P1/bt2+vtExwcLEnq0KFDs4/v6+vrtGyxWJp0C6p9+/bKysrSnj179OGHH+rll1/Wb37zG+3bt0+9e/dudn0A3IcrOwC8QnR0tMrLyzV06FDZ7Xb5+PjommuucXp0795dkjR48GBlZ2c32M/AgQN14cIF7du3z7HuzJkzysvLU3R09GXVEhQUpPDwcKc+Lly4oNzcXKd2FotFt9xyixYtWqRDhw7Jz8/PaV4PAO/AlR0ArerMmTO6++67NX36dA0ePFhdunTRwYMHlZaWpoSEBMXGxspmsykxMVFpaWnq16+fCgoKHJOShw8froULF2r06NHq06ePJk2apAsXLmjjxo2aN2+e+vbtq4SEBD344IN69dVX1aVLF/3617/Wj370IyUkJFx2nbNnz9Yzzzyjvn37asCAAVq2bJlKSkoc2/ft26fs7GyNGTNGoaGh2rdvn06fPq2BAwe2wF8NQHMQdgC0qs6dOysmJkYvvviijh8/rurqakVGRurBBx/U//t//08Wi0UbN27Ub37zG91///06ffq0rFarbrvtNoWFhUmSfvzjH2vdunVasmSJnnnmGQUGBuq2225zHCMjI0OzZ8/WnXfeqaqqKt12223auHFjvVtXF/PII4+osLBQycnJateunaZPn64JEyaotLRUkhQYGKidO3dq+fLlKisrU1RUlF544QWNGzfOvX8wAM1mMZpykxoAAKCNYc4OAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwNcIOAAAwtf8PUDWqNddv1tgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_duration_histogram(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0476cdf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def standardise_length(df, seconds):\n",
    "\n",
    "    recListAdj = []\n",
    "    lenListAdj = []\n",
    "\n",
    "    for idx, row in df.iterrows():\n",
    "        rec = row['AudioData']\n",
    "        sr = row['SampleRates']\n",
    "        rec, index = librosa.effects.trim(rec, top_db=30)\n",
    "        threshold = sr * seconds \n",
    "        if len(rec) > threshold:\n",
    "            rec = rec[:threshold]\n",
    "        recListAdj.append(rec)\n",
    "        lenListAdj.append(len(rec)/sr)\n",
    "    df['AudioData'] = recListAdj\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "74e86885",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = standardise_length(df, 3)\n",
    "plot_duration_histogram(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "83c85e89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_all_audios(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6f1e7533",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_split_spectrogram(audio, sr, n_blocks):\n",
    "    frequencies, times, spec = spectrogram(audio, fs=sr)\n",
    "    times = (times * 1000).astype(int) #milliseconds\n",
    "    frequencies = frequencies.astype(int)\n",
    "    fig, ax = plt.subplots(1,1)\n",
    "    plt.pcolormesh(times, frequencies, 10*np.log10(spec)) #changed the scale to dB\n",
    "    plt.colorbar()\n",
    "    time_bounds = list()\n",
    "    freq_bounds = list()\n",
    "    #n_blocks is the same vertically and horizontally\n",
    "    #starting from vertical lines to create time bins:\n",
    "    v_step = int(times[len(times)-1] / n_blocks)\n",
    "    \n",
    "    for v in range(0, times[len(times)-1], v_step):\n",
    "        ax.axvline(v, color = 'red')\n",
    "        time_bounds.append(v)\n",
    "    #and then horizontal lines to create frequency bins:\n",
    "    h_step = int(frequencies[len(frequencies)-1]/n_blocks)\n",
    "    for h in range(0, frequencies[len(frequencies)-1], h_step):\n",
    "        ax.axhline(h, color = 'red')\n",
    "        freq_bounds.append(h)\n",
    "    plt.xlabel(\"Time (ms)\")\n",
    "    plt.xlim(left=0)\n",
    "    plt.ylabel(\"Frequency (Hz)\")\n",
    "    plt.show()\n",
    "\n",
    "def split_spectrogram(audio, sr, n_blocks):\n",
    "    frequencies, times, spec = spectrogram(audio, fs=sr)\n",
    "    times = (times * 1000).astype(int) #milliseconds\n",
    "    frequencies = frequencies.astype(int)\n",
    "        \n",
    "    \n",
    "    # Split the matrix into sub-arrays of size N along the first axis (rows)\n",
    "    rows_blocks = np.array_split(spec, n_blocks, axis=0)\n",
    "    #this is a list: each element of the list is a sub-matrix obtained slicing horizontally the spectrogram\n",
    "    # into n_blocks slices\n",
    "\n",
    "    # Split the sub_matrices into sub-arrays of size NxN along the second axis (columns)\n",
    "    blocks = [ np.array_split(block, n_blocks, axis=1) for block in rows_blocks]\n",
    "    #This is, instead, a list of lists: the first element is a list containing all the blocks obtained from \n",
    "    #the first vertical slice\n",
    "\n",
    "\n",
    "    # Initialize arrays to store the mean and standard deviation of each block\n",
    "    block_means = np.zeros((n_blocks,n_blocks))\n",
    "    block_stds = np.zeros((n_blocks,n_blocks))\n",
    "\n",
    "    \n",
    "\n",
    "    # Iterate over blocks\n",
    "    for i in range(0, n_blocks):\n",
    "        for j in range(0, n_blocks):\n",
    "            # Compute mean and standard deviation of the block\n",
    "             #i is the index of the horizontal split  (indicating which column)\n",
    "            block = blocks[i][j] #j is the index of the vertical split (indicating which row)\n",
    "            mean = np.mean(block)\n",
    "            std = np.std(block)\n",
    "            minimum = np.min(block)\n",
    "            maximum = np.max(block)\n",
    "            block_means[i,j] = mean\n",
    "            block_stds[i,j] = std\n",
    "    return block_means.tolist(), block_stds.tolist(), blocks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9c7f11cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#adding the statistics of the blocks into the dataframe\n",
    "\n",
    "def add_stats_df(df,sr,n_blocks):\n",
    "    for i in range(0,n_blocks):\n",
    "        for j in range(0, n_blocks):\n",
    "            idx = j+n_blocks*i\n",
    "            col_name_mean = 'block_mean_'+str(idx)\n",
    "            col_name_dev = 'block_dev_'+str(idx)\n",
    "            df[col_name_mean] = ''\n",
    "            df[col_name_dev] = ''\n",
    "    all_means = list()\n",
    "    all_std_devs = list()\n",
    "    all_mins = list()\n",
    "    all_maxs = list()\n",
    "\n",
    "    for index, row in df.iterrows():\n",
    "        audio = row['AudioData']\n",
    "        meanTemp, stdTemp, blocksTemps = split_spectrogram(audio, sr, n_blocks)\n",
    "        for i in range(0,n_blocks):\n",
    "            for j in range(0, n_blocks):\n",
    "                idx = j+n_blocks*i\n",
    "                col_name_mean = 'block_mean_'+str(idx)\n",
    "                col_name_dev = 'block_dev_'+str(idx)\n",
    "                df.loc[index, col_name_mean] = meanTemp[i][j]\n",
    "                df.loc[index, col_name_dev] = stdTemp[i][j]\n",
    "        \n",
    "   \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b4f5e93c",
   "metadata": {},
   "outputs": [],
   "source": [
    "label_dictionary = {'increase volume': 0, 'decrease volume': 1, 'increase heat': 2, 'decrease heat':3, 'change language none':4,'activate music':5, 'deactivate lights':6 }\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c0ea549f",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "# random forest hyperparameters:\n",
    "RF_params = {\n",
    "    'n_estimators' : [50, 100, 150, 200],\n",
    "    'criterion' : ['gini', 'entropy'],\n",
    "    'max_depth' : [5, 10, 20, 50]\n",
    "}\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e02551d",
   "metadata": {},
   "source": [
    "# Option 1 \n",
    "For each possible number of blocks, search for the best hyperparameter configuration (this is super slow)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "529935a9",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# blocks:  2\n",
      "The best accuracy for  2  blocks is  0.3419583967529173  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  4\n",
      "The best accuracy for  4  blocks is  0.480466768138001  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  6\n",
      "The best accuracy for  6  blocks is  0.5591070522577372  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  8\n",
      "The best accuracy for  8  blocks is  0.604261796042618  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  10\n",
      "The best accuracy for  10  blocks is  0.671740233384069  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  12\n",
      "The best accuracy for  12  blocks is  0.6884830035514967  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "# blocks:  14\n",
      "The best accuracy for  14  blocks is  0.7057331303906647  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n",
      "The best result in terms of accuracy is  0.7057331303906647 , obtained with n =  14  and the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n"
     ]
    }
   ],
   "source": [
    "df1 = df.copy() #working every time with a new copy of the original df\n",
    "n_blocks = range(2, 16, 2)\n",
    "best_configs = list() #best configuration for each different number of blocks\n",
    "best_accuracies = list() #higher accuracy for each different number of blocks\n",
    "for n in n_blocks:\n",
    "    print('# blocks: ', n)\n",
    "    add_stats_df(df1, 16000, n)\n",
    "    y = df1['action'] + ' ' + df1['object']\n",
    "    y = y.replace(label_dictionary)\n",
    "    X = df1.drop(['action', 'object', 'path', 'speakerId', 'AudioData'], axis = 1)\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=31)\n",
    "    accuracies_rf = list()\n",
    "    configs_rf = list()\n",
    "    for config in ParameterGrid(RF_params):\n",
    "        rf = RandomForestClassifier(**config)\n",
    "        rf.fit(X_train, y_train)\n",
    "        acc = accuracy_score(y_test, rf.predict(X_test))\n",
    "        accuracies_rf.append(acc)\n",
    "        configs_rf.append(config)\n",
    "    max_val = max(accuracies_rf)\n",
    "    max_idx = accuracies_rf.index(max_val)\n",
    "    best_conf = configs_rf[max_idx]\n",
    "    best_configs.append(best_conf)\n",
    "    best_accuracies.append(max_val)\n",
    "    print('The best accuracy for ', n, ' blocks is ', max_val , ' obtained with the configuration ', config)\n",
    "\n",
    "best_acc = max(best_accuracies)    \n",
    "best_idx = best_accuracies.index(best_acc)\n",
    "best_tuning = best_configs[best_idx]\n",
    "best_n = n_blocks[best_idx]\n",
    "print('The best result in terms of accuracy is ', best_acc, ', obtained with n = ', best_n, ' and the configuration ', best_tuning )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "865e1297",
   "metadata": {},
   "source": [
    "# Option 2:\n",
    "Look for the most accurate configuration without blocks (meaning only one block), then use that configuration to tune the number of blocks to used based on which one returns a higher accuracy (this is faster - the configuration returned is the same as before (entropy-50-200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dceae1d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best accuracy is  0.313039066463724  obtained with the configuration  {'criterion': 'entropy', 'max_depth': 50, 'n_estimators': 200}\n"
     ]
    }
   ],
   "source": [
    "df2 = df.copy()\n",
    "\n",
    "add_stats_df(df2, 16000, 1)\n",
    "y = df2['action'] + ' ' + df2['object']\n",
    "y = y.replace(label_dictionary)\n",
    "X = df2.drop(['action', 'object', 'path', 'speakerId', 'AudioData'], axis = 1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=31)\n",
    "accuracies_rf = list()\n",
    "configs_rf = list()\n",
    "\n",
    "for config in ParameterGrid(RF_params):\n",
    "    rf = RandomForestClassifier(**config)\n",
    "    rf.fit(X_train, y_train)\n",
    "    acc = accuracy_score(y_test, rf.predict(X_test))\n",
    "    accuracies_rf.append(acc)\n",
    "    configs_rf.append(config)\n",
    "\n",
    "max_val = max(accuracies_rf)\n",
    "max_idx = accuracies_rf.index(max_val)\n",
    "best_conf = configs_rf[max_idx]\n",
    "print('The best accuracy is ', max_val , ' obtained with the configuration ', config)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "e0a1308a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# blocks:  2\n",
      "# blocks:  4\n",
      "# blocks:  6\n",
      "# blocks:  8\n",
      "# blocks:  10\n",
      "# blocks:  12\n",
      "# blocks:  14\n",
      "# blocks:  16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/lib/python3.10/site-packages/numpy/core/fromnumeric.py:3432: RuntimeWarning: Mean of empty slice.\n",
      "  return _methods._mean(a, axis=axis, dtype=dtype,\n",
      "/opt/homebrew/lib/python3.10/site-packages/numpy/core/_methods.py:190: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n",
      "/opt/homebrew/lib/python3.10/site-packages/numpy/core/_methods.py:265: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  ret = _var(a, axis=axis, dtype=dtype, out=out, ddof=ddof,\n",
      "/opt/homebrew/lib/python3.10/site-packages/numpy/core/_methods.py:223: RuntimeWarning: invalid value encountered in divide\n",
      "  arrmean = um.true_divide(arrmean, div, out=arrmean, casting='unsafe',\n",
      "/opt/homebrew/lib/python3.10/site-packages/numpy/core/_methods.py:257: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "zero-size array to reduction operation minimum which has no identity",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn [21], line 6\u001b[0m\n\u001b[1;32m      4\u001b[0m df2 \u001b[38;5;241m=\u001b[39m df\u001b[38;5;241m.\u001b[39mcopy() \u001b[38;5;66;03m#working every time with a new copy of the original df\u001b[39;00m\n\u001b[1;32m      5\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m# blocks: \u001b[39m\u001b[38;5;124m'\u001b[39m, n)\n\u001b[0;32m----> 6\u001b[0m add_stats_df(df2, \u001b[38;5;241m16000\u001b[39m, n)\n\u001b[1;32m      7\u001b[0m y \u001b[38;5;241m=\u001b[39m df2[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124maction\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m+\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m \u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;241m+\u001b[39m df2[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mobject\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[1;32m      8\u001b[0m X \u001b[38;5;241m=\u001b[39m df2\u001b[38;5;241m.\u001b[39mdrop([\u001b[38;5;124m'\u001b[39m\u001b[38;5;124maction\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mobject\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpath\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mspeakerId\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mAudioData\u001b[39m\u001b[38;5;124m'\u001b[39m], axis \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m)\n",
      "Cell \u001b[0;32mIn [16], line 18\u001b[0m, in \u001b[0;36madd_stats_df\u001b[0;34m(df, sr, n_blocks)\u001b[0m\n\u001b[1;32m     16\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m index, row \u001b[38;5;129;01min\u001b[39;00m df\u001b[38;5;241m.\u001b[39miterrows():\n\u001b[1;32m     17\u001b[0m     audio \u001b[38;5;241m=\u001b[39m row[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mAudioData\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[0;32m---> 18\u001b[0m     meanTemp, stdTemp, blocksTemps \u001b[38;5;241m=\u001b[39m \u001b[43msplit_spectrogram\u001b[49m\u001b[43m(\u001b[49m\u001b[43maudio\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43msr\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mn_blocks\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     19\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;241m0\u001b[39m,n_blocks):\n\u001b[1;32m     20\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m j \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;241m0\u001b[39m, n_blocks):\n",
      "Cell \u001b[0;32mIn [11], line 58\u001b[0m, in \u001b[0;36msplit_spectrogram\u001b[0;34m(audio, sr, n_blocks)\u001b[0m\n\u001b[1;32m     56\u001b[0m mean \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mmean(block)\n\u001b[1;32m     57\u001b[0m std \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mstd(block)\n\u001b[0;32m---> 58\u001b[0m minimum \u001b[38;5;241m=\u001b[39m \u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmin\u001b[49m\u001b[43m(\u001b[49m\u001b[43mblock\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     59\u001b[0m maximum \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mmax(block)\n\u001b[1;32m     60\u001b[0m block_means[i,j] \u001b[38;5;241m=\u001b[39m mean\n",
      "File \u001b[0;32m<__array_function__ internals>:180\u001b[0m, in \u001b[0;36mamin\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
      "File \u001b[0;32m/opt/homebrew/lib/python3.10/site-packages/numpy/core/fromnumeric.py:2918\u001b[0m, in \u001b[0;36mamin\u001b[0;34m(a, axis, out, keepdims, initial, where)\u001b[0m\n\u001b[1;32m   2802\u001b[0m \u001b[38;5;129m@array_function_dispatch\u001b[39m(_amin_dispatcher)\n\u001b[1;32m   2803\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mamin\u001b[39m(a, axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, out\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, keepdims\u001b[38;5;241m=\u001b[39mnp\u001b[38;5;241m.\u001b[39m_NoValue, initial\u001b[38;5;241m=\u001b[39mnp\u001b[38;5;241m.\u001b[39m_NoValue,\n\u001b[1;32m   2804\u001b[0m          where\u001b[38;5;241m=\u001b[39mnp\u001b[38;5;241m.\u001b[39m_NoValue):\n\u001b[1;32m   2805\u001b[0m     \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m   2806\u001b[0m \u001b[38;5;124;03m    Return the minimum of an array or minimum along an axis.\u001b[39;00m\n\u001b[1;32m   2807\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   2916\u001b[0m \u001b[38;5;124;03m    6\u001b[39;00m\n\u001b[1;32m   2917\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m-> 2918\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_wrapreduction\u001b[49m\u001b[43m(\u001b[49m\u001b[43ma\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mminimum\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mmin\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mout\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2919\u001b[0m \u001b[43m                          \u001b[49m\u001b[43mkeepdims\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mkeepdims\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minitial\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minitial\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mwhere\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mwhere\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/homebrew/lib/python3.10/site-packages/numpy/core/fromnumeric.py:86\u001b[0m, in \u001b[0;36m_wrapreduction\u001b[0;34m(obj, ufunc, method, axis, dtype, out, **kwargs)\u001b[0m\n\u001b[1;32m     83\u001b[0m         \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m     84\u001b[0m             \u001b[38;5;28;01mreturn\u001b[39;00m reduction(axis\u001b[38;5;241m=\u001b[39maxis, out\u001b[38;5;241m=\u001b[39mout, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mpasskwargs)\n\u001b[0;32m---> 86\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mufunc\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreduce\u001b[49m\u001b[43m(\u001b[49m\u001b[43mobj\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdtype\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mout\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[43mpasskwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mValueError\u001b[0m: zero-size array to reduction operation minimum which has no identity"
     ]
    }
   ],
   "source": [
    "n_blocks = range(2, 22, 2)\n",
    "accuracies_rf_n = list()\n",
    "for n in n_blocks:\n",
    "    df2 = df.copy() #working every time with a new copy of the original df\n",
    "    print('# blocks: ', n)\n",
    "    add_stats_df(df2, 16000, n)\n",
    "    y = df2['action'] + ' ' + df2['object']\n",
    "    X = df2.drop(['action', 'object', 'path', 'speakerId', 'AudioData'], axis = 1)\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=31)\n",
    "    rf = RandomForestClassifier(criterion='entropy', max_depth=50, n_estimators = 200)\n",
    "    rf.fit(X_train, y_train)\n",
    "    acc = accuracy_score(y_test, rf.predict(X_test))\n",
    "    accuracies_rf_n.append(acc)\n",
    "    \n",
    "max_val_n = max(accuracies_rf_n)\n",
    "max_idx_n = accuracies_rf_n.index(max_val_n)\n",
    "best_n = n_blocks[max_idx_n]\n",
    "print('The best accuracy is ', max_val_n , ', obtained with n = ', best_n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3572d5f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.3363774733637747,\n",
       " 0.4682902080162354,\n",
       " 0.5509893455098934,\n",
       " 0.6047691527143582,\n",
       " 0.6712328767123288,\n",
       " 0.6839167935058346,\n",
       " 0.6986301369863014]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#it still gives the same error with n = 16 ???\n",
    "accuracies_rf_n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "12cd0b5d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***** DATAFRAME INFO *****\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1455 entries, 0 to 1454\n",
      "Data columns (total 8 columns):\n",
      " #   Column                                 Non-Null Count  Dtype \n",
      "---  ------                                 --------------  ----- \n",
      " 0   Id                                     1455 non-null   int64 \n",
      " 1   path                                   1455 non-null   object\n",
      " 2   speakerId                              1455 non-null   object\n",
      " 3   Self-reported fluency level            1455 non-null   object\n",
      " 4   First Language spoken                  1455 non-null   object\n",
      " 5   Current language used for work/school  1455 non-null   object\n",
      " 6   gender                                 1455 non-null   object\n",
      " 7   ageRange                               1455 non-null   object\n",
      "dtypes: int64(1), object(7)\n",
      "memory usage: 91.1+ KB\n",
      "None\n",
      "\n",
      "\n",
      "***** HEAD *****\n",
      "\n",
      "   Id                                               path          speakerId  \\\n",
      "0   0  dsl_data/audio/speakers/NgQEvO2x7Vh3xy2xz/f53c...  NgQEvO2x7Vh3xy2xz   \n",
      "1   1  dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/1d5f8...   k5bqyxx2lzIbrlg9   \n",
      "2   2  dsl_data/audio/speakers/7B4XmNppyrCK977p/1c0d5...   7B4XmNppyrCK977p   \n",
      "3   3  dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/275c3...   k5bqyxx2lzIbrlg9   \n",
      "4   4  dsl_data/audio/speakers/V4ZbwLm9G5irobWn/b7c7a...   V4ZbwLm9G5irobWn   \n",
      "\n",
      "  Self-reported fluency level     First Language spoken  \\\n",
      "0                       native  English (United States)   \n",
      "1                       native  English (United States)   \n",
      "2                       native  English (United States)   \n",
      "3                       native  English (United States)   \n",
      "4                       native  English (United States)   \n",
      "\n",
      "  Current language used for work/school  gender ageRange  \n",
      "0               English (United States)    male    22-40  \n",
      "1               English (United States)    male    22-40  \n",
      "2               English (United States)    male    41-65  \n",
      "3               English (United States)    male    22-40  \n",
      "4               English (United States)  female    41-65  \n",
      "\n",
      "\n",
      " ***** POSSIBLE CLASSIFICATION OUTCOMES *****\n",
      "\n",
      "\n",
      "\n",
      " ***** SAMPLE RATES *****\n",
      "SampleRates\n",
      "16000    1455\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Self-reported fluency level *****\n",
      "Self-reported fluency level \n",
      "native    1455\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** First Language spoken *****\n",
      "First Language spoken\n",
      "English (United States)    1455\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Current language used for work/school *****\n",
      "Current language used for work/school\n",
      "English (United States)    1455\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Gender *****\n",
      "gender\n",
      "female    773\n",
      "male      682\n",
      "dtype: int64\n",
      "\n",
      "\n",
      " ***** Age range *****\n",
      "ageRange\n",
      "22-40    907\n",
      "41-65    548\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# assuming the best result is obtained with criterion='entropy', max_depth=50, n_estimators = 200, n = 14:\n",
    "\n",
    "n_blocks = 14\n",
    "eval_df = load_dataframe(eval_path)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "ab7aa107",
   "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>Id</th>\n",
       "      <th>Self-reported fluency level</th>\n",
       "      <th>First Language spoken</th>\n",
       "      <th>Current language used for work/school</th>\n",
       "      <th>gender</th>\n",
       "      <th>ageRange</th>\n",
       "      <th>SampleRates</th>\n",
       "      <th>block_mean_0</th>\n",
       "      <th>block_dev_0</th>\n",
       "      <th>block_mean_1</th>\n",
       "      <th>...</th>\n",
       "      <th>block_mean_191</th>\n",
       "      <th>block_dev_191</th>\n",
       "      <th>block_mean_192</th>\n",
       "      <th>block_dev_192</th>\n",
       "      <th>block_mean_193</th>\n",
       "      <th>block_dev_193</th>\n",
       "      <th>block_mean_194</th>\n",
       "      <th>block_dev_194</th>\n",
       "      <th>block_mean_195</th>\n",
       "      <th>block_dev_195</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3559</th>\n",
       "      <td>3559</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000966</td>\n",
       "      <td>0.003437</td>\n",
       "      <td>0.005358</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8274</th>\n",
       "      <td>8274</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000686</td>\n",
       "      <td>0.001362</td>\n",
       "      <td>0.001413</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5344</th>\n",
       "      <td>5344</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>0.001968</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>297</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000032</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>0.004226</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>0.000274</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>0.00005</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1124</th>\n",
       "      <td>1124</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.002902</td>\n",
       "      <td>0.003019</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1460</th>\n",
       "      <td>1460</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004585</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3271</th>\n",
       "      <td>3271</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000988</td>\n",
       "      <td>0.001162</td>\n",
       "      <td>0.00015</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4764</th>\n",
       "      <td>4764</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000437</td>\n",
       "      <td>0.001181</td>\n",
       "      <td>0.001544</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8058</th>\n",
       "      <td>8058</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.001087</td>\n",
       "      <td>0.003524</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9503</th>\n",
       "      <td>9503</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.005892</td>\n",
       "      <td>0.019757</td>\n",
       "      <td>0.031482</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3654</th>\n",
       "      <td>3654</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1059</th>\n",
       "      <td>1059</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.002084</td>\n",
       "      <td>0.006662</td>\n",
       "      <td>0.001221</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9456</th>\n",
       "      <td>9456</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.004093</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000161</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8290</th>\n",
       "      <td>8290</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000071</td>\n",
       "      <td>0.000294</td>\n",
       "      <td>0.004128</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>0.000021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7625</th>\n",
       "      <td>7625</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000096</td>\n",
       "      <td>0.000337</td>\n",
       "      <td>0.000996</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>143</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.000026</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000098</td>\n",
       "      <td>0.000118</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9657</th>\n",
       "      <td>9657</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000209</td>\n",
       "      <td>0.000759</td>\n",
       "      <td>0.001096</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5643</th>\n",
       "      <td>5643</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000403</td>\n",
       "      <td>0.001066</td>\n",
       "      <td>0.003314</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2909</th>\n",
       "      <td>2909</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000043</td>\n",
       "      <td>0.005263</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3819</th>\n",
       "      <td>3819</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000151</td>\n",
       "      <td>0.000384</td>\n",
       "      <td>0.002949</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 399 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Id  Self-reported fluency level   First Language spoken  \\\n",
       "3559  3559                             0                      0   \n",
       "8274  8274                             0                      0   \n",
       "5344  5344                             0                      0   \n",
       "297    297                             0                      0   \n",
       "1124  1124                             0                      0   \n",
       "1460  1460                             0                      0   \n",
       "3271  3271                             0                      0   \n",
       "4764  4764                             0                      0   \n",
       "8058  8058                             0                      0   \n",
       "9503  9503                             0                      0   \n",
       "3654  3654                             0                      0   \n",
       "1059  1059                             0                      0   \n",
       "9456  9456                             0                      0   \n",
       "8290  8290                             0                      0   \n",
       "7625  7625                             0                      0   \n",
       "143    143                             0                      0   \n",
       "9657  9657                             0                      0   \n",
       "5643  5643                             0                      0   \n",
       "2909  2909                             0                      0   \n",
       "3819  3819                             0                      0   \n",
       "\n",
       "      Current language used for work/school  gender  ageRange  SampleRates  \\\n",
       "3559                                      0       1         0        16000   \n",
       "8274                                      0       0         1        16000   \n",
       "5344                                      0       1         0        16000   \n",
       "297                                       0       0         1        16000   \n",
       "1124                                      0       1         1        16000   \n",
       "1460                                      0       1         0        16000   \n",
       "3271                                      0       0         0        16000   \n",
       "4764                                      0       0         1        16000   \n",
       "8058                                      0       1         0        16000   \n",
       "9503                                      0       0         1        16000   \n",
       "3654                                      0       0         0        16000   \n",
       "1059                                      0       1         0        16000   \n",
       "9456                                      0       0         1        16000   \n",
       "8290                                      0       0         1        16000   \n",
       "7625                                      0       0         0        16000   \n",
       "143                                       0       0         0        16000   \n",
       "9657                                      0       0         0        16000   \n",
       "5643                                      0       1         0        16000   \n",
       "2909                                      0       1         0        16000   \n",
       "3819                                      0       0         1        16000   \n",
       "\n",
       "     block_mean_0 block_dev_0 block_mean_1  ... block_mean_191 block_dev_191  \\\n",
       "3559     0.000966    0.003437     0.005358  ...            0.0           0.0   \n",
       "8274     0.000686    0.001362     0.001413  ...            0.0           0.0   \n",
       "5344     0.000012    0.000022     0.001968  ...            0.0           0.0   \n",
       "297      0.000032    0.000073     0.004226  ...            0.0           0.0   \n",
       "1124        0.001    0.002902     0.003019  ...            0.0           0.0   \n",
       "1460          0.0         0.0     0.004585  ...            0.0           0.0   \n",
       "3271     0.000988    0.001162      0.00015  ...            0.0           0.0   \n",
       "4764     0.000437    0.001181     0.001544  ...            0.0           0.0   \n",
       "8058     0.001087    0.003524     0.000008  ...       0.000004      0.000005   \n",
       "9503     0.005892    0.019757     0.031482  ...            0.0           0.0   \n",
       "3654     0.000003     0.00001          0.0  ...            0.0           0.0   \n",
       "1059     0.002084    0.006662     0.001221  ...            0.0           0.0   \n",
       "9456     0.000001    0.000003     0.004093  ...            0.0           0.0   \n",
       "8290     0.000071    0.000294     0.004128  ...            0.0           0.0   \n",
       "7625     0.000096    0.000337     0.000996  ...            0.0           0.0   \n",
       "143      0.000009    0.000026          0.0  ...       0.000098      0.000118   \n",
       "9657     0.000209    0.000759     0.001096  ...            0.0           0.0   \n",
       "5643     0.000403    0.001066     0.003314  ...            0.0           0.0   \n",
       "2909     0.000013    0.000043     0.005263  ...            0.0           0.0   \n",
       "3819     0.000151    0.000384     0.002949  ...            0.0           0.0   \n",
       "\n",
       "     block_mean_192 block_dev_192 block_mean_193 block_dev_193 block_mean_194  \\\n",
       "3559            0.0           0.0            0.0           0.0            0.0   \n",
       "8274            0.0           0.0            0.0           0.0            0.0   \n",
       "5344            0.0           0.0            0.0           0.0            0.0   \n",
       "297             0.0      0.000001       0.000159      0.000274       0.000035   \n",
       "1124            0.0           0.0            0.0           0.0            0.0   \n",
       "1460            0.0           0.0            0.0           0.0            0.0   \n",
       "3271            0.0           0.0            0.0           0.0            0.0   \n",
       "4764            0.0           0.0            0.0           0.0            0.0   \n",
       "8058       0.000001      0.000001            0.0      0.000001            0.0   \n",
       "9503            0.0           0.0            0.0           0.0            0.0   \n",
       "3654            0.0           0.0            0.0           0.0            0.0   \n",
       "1059            0.0           0.0            0.0           0.0            0.0   \n",
       "9456            0.0           0.0            0.0           0.0       0.000044   \n",
       "8290            0.0           0.0            0.0           0.0            0.0   \n",
       "7625            0.0           0.0            0.0           0.0            0.0   \n",
       "143             0.0      0.000001            0.0           0.0            0.0   \n",
       "9657            0.0           0.0            0.0           0.0            0.0   \n",
       "5643            0.0           0.0            0.0           0.0            0.0   \n",
       "2909            0.0           0.0            0.0           0.0            0.0   \n",
       "3819            0.0           0.0            0.0           0.0            0.0   \n",
       "\n",
       "     block_dev_194 block_mean_195 block_dev_195  \n",
       "3559           0.0            0.0           0.0  \n",
       "8274           0.0       0.000019      0.000071  \n",
       "5344           0.0            0.0           0.0  \n",
       "297        0.00005       0.000002      0.000005  \n",
       "1124           0.0       0.000001      0.000001  \n",
       "1460           0.0            0.0           0.0  \n",
       "3271           0.0            0.0           0.0  \n",
       "4764           0.0            0.0           0.0  \n",
       "8058           0.0            0.0           0.0  \n",
       "9503           0.0            0.0      0.000001  \n",
       "3654           0.0            0.0           0.0  \n",
       "1059           0.0            0.0           0.0  \n",
       "9456      0.000161            0.0           0.0  \n",
       "8290           0.0       0.000007      0.000021  \n",
       "7625           0.0            0.0           0.0  \n",
       "143            0.0            0.0           0.0  \n",
       "9657           0.0            0.0           0.0  \n",
       "5643           0.0            0.0           0.0  \n",
       "2909           0.0            0.0           0.0  \n",
       "3819           0.0            0.0           0.0  \n",
       "\n",
       "[20 rows x 399 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "48a29e25",
   "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>Id</th>\n",
       "      <th>path</th>\n",
       "      <th>speakerId</th>\n",
       "      <th>Self-reported fluency level</th>\n",
       "      <th>First Language spoken</th>\n",
       "      <th>Current language used for work/school</th>\n",
       "      <th>gender</th>\n",
       "      <th>ageRange</th>\n",
       "      <th>SampleRates</th>\n",
       "      <th>AudioData</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>dsl_data/audio/speakers/NgQEvO2x7Vh3xy2xz/f53c...</td>\n",
       "      <td>NgQEvO2x7Vh3xy2xz</td>\n",
       "      <td>native</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>male</td>\n",
       "      <td>22-40</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/1d5f8...</td>\n",
       "      <td>k5bqyxx2lzIbrlg9</td>\n",
       "      <td>native</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>male</td>\n",
       "      <td>22-40</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>dsl_data/audio/speakers/7B4XmNppyrCK977p/1c0d5...</td>\n",
       "      <td>7B4XmNppyrCK977p</td>\n",
       "      <td>native</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>male</td>\n",
       "      <td>41-65</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/275c3...</td>\n",
       "      <td>k5bqyxx2lzIbrlg9</td>\n",
       "      <td>native</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>male</td>\n",
       "      <td>22-40</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>dsl_data/audio/speakers/V4ZbwLm9G5irobWn/b7c7a...</td>\n",
       "      <td>V4ZbwLm9G5irobWn</td>\n",
       "      <td>native</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>English (United States)</td>\n",
       "      <td>female</td>\n",
       "      <td>41-65</td>\n",
       "      <td>16000</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id                                               path          speakerId  \\\n",
       "0   0  dsl_data/audio/speakers/NgQEvO2x7Vh3xy2xz/f53c...  NgQEvO2x7Vh3xy2xz   \n",
       "1   1  dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/1d5f8...   k5bqyxx2lzIbrlg9   \n",
       "2   2  dsl_data/audio/speakers/7B4XmNppyrCK977p/1c0d5...   7B4XmNppyrCK977p   \n",
       "3   3  dsl_data/audio/speakers/k5bqyxx2lzIbrlg9/275c3...   k5bqyxx2lzIbrlg9   \n",
       "4   4  dsl_data/audio/speakers/V4ZbwLm9G5irobWn/b7c7a...   V4ZbwLm9G5irobWn   \n",
       "\n",
       "  Self-reported fluency level     First Language spoken  \\\n",
       "0                       native  English (United States)   \n",
       "1                       native  English (United States)   \n",
       "2                       native  English (United States)   \n",
       "3                       native  English (United States)   \n",
       "4                       native  English (United States)   \n",
       "\n",
       "  Current language used for work/school  gender ageRange  SampleRates  \\\n",
       "0               English (United States)    male    22-40        16000   \n",
       "1               English (United States)    male    22-40        16000   \n",
       "2               English (United States)    male    41-65        16000   \n",
       "3               English (United States)    male    22-40        16000   \n",
       "4               English (United States)  female    41-65        16000   \n",
       "\n",
       "                                           AudioData  \n",
       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "1  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "2  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "3  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
       "4  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "10dd9d05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SampleRates\n",
      "16000    1455\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "target_rate = 16000\n",
    "edit_sample_rates(eval_df, target_rate) #bring all sr's to 16kHz\n",
    "encode_categorical(eval_df)\n",
    "eval_df = normalise_audios(eval_df)\n",
    "eval_df = standardise_length(eval_df, 3)\n",
    "add_stats_df(eval_df, target_rate, n_blocks)\n",
    "\n",
    "\n",
    "eval_X = eval_df.drop(['path', 'speakerId', 'AudioData'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "2764e782",
   "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>Id</th>\n",
       "      <th>Self-reported fluency level</th>\n",
       "      <th>First Language spoken</th>\n",
       "      <th>Current language used for work/school</th>\n",
       "      <th>gender</th>\n",
       "      <th>ageRange</th>\n",
       "      <th>SampleRates</th>\n",
       "      <th>block_mean_0</th>\n",
       "      <th>block_dev_0</th>\n",
       "      <th>block_mean_1</th>\n",
       "      <th>...</th>\n",
       "      <th>block_mean_191</th>\n",
       "      <th>block_dev_191</th>\n",
       "      <th>block_mean_192</th>\n",
       "      <th>block_dev_192</th>\n",
       "      <th>block_mean_193</th>\n",
       "      <th>block_dev_193</th>\n",
       "      <th>block_mean_194</th>\n",
       "      <th>block_dev_194</th>\n",
       "      <th>block_mean_195</th>\n",
       "      <th>block_dev_195</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000734</td>\n",
       "      <td>0.001809</td>\n",
       "      <td>0.004086</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000838</td>\n",
       "      <td>0.004087</td>\n",
       "      <td>0.002369</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.003844</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.004045</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.001288</td>\n",
       "      <td>0.00328</td>\n",
       "      <td>0.004587</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 399 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Self-reported fluency level   First Language spoken  \\\n",
       "0   0                             0                      0   \n",
       "1   1                             0                      0   \n",
       "2   2                             0                      0   \n",
       "3   3                             0                      0   \n",
       "4   4                             0                      0   \n",
       "\n",
       "   Current language used for work/school  gender  ageRange  SampleRates  \\\n",
       "0                                      0       1         0        16000   \n",
       "1                                      0       1         0        16000   \n",
       "2                                      0       1         1        16000   \n",
       "3                                      0       1         0        16000   \n",
       "4                                      0       0         1        16000   \n",
       "\n",
       "  block_mean_0 block_dev_0 block_mean_1  ... block_mean_191 block_dev_191  \\\n",
       "0     0.000734    0.001809     0.004086  ...            0.0           0.0   \n",
       "1     0.000838    0.004087     0.002369  ...            0.0           0.0   \n",
       "2     0.000006    0.000011     0.003844  ...            0.0           0.0   \n",
       "3     0.000001    0.000003     0.004045  ...            0.0           0.0   \n",
       "4     0.001288     0.00328     0.004587  ...            0.0           0.0   \n",
       "\n",
       "  block_mean_192 block_dev_192 block_mean_193 block_dev_193 block_mean_194  \\\n",
       "0            0.0           0.0            0.0           0.0            0.0   \n",
       "1            0.0           0.0            0.0           0.0            0.0   \n",
       "2            0.0           0.0            0.0           0.0            0.0   \n",
       "3            0.0           0.0            0.0           0.0            0.0   \n",
       "4            0.0           0.0            0.0           0.0            0.0   \n",
       "\n",
       "  block_dev_194 block_mean_195 block_dev_195  \n",
       "0           0.0            0.0           0.0  \n",
       "1           0.0            0.0           0.0  \n",
       "2           0.0            0.0           0.0  \n",
       "3           0.0            0.0           0.0  \n",
       "4           0.0            0.0           0.0  \n",
       "\n",
       "[5 rows x 399 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "d3abdd62",
   "metadata": {},
   "outputs": [],
   "source": [
    "add_stats_df(df, 16000, 14)\n",
    "y = df['action'] + ' ' + df['object']\n",
    "y = y.replace(label_dictionary)\n",
    "X = df.drop(['action', 'object', 'path', 'speakerId', 'AudioData'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "95a2f357",
   "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>Id</th>\n",
       "      <th>Self-reported fluency level</th>\n",
       "      <th>First Language spoken</th>\n",
       "      <th>Current language used for work/school</th>\n",
       "      <th>gender</th>\n",
       "      <th>ageRange</th>\n",
       "      <th>SampleRates</th>\n",
       "      <th>block_mean_0</th>\n",
       "      <th>block_dev_0</th>\n",
       "      <th>block_mean_1</th>\n",
       "      <th>...</th>\n",
       "      <th>block_mean_191</th>\n",
       "      <th>block_dev_191</th>\n",
       "      <th>block_mean_192</th>\n",
       "      <th>block_dev_192</th>\n",
       "      <th>block_mean_193</th>\n",
       "      <th>block_dev_193</th>\n",
       "      <th>block_mean_194</th>\n",
       "      <th>block_dev_194</th>\n",
       "      <th>block_mean_195</th>\n",
       "      <th>block_dev_195</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000072</td>\n",
       "      <td>0.00028</td>\n",
       "      <td>0.004955</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.00038</td>\n",
       "      <td>0.001109</td>\n",
       "      <td>0.004402</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.001429</td>\n",
       "      <td>0.003511</td>\n",
       "      <td>0.000686</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.000055</td>\n",
       "      <td>0.000091</td>\n",
       "      <td>0.000124</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16000</td>\n",
       "      <td>0.001332</td>\n",
       "      <td>0.003677</td>\n",
       "      <td>0.003191</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 399 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Self-reported fluency level   First Language spoken  \\\n",
       "0   0                             1                      0   \n",
       "1   1                             1                      0   \n",
       "2   2                             1                      0   \n",
       "3   3                             1                      0   \n",
       "4   4                             1                      0   \n",
       "\n",
       "   Current language used for work/school  gender  ageRange  SampleRates  \\\n",
       "0                                      0       0         0        16000   \n",
       "1                                      0       0         0        16000   \n",
       "2                                      0       0         0        16000   \n",
       "3                                      0       0         0        16000   \n",
       "4                                      0       0         0        16000   \n",
       "\n",
       "  block_mean_0 block_dev_0 block_mean_1  ... block_mean_191 block_dev_191  \\\n",
       "0     0.000072     0.00028     0.004955  ...            0.0           0.0   \n",
       "1      0.00038    0.001109     0.004402  ...            0.0           0.0   \n",
       "2     0.001429    0.003511     0.000686  ...            0.0           0.0   \n",
       "3     0.000055    0.000091     0.000124  ...            0.0           0.0   \n",
       "4     0.001332    0.003677     0.003191  ...            0.0           0.0   \n",
       "\n",
       "  block_mean_192 block_dev_192 block_mean_193 block_dev_193 block_mean_194  \\\n",
       "0            0.0           0.0            0.0           0.0       0.000003   \n",
       "1            0.0           0.0            0.0           0.0            0.0   \n",
       "2       0.000019      0.000044       0.000028      0.000031       0.000012   \n",
       "3            0.0           0.0       0.000013      0.000017            0.0   \n",
       "4            0.0           0.0            0.0           0.0            0.0   \n",
       "\n",
       "  block_dev_194 block_mean_195 block_dev_195  \n",
       "0      0.000003       0.000001      0.000001  \n",
       "1           0.0            0.0           0.0  \n",
       "2      0.000014       0.000002      0.000005  \n",
       "3           0.0            0.0           0.0  \n",
       "4           0.0            0.0           0.0  \n",
       "\n",
       "[5 rows x 399 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "957db208",
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestClassifier(criterion='entropy', max_depth=50, n_estimators = 200)\n",
    "rf.fit(X, y)\n",
    "eval_y = rf.predict(eval_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "8e2e7110",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 2, ..., 2, 0, 5])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "817ba919",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 'increase volume',\n",
       " 1: 'decrease volume',\n",
       " 2: 'increase heat',\n",
       " 3: 'decrease heat',\n",
       " 4: 'change language none',\n",
       " 5: 'activate music',\n",
       " 6: 'deactivate lights'}"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inverse_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "8e766b62",
   "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>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>increase volume</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>increase heat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>increase heat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>deactivate lights</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>decrease volume</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Predicted\n",
       "0    increase volume\n",
       "1      increase heat\n",
       "2      increase heat\n",
       "3  deactivate lights\n",
       "4    decrease volume"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inverse_dict = {v: k for k, v in label_dictionary.items()}\n",
    "solution = pd.DataFrame(eval_y, index = eval_X.index, columns = ['Predicted'])\n",
    "solution['Predicted'].replace(inverse_dict, inplace=True)\n",
    "solution.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "8a939d66",
   "metadata": {},
   "outputs": [],
   "source": [
    "solution.to_csv('draft_solution.csv', index_label='Id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a759ead",
   "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}