4_models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deep Learning Models for Speech Recognition"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"padding-left: 28px;\">**<font size=4>Data Science - Capstone Project Submission</font>**</span>\n",
    "\n",
    "* Student Name: **James Toop**\n",
    "* Student Pace: **Self Paced**\n",
    "* Scheduled project review date/time: **29th October 2021 @ 21:30 BST**\n",
    "* Instructor name: **Jeff Herman / James Irving**\n",
    "* Blog URL: **https://toopster.github.io/**\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import relevant libraries and modules for creating and training networks\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import wave\n",
    "import soundfile as sf\n",
    "import librosa, librosa.display\n",
    "import IPython.display as ipd\n",
    "import os\n",
    "import json\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.layers.experimental import preprocessing\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import models\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix\n",
    "\n",
    "import logging\n",
    "logging.getLogger(\"tensorflow\").setLevel(logging.ERROR)\n",
    "\n",
    "import pathlib\n",
    "from pathlib import Path\n",
    "\n",
    "import shared_functions.preprocessing as preprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set seed for reproducibility\n",
    "seed = 123\n",
    "tf.random.set_seed(seed)\n",
    "np.random.seed(seed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create model evaluation functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualise_results(results):\n",
    "    '''\n",
    "    Visualise the results from the model history plotting training and\n",
    "    validation accuracy and loss vs. epoch\n",
    "        \n",
    "        Params:\n",
    "            results: Model history\n",
    "    '''     \n",
    "    history = results.history\n",
    "\n",
    "    plt.figure(figsize=(20,8))\n",
    "    plt.xticks(fontsize=12)\n",
    "    plt.yticks(fontsize=12)\n",
    "    \n",
    "    plt.subplot(1, 2, 1)\n",
    "    plt.plot(history['val_loss'])\n",
    "    plt.plot(history['loss'])\n",
    "    plt.legend(['Validation Loss', 'Training Loss'], fontsize=12)\n",
    "    plt.title('Loss', fontsize=18)\n",
    "    plt.xlabel('Epochs', fontsize=14)\n",
    "    plt.ylabel('Loss', fontsize=14)\n",
    "    \n",
    "    plt.subplot(1, 2, 2)\n",
    "    plt.plot(history['val_acc'])\n",
    "    plt.plot(history['acc'])\n",
    "    plt.legend(['Validation Accuracy', 'Training Accuracy'], fontsize=12)\n",
    "    plt.title('Accuracy', fontsize=18)\n",
    "    plt.xlabel('Epochs', fontsize=14)\n",
    "    plt.ylabel('Accuracy', fontsize=14)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_performance(input_model,\n",
    "                      input_train_X,\n",
    "                      input_train_y,\n",
    "                      input_test_X, \n",
    "                      input_test_y\n",
    "                     ): \n",
    "    '''\n",
    "    Visualise the results from the model history plotting training and\n",
    "    validation accuracy and loss vs. epoch\n",
    "        \n",
    "        Params:\n",
    "            input_model:\n",
    "            input_train_X (ndarray): Inputs for the training dataset\n",
    "            input_train_y (ndarray): Targets for the training dataset\n",
    "            input_test_X (ndarray): Inputs for the test dataset\n",
    "            input_test_y (ndarray): Targets for the test dataset\n",
    "    '''    \n",
    "    # Evaluate training results\n",
    "    results_train = input_model.evaluate(input_train_X, \n",
    "                                         input_train_y, \n",
    "                                         verbose=0)\n",
    "    \n",
    "    # Evaluate test results\n",
    "    results_test = input_model.evaluate(input_test_X, \n",
    "                                        input_test_y, \n",
    "                                        verbose=0)\n",
    "    \n",
    "    # Create predictions based on model\n",
    "    predictions = input_model.predict(input_test_X)\n",
    "    \n",
    "    # Calculate confusion matrix\n",
    "    conf_matrix = confusion_matrix(input_test_y.argmax(axis=1), \n",
    "                                   np.round(predictions.argmax(axis=1)))\n",
    "    \n",
    "    # Output performance evaluation\n",
    "    print('CONFUSION MATRIX ------------------')\n",
    "    print(conf_matrix)\n",
    "\n",
    "    print('\\nTRAIN METRICS ----------------------')\n",
    "    print('Loss: {}'.format(results_train[0]))\n",
    "    print('Accuracy: {}%'.format(np.round(results_train[1]*100), 2))\n",
    "\n",
    "    print('\\nTEST METRICS ----------------------')\n",
    "    print('Loss: {}'.format(results_test[0]))\n",
    "    print('Accuracy: {}%'.format(np.round(results_test[1]*100), 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_array_shapes(X_train, y_train, X_val, y_val, X_test, y_test):\n",
    "    '''\n",
    "    Output the training, test and validation dataset shapes\n",
    "        \n",
    "        Params:\n",
    "            X_train (ndarray): Inputs for the training dataset\n",
    "            y_train (ndarray): Targets for the training dataset\n",
    "            X_val (ndarray): Inputs for the validation dataset\n",
    "            y_val (ndarray): Targets for the validation dataset\n",
    "            X_test (ndarray): Inputs for the test dataset\n",
    "            y_test (ndarray): Targets for the test dataset\n",
    "    '''       \n",
    "    print (\"Number of training samples: \" + str(X_train.shape[0]))\n",
    "    print (\"Number of testing samples: \" + str(X_test.shape[0]))\n",
    "    print (\"Number of validation samples: \" + str(X_val.shape[0]))\n",
    "    print (\"X_train shape: \" + str(X_train.shape))\n",
    "    print (\"y_train shape: \" + str(y_train.shape))\n",
    "    print (\"X_test shape: \" + str(X_test.shape))\n",
    "    print (\"y_test shape: \" + str(y_test.shape))\n",
    "    print (\"X_val shape: \" + str(X_val.shape))\n",
    "    print (\"y_val shape: \" + str(y_val.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def reformat_y(y):\n",
    "    '''\n",
    "    Reformats / One Hot Encodes targets\n",
    "        \n",
    "        Params:\n",
    "            y (ndarray): Input targets\n",
    "        \n",
    "        Returns:\n",
    "            y (ndarray): One hot encoded targets\n",
    "    '''       \n",
    "    y = LabelEncoder().fit_transform(y)\n",
    "    y = tf.keras.utils.to_categorical(y)\n",
    "    return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_model(save_model, save_path):\n",
    "    '''\n",
    "    Save the model\n",
    "        \n",
    "        Params:\n",
    "            save_model : Input Tensorflow model\n",
    "            save_path (str): Path to save model including file extension .h5\n",
    "    '''    \n",
    "    save_model.save(save_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create training, test and validation datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Datasets loaded...\n"
     ]
    }
   ],
   "source": [
    "# See shared_functions/preprocessing.py for create_train_test function\n",
    "# Create the train, test and validation datasets for the Speech Commands dataset using MFCCs\n",
    "sc_data_path = \"speech_commands_data.json\"\n",
    "(\n",
    "    sc_X_train,\n",
    "    sc_y_train,\n",
    "    sc_X_val,\n",
    "    sc_y_val,\n",
    "    sc_X_test,\n",
    "    sc_y_test,\n",
    ") = preprocess.create_train_test(sc_data_path, 'MFCCs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Datasets loaded...\n"
     ]
    }
   ],
   "source": [
    "# See shared_functions/preprocessing.py for create_train_test function\n",
    "# Create the train, test and validation datasets for the Ultrasuite Top 35 dataset using MFCCs\n",
    "us_data_path = 'ultrasuite_top35_data.json'\n",
    "(\n",
    "    us_X_train, \n",
    "    us_y_train, \n",
    "    us_X_val, \n",
    "    us_y_val, \n",
    "    us_X_test, \n",
    "    us_y_test,\n",
    ") = preprocess.create_train_test(us_data_path, 'MFCCs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training samples: 61052\n",
      "Number of testing samples: 19079\n",
      "Number of validation samples: 15263\n",
      "X_train shape: (61052, 44, 13, 1)\n",
      "y_train shape: (61052,)\n",
      "X_test shape: (19079, 44, 13, 1)\n",
      "y_test shape: (19079,)\n",
      "X_val shape: (15263, 44, 13, 1)\n",
      "y_val shape: (15263,)\n"
     ]
    }
   ],
   "source": [
    "# Check the array shapes for the Ultrasuite dataset\n",
    "check_array_shapes(sc_X_train, \n",
    "                   sc_y_train, \n",
    "                   sc_X_val, \n",
    "                   sc_y_val, \n",
    "                   sc_X_test, \n",
    "                   sc_y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training samples: 3942\n",
      "Number of testing samples: 1233\n",
      "Number of validation samples: 986\n",
      "X_train shape: (3942, 44, 13, 1)\n",
      "y_train shape: (3942,)\n",
      "X_test shape: (1233, 44, 13, 1)\n",
      "y_test shape: (1233,)\n",
      "X_val shape: (986, 44, 13, 1)\n",
      "y_val shape: (986,)\n"
     ]
    }
   ],
   "source": [
    "# Check the array shapes for the Ultrasuite dataset\n",
    "check_array_shapes(us_X_train, \n",
    "                   us_y_train, \n",
    "                   us_X_val, \n",
    "                   us_y_val, \n",
    "                   us_X_test, \n",
    "                   us_y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model 1: Create a simple baseline model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_baseline_model(input_shape, \n",
    "                         output_units, \n",
    "                         loss_func='categorical_crossentropy', \n",
    "                         learning_rate=0.0001):\n",
    "    '''\n",
    "    Build a baseline model\n",
    "        \n",
    "        Params:\n",
    "            input_shape (tuple): Shape of array representing a sample train\n",
    "            output_units (int): Number of targets / categories\n",
    "            loss_func (str): Loss function to use\n",
    "            learning_rate (float): Learning rate\n",
    "        \n",
    "        Returns:\n",
    "            baseline_model: Tensorflow model\n",
    "    '''     \n",
    "    baseline_model = tf.keras.models.Sequential()\n",
    "    baseline_model.add(tf.keras.layers.InputLayer(input_shape=input_shape))\n",
    "    baseline_model.add(tf.keras.layers.Flatten())\n",
    "    baseline_model.add(tf.keras.layers.BatchNormalization())\n",
    "    baseline_model.add(tf.keras.layers.Dense(output_units, activation='softmax'))\n",
    "    \n",
    "    # Set optimizer and learning rate\n",
    "    optimiser = tf.optimizers.Adam(learning_rate=learning_rate)\n",
    "    \n",
    "    # Compile the baseline model\n",
    "    baseline_model.compile(loss=loss_func,\n",
    "                           optimizer=optimiser,\n",
    "                           metrics=['acc'])\n",
    "\n",
    "    # Print summary for model\n",
    "    baseline_model.summary()\n",
    "    \n",
    "    return baseline_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function for fitting the model\n",
    "def fit_model(model, \n",
    "              epochs, \n",
    "              batch_size, \n",
    "              patience, \n",
    "              X_train, \n",
    "              y_train, \n",
    "              X_val, \n",
    "              y_val):\n",
    "    '''\n",
    "    Fit the model\n",
    "        \n",
    "        Params:\n",
    "            model : Input Tensorflow model\n",
    "            epochs (int): Number of training epochs\n",
    "            batch_size (int): Number of samples per batch\n",
    "            patience (int): Number of epochs to wait before early stop, \n",
    "                            if there no improvement on accuracy\n",
    "            X_train (ndarray): Inputs for the training dataset\n",
    "            y_train (ndarray): Targets for the training dataset\n",
    "            X_val (ndarray): Inputs for the validation dataset\n",
    "            y_val (ndarray): Targets for the training dataset\n",
    "        \n",
    "        Returns:\n",
    "            results: Training history\n",
    "    '''    \n",
    "    # Define early stopping criteria\n",
    "    early_stopping = tf.keras.callbacks.EarlyStopping(monitor='accuracy', \n",
    "                                                      min_delta=0.001, \n",
    "                                                      patience=patience)\n",
    "\n",
    "    # Fit the model\n",
    "    results = model.fit(X_train,\n",
    "                        y_train,\n",
    "                        epochs=epochs,\n",
    "                        batch_size=batch_size,\n",
    "                        validation_data=(X_val, y_val),\n",
    "                        callbacks=[early_stopping])\n",
    "    return results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baseline model for the Speech Commands dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten (Flatten)            (None, 572)               0         \n",
      "_________________________________________________________________\n",
      "batch_normalization (BatchNo (None, 572)               2288      \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 35)                20055     \n",
      "=================================================================\n",
      "Total params: 22,343\n",
      "Trainable params: 21,199\n",
      "Non-trainable params: 1,144\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc6d73bd40> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "1884/1908 [============================>.] - ETA: 0s - loss: 3.0228 - acc: 0.1989WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc5599b710> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 3.0156 - acc: 0.2003 - val_loss: 2.4874 - val_acc: 0.3122\n",
      "Epoch 2/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.3892 - acc: 0.3389 - val_loss: 2.2717 - val_acc: 0.3737\n",
      "Epoch 3/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.2549 - acc: 0.3775 - val_loss: 2.1956 - val_acc: 0.3955\n",
      "Epoch 4/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.1982 - acc: 0.3940 - val_loss: 2.1577 - val_acc: 0.4074\n",
      "Epoch 5/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.1640 - acc: 0.4051 - val_loss: 2.1347 - val_acc: 0.4107\n",
      "Epoch 6/40\n",
      "1908/1908 [==============================] - 5s 2ms/step - loss: 2.1457 - acc: 0.4112 - val_loss: 2.1224 - val_acc: 0.4165\n",
      "Epoch 7/40\n",
      "1908/1908 [==============================] - 5s 2ms/step - loss: 2.1294 - acc: 0.4178 - val_loss: 2.1136 - val_acc: 0.4181\n",
      "Epoch 8/40\n",
      "1908/1908 [==============================] - 5s 2ms/step - loss: 2.1162 - acc: 0.4216 - val_loss: 2.1055 - val_acc: 0.4213\n",
      "Epoch 9/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.1103 - acc: 0.4227 - val_loss: 2.1000 - val_acc: 0.4249\n",
      "Epoch 10/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.1029 - acc: 0.4248 - val_loss: 2.0958 - val_acc: 0.4248\n",
      "Epoch 11/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0941 - acc: 0.4270 - val_loss: 2.0924 - val_acc: 0.4273\n",
      "Epoch 12/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0915 - acc: 0.4278 - val_loss: 2.0919 - val_acc: 0.4268\n",
      "Epoch 13/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0879 - acc: 0.4304 - val_loss: 2.0889 - val_acc: 0.4298\n",
      "Epoch 14/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0833 - acc: 0.4317 - val_loss: 2.0877 - val_acc: 0.4292\n",
      "Epoch 15/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0790 - acc: 0.4303 - val_loss: 2.0874 - val_acc: 0.4283\n",
      "Epoch 16/40\n",
      "1908/1908 [==============================] - 8s 4ms/step - loss: 2.0756 - acc: 0.4302 - val_loss: 2.0858 - val_acc: 0.4291\n",
      "Epoch 17/40\n",
      "1908/1908 [==============================] - 8s 4ms/step - loss: 2.0738 - acc: 0.4333 - val_loss: 2.0858 - val_acc: 0.4295\n",
      "Epoch 18/40\n",
      "1908/1908 [==============================] - 7s 4ms/step - loss: 2.0734 - acc: 0.4334 - val_loss: 2.0847 - val_acc: 0.4310\n",
      "Epoch 19/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0651 - acc: 0.4359 - val_loss: 2.0816 - val_acc: 0.4313\n",
      "Epoch 20/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0668 - acc: 0.4365 - val_loss: 2.0849 - val_acc: 0.4312\n",
      "Epoch 21/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0663 - acc: 0.4373 - val_loss: 2.0836 - val_acc: 0.4289\n",
      "Epoch 22/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0614 - acc: 0.4385 - val_loss: 2.0837 - val_acc: 0.4284\n",
      "Epoch 23/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0599 - acc: 0.4374 - val_loss: 2.0803 - val_acc: 0.4321\n",
      "Epoch 24/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0591 - acc: 0.4399 - val_loss: 2.0816 - val_acc: 0.4297\n",
      "Epoch 25/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0589 - acc: 0.4379 - val_loss: 2.0830 - val_acc: 0.4284\n",
      "Epoch 26/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0573 - acc: 0.4379 - val_loss: 2.0820 - val_acc: 0.4317\n",
      "Epoch 27/40\n",
      "1908/1908 [==============================] - 4s 2ms/step - loss: 2.0570 - acc: 0.4389 - val_loss: 2.0817 - val_acc: 0.4331\n",
      "Epoch 28/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0542 - acc: 0.4389 - val_loss: 2.0832 - val_acc: 0.4323\n",
      "Epoch 29/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.0528 - acc: 0.4386 - val_loss: 2.0824 - val_acc: 0.4318\n",
      "Epoch 30/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.0518 - acc: 0.4409 - val_loss: 2.0826 - val_acc: 0.4320\n",
      "Epoch 31/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0468 - acc: 0.4410 - val_loss: 2.0800 - val_acc: 0.4312\n",
      "Epoch 32/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0471 - acc: 0.4424 - val_loss: 2.0802 - val_acc: 0.4333\n",
      "Epoch 33/40\n",
      "1908/1908 [==============================] - 6s 3ms/step - loss: 2.0481 - acc: 0.4413 - val_loss: 2.0820 - val_acc: 0.4312\n",
      "Epoch 34/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0480 - acc: 0.4419 - val_loss: 2.0800 - val_acc: 0.4329\n",
      "Epoch 35/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0459 - acc: 0.4434 - val_loss: 2.0808 - val_acc: 0.4312\n",
      "Epoch 36/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0464 - acc: 0.4397 - val_loss: 2.0807 - val_acc: 0.4309\n",
      "Epoch 37/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0406 - acc: 0.4457 - val_loss: 2.0802 - val_acc: 0.4325\n",
      "Epoch 38/40\n",
      "1908/1908 [==============================] - 8s 4ms/step - loss: 2.0462 - acc: 0.4429 - val_loss: 2.0830 - val_acc: 0.4303\n",
      "Epoch 39/40\n",
      "1908/1908 [==============================] - 9s 5ms/step - loss: 2.0429 - acc: 0.4423 - val_loss: 2.0818 - val_acc: 0.4310\n",
      "Epoch 40/40\n",
      "1908/1908 [==============================] - 5s 3ms/step - loss: 2.0412 - acc: 0.4422 - val_loss: 2.0844 - val_acc: 0.4306\n"
     ]
    }
   ],
   "source": [
    "# One-hot encode Speech Commands labels\n",
    "sc_train_y = reformat_y(sc_y_train)\n",
    "sc_test_y = reformat_y(sc_y_test)\n",
    "sc_val_y = reformat_y(sc_y_val)\n",
    "\n",
    "# Create baseline model for Speech Commands dataset\n",
    "sc_input_shape = (sc_X_train[0].shape)\n",
    "sc_output_units = 35\n",
    "sc_baseline_model = build_baseline_model(sc_input_shape, \n",
    "                                         sc_output_units, \n",
    "                                         learning_rate=0.0001)\n",
    "\n",
    "# Fit model\n",
    "sc_epochs = 40\n",
    "sc_batch_size = 32\n",
    "sc_patience = 5\n",
    "sc_baseline_results = fit_model(sc_baseline_model, \n",
    "                      sc_epochs, \n",
    "                      sc_batch_size, \n",
    "                      sc_patience, \n",
    "                      sc_X_train, \n",
    "                      sc_train_y, \n",
    "                      sc_X_val,\n",
    "                      sc_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(sc_baseline_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[233  25   6 ...   4   6  10]\n",
      " [ 12 357   1 ...   0   5   2]\n",
      " [ 11  10 124 ...   0  24   3]\n",
      " ...\n",
      " [  6   1   1 ...  94   2  59]\n",
      " [  5   0   3 ...   4 365  18]\n",
      " [  4   1   3 ...  29  20 313]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 1.947283148765564\n",
      "Accuracy: 47.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 2.0892109870910645\n",
      "Accuracy: 43.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(sc_baseline_model, \n",
    "                  sc_X_train, \n",
    "                  sc_train_y,\n",
    "                  sc_X_test, \n",
    "                  sc_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baseline model for the Ultrasuite dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten_1 (Flatten)          (None, 572)               0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 572)               2288      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 35)                20055     \n",
      "=================================================================\n",
      "Total params: 22,343\n",
      "Trainable params: 21,199\n",
      "Non-trainable params: 1,144\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc57590680> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "116/124 [===========================>..] - ETA: 0s - loss: 3.9106 - acc: 0.0550WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc55f8b3b0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "124/124 [==============================] - 1s 7ms/step - loss: 3.8920 - acc: 0.0568 - val_loss: 3.6495 - val_acc: 0.1166\n",
      "Epoch 2/40\n",
      "124/124 [==============================] - 1s 4ms/step - loss: 3.2134 - acc: 0.1502 - val_loss: 2.9252 - val_acc: 0.2231\n",
      "Epoch 3/40\n",
      "124/124 [==============================] - 1s 7ms/step - loss: 2.7694 - acc: 0.2466 - val_loss: 2.5659 - val_acc: 0.3215\n",
      "Epoch 4/40\n",
      "124/124 [==============================] - 1s 5ms/step - loss: 2.4499 - acc: 0.3409 - val_loss: 2.3223 - val_acc: 0.3925\n",
      "Epoch 5/40\n",
      "124/124 [==============================] - 1s 6ms/step - loss: 2.2293 - acc: 0.4087 - val_loss: 2.1367 - val_acc: 0.4533\n",
      "Epoch 6/40\n",
      "124/124 [==============================] - 1s 4ms/step - loss: 2.0528 - acc: 0.4612 - val_loss: 1.9904 - val_acc: 0.4888\n",
      "Epoch 7/40\n",
      "124/124 [==============================] - 1s 5ms/step - loss: 1.9030 - acc: 0.5084 - val_loss: 1.8734 - val_acc: 0.5172\n",
      "Epoch 8/40\n",
      "124/124 [==============================] - 1s 6ms/step - loss: 1.7761 - acc: 0.5492 - val_loss: 1.7807 - val_acc: 0.5436\n",
      "Epoch 9/40\n",
      "124/124 [==============================] - 1s 6ms/step - loss: 1.6846 - acc: 0.5667 - val_loss: 1.7081 - val_acc: 0.5669\n",
      "Epoch 10/40\n",
      "124/124 [==============================] - 1s 4ms/step - loss: 1.5966 - acc: 0.5916 - val_loss: 1.6438 - val_acc: 0.5842\n",
      "Epoch 11/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.5286 - acc: 0.6126 - val_loss: 1.5877 - val_acc: 0.6034\n",
      "Epoch 12/40\n",
      "124/124 [==============================] - 1s 5ms/step - loss: 1.4553 - acc: 0.6347 - val_loss: 1.5401 - val_acc: 0.6136\n",
      "Epoch 13/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.4132 - acc: 0.6461 - val_loss: 1.5008 - val_acc: 0.6217\n",
      "Epoch 14/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 1.3713 - acc: 0.6540 - val_loss: 1.4676 - val_acc: 0.6308\n",
      "Epoch 15/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 1.3173 - acc: 0.6730 - val_loss: 1.4343 - val_acc: 0.6450\n",
      "Epoch 16/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.2850 - acc: 0.6824 - val_loss: 1.4051 - val_acc: 0.6531\n",
      "Epoch 17/40\n",
      "124/124 [==============================] - 1s 4ms/step - loss: 1.2502 - acc: 0.6842 - val_loss: 1.3828 - val_acc: 0.6613\n",
      "Epoch 18/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.2177 - acc: 0.6892 - val_loss: 1.3603 - val_acc: 0.6633\n",
      "Epoch 19/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.1962 - acc: 0.6986 - val_loss: 1.3419 - val_acc: 0.6643\n",
      "Epoch 20/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 1.1603 - acc: 0.7121 - val_loss: 1.3245 - val_acc: 0.6704\n",
      "Epoch 21/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.1342 - acc: 0.7184 - val_loss: 1.3102 - val_acc: 0.6714\n",
      "Epoch 22/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.1176 - acc: 0.7215 - val_loss: 1.2954 - val_acc: 0.6775\n",
      "Epoch 23/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.0836 - acc: 0.7326 - val_loss: 1.2811 - val_acc: 0.6775\n",
      "Epoch 24/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.0776 - acc: 0.7258 - val_loss: 1.2710 - val_acc: 0.6785\n",
      "Epoch 25/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.0568 - acc: 0.7369 - val_loss: 1.2587 - val_acc: 0.6815\n",
      "Epoch 26/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 1.0323 - acc: 0.7407 - val_loss: 1.2502 - val_acc: 0.6826\n",
      "Epoch 27/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 1.0221 - acc: 0.7428 - val_loss: 1.2427 - val_acc: 0.6826\n",
      "Epoch 28/40\n",
      "124/124 [==============================] - 0s 2ms/step - loss: 1.0037 - acc: 0.7557 - val_loss: 1.2370 - val_acc: 0.6897\n",
      "Epoch 29/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 0.9924 - acc: 0.7539 - val_loss: 1.2281 - val_acc: 0.6907\n",
      "Epoch 30/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 0.9706 - acc: 0.7638 - val_loss: 1.2201 - val_acc: 0.6937\n",
      "Epoch 31/40\n",
      "124/124 [==============================] - 1s 4ms/step - loss: 0.9653 - acc: 0.7593 - val_loss: 1.2168 - val_acc: 0.6907\n",
      "Epoch 32/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.9510 - acc: 0.7641 - val_loss: 1.2117 - val_acc: 0.6927\n",
      "Epoch 33/40\n",
      "124/124 [==============================] - 0s 4ms/step - loss: 0.9323 - acc: 0.7661 - val_loss: 1.2064 - val_acc: 0.6968\n",
      "Epoch 34/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.9226 - acc: 0.7686 - val_loss: 1.1977 - val_acc: 0.6988\n",
      "Epoch 35/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.9120 - acc: 0.7768 - val_loss: 1.1941 - val_acc: 0.6988\n",
      "Epoch 36/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.9096 - acc: 0.7747 - val_loss: 1.1908 - val_acc: 0.7018\n",
      "Epoch 37/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.8928 - acc: 0.7780 - val_loss: 1.1880 - val_acc: 0.7039\n",
      "Epoch 38/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.8849 - acc: 0.7854 - val_loss: 1.1835 - val_acc: 0.7059\n",
      "Epoch 39/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.8743 - acc: 0.7841 - val_loss: 1.1799 - val_acc: 0.7028\n",
      "Epoch 40/40\n",
      "124/124 [==============================] - 0s 3ms/step - loss: 0.8658 - acc: 0.7859 - val_loss: 1.1793 - val_acc: 0.7018\n"
     ]
    }
   ],
   "source": [
    "# One-hot encode Ultrasuite Top 35 labels\n",
    "us_train_y = reformat_y(us_y_train)\n",
    "us_test_y = reformat_y(us_y_test)\n",
    "us_val_y = reformat_y(us_y_val)\n",
    "\n",
    "# Create baseline model for Ultrasuite dataset\n",
    "us_input_shape = (us_X_train[0].shape)\n",
    "us_output_units = 35\n",
    "us_baseline_model = build_baseline_model(us_input_shape, \n",
    "                                         us_output_units, \n",
    "                                         learning_rate=0.0001)\n",
    "\n",
    "# Fit model\n",
    "us_epochs = 40\n",
    "us_batch_size = 32\n",
    "us_patience = 5\n",
    "us_baseline_results = fit_model(us_baseline_model, \n",
    "                      us_epochs, \n",
    "                      us_batch_size, \n",
    "                      us_patience, \n",
    "                      us_X_train, \n",
    "                      us_train_y, \n",
    "                      us_X_val, \n",
    "                      us_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(us_baseline_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[15  0  0 ...  0  0  0]\n",
      " [ 0 32  0 ...  1  2  0]\n",
      " [ 1  1 24 ...  0  0  0]\n",
      " ...\n",
      " [ 0  1  0 ... 13  1  2]\n",
      " [ 0  1  0 ...  0 40  0]\n",
      " [ 0  0  0 ...  0  0 12]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 0.7894143462181091\n",
      "Accuracy: 81.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 1.254226803779602\n",
      "Accuracy: 70.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(us_baseline_model, \n",
    "                  us_X_train, \n",
    "                  us_train_y,\n",
    "                  us_X_test, \n",
    "                  us_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It's starting point with the training accuracy for the Ultrasuite dataset being 81%. The model is clearly overfitting the training data and is not generalising well when shown unseen data given the validation accuracy of 70%.\n",
    "\n",
    "What is of particular note is the vastly different accuracies when compared to the Speech Commands dataset which are both under 50%.  \n",
    "\n",
    "This could suggest that a more simple model works better for audio samples from children with a speech sound disorder."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model 2: Baseline model with increased learning rate and batch size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten_2 (Flatten)          (None, 572)               0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 572)               2288      \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 35)                20055     \n",
      "=================================================================\n",
      "Total params: 22,343\n",
      "Trainable params: 21,199\n",
      "Non-trainable params: 1,144\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc592b58c0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "62/62 [==============================] - ETA: 0s - loss: 2.7641 - acc: 0.2808WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc58f88f80> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "62/62 [==============================] - 1s 6ms/step - loss: 2.7641 - acc: 0.2808 - val_loss: 2.1657 - val_acc: 0.4249\n",
      "Epoch 2/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 1.5795 - acc: 0.5893 - val_loss: 1.5551 - val_acc: 0.5872\n",
      "Epoch 3/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 1.2494 - acc: 0.6877 - val_loss: 1.3764 - val_acc: 0.6379\n",
      "Epoch 4/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 1.1008 - acc: 0.7275 - val_loss: 1.2806 - val_acc: 0.6613\n",
      "Epoch 5/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.9909 - acc: 0.7529 - val_loss: 1.2235 - val_acc: 0.6856\n",
      "Epoch 6/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.9158 - acc: 0.7768 - val_loss: 1.2088 - val_acc: 0.6815\n",
      "Epoch 7/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.8605 - acc: 0.7851 - val_loss: 1.1870 - val_acc: 0.6927\n",
      "Epoch 8/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.8025 - acc: 0.8009 - val_loss: 1.1821 - val_acc: 0.6876\n",
      "Epoch 9/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.7830 - acc: 0.8044 - val_loss: 1.1723 - val_acc: 0.6978\n",
      "Epoch 10/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.7397 - acc: 0.8176 - val_loss: 1.1765 - val_acc: 0.6957\n",
      "Epoch 11/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.7050 - acc: 0.8267 - val_loss: 1.1781 - val_acc: 0.6927\n",
      "Epoch 12/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.6858 - acc: 0.8369 - val_loss: 1.1806 - val_acc: 0.7018\n",
      "Epoch 13/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.6604 - acc: 0.8453 - val_loss: 1.1824 - val_acc: 0.7018\n",
      "Epoch 14/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.6372 - acc: 0.8450 - val_loss: 1.1896 - val_acc: 0.6957\n",
      "Epoch 15/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.6124 - acc: 0.8544 - val_loss: 1.1892 - val_acc: 0.6947\n",
      "Epoch 16/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.5990 - acc: 0.8531 - val_loss: 1.2017 - val_acc: 0.6907\n",
      "Epoch 17/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.5763 - acc: 0.8612 - val_loss: 1.2190 - val_acc: 0.6957\n",
      "Epoch 18/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.5740 - acc: 0.8643 - val_loss: 1.2118 - val_acc: 0.6907\n",
      "Epoch 19/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.5543 - acc: 0.8671 - val_loss: 1.2177 - val_acc: 0.6947\n",
      "Epoch 20/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.5408 - acc: 0.8696 - val_loss: 1.2310 - val_acc: 0.6856\n",
      "Epoch 21/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.5218 - acc: 0.8742 - val_loss: 1.2423 - val_acc: 0.6876\n",
      "Epoch 22/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.5104 - acc: 0.8833 - val_loss: 1.2609 - val_acc: 0.6897\n",
      "Epoch 23/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.4964 - acc: 0.8823 - val_loss: 1.2605 - val_acc: 0.6957\n",
      "Epoch 24/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.4883 - acc: 0.8884 - val_loss: 1.2788 - val_acc: 0.6866\n",
      "Epoch 25/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.4892 - acc: 0.8818 - val_loss: 1.2735 - val_acc: 0.6866\n",
      "Epoch 26/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.4697 - acc: 0.8836 - val_loss: 1.2920 - val_acc: 0.6836\n",
      "Epoch 27/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.4610 - acc: 0.8937 - val_loss: 1.3068 - val_acc: 0.6826\n",
      "Epoch 28/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.4517 - acc: 0.8932 - val_loss: 1.3016 - val_acc: 0.6856\n",
      "Epoch 29/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.4430 - acc: 0.8904 - val_loss: 1.3343 - val_acc: 0.6897\n",
      "Epoch 30/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.4351 - acc: 0.8978 - val_loss: 1.3254 - val_acc: 0.6795\n",
      "Epoch 31/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.4221 - acc: 0.9016 - val_loss: 1.3584 - val_acc: 0.6805\n",
      "Epoch 32/40\n",
      "62/62 [==============================] - 0s 3ms/step - loss: 0.4213 - acc: 0.9021 - val_loss: 1.3607 - val_acc: 0.6826\n",
      "Epoch 33/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.4121 - acc: 0.9069 - val_loss: 1.3782 - val_acc: 0.6755\n",
      "Epoch 34/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.4025 - acc: 0.9041 - val_loss: 1.3845 - val_acc: 0.6744\n",
      "Epoch 35/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.3882 - acc: 0.9110 - val_loss: 1.3855 - val_acc: 0.6694\n",
      "Epoch 36/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.3918 - acc: 0.9074 - val_loss: 1.4073 - val_acc: 0.6694\n",
      "Epoch 37/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.3834 - acc: 0.9102 - val_loss: 1.4053 - val_acc: 0.6765\n",
      "Epoch 38/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.3811 - acc: 0.9112 - val_loss: 1.4276 - val_acc: 0.6643\n",
      "Epoch 39/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.3732 - acc: 0.9117 - val_loss: 1.4245 - val_acc: 0.6673\n",
      "Epoch 40/40\n",
      "62/62 [==============================] - 0s 4ms/step - loss: 0.3664 - acc: 0.9130 - val_loss: 1.4384 - val_acc: 0.6744\n"
     ]
    }
   ],
   "source": [
    "# Create second baseline model on Ultrasuite dataset changing the learning rate and batch size\n",
    "us_input_shape = (us_X_train[0].shape)\n",
    "us_output_units = 35\n",
    "us_baseline_model_2 = build_baseline_model(us_input_shape, \n",
    "                                           us_output_units, \n",
    "                                           learning_rate=0.001)\n",
    "\n",
    " \n",
    "# Fit model\n",
    "us_epochs = 40\n",
    "us_batch_size = 64\n",
    "us_patience = 5\n",
    "us_baseline_results_2 = fit_model(us_baseline_model_2, \n",
    "                      us_epochs, \n",
    "                      us_batch_size, \n",
    "                      us_patience, \n",
    "                      us_X_train, \n",
    "                      us_train_y, \n",
    "                      us_X_val, \n",
    "                      us_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(us_baseline_results_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[14  0  0 ...  0  0  1]\n",
      " [ 0 28  0 ...  3  1  0]\n",
      " [ 1  0 28 ...  0  0  0]\n",
      " ...\n",
      " [ 1  1  0 ... 13  0  1]\n",
      " [ 1  0  0 ...  0 43  0]\n",
      " [ 0  0  0 ...  0  0 18]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 0.28699809312820435\n",
      "Accuracy: 94.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 1.5567854642868042\n",
      "Accuracy: 65.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(us_baseline_model_2, \n",
    "                  us_X_train, \n",
    "                  us_train_y,\n",
    "                  us_X_test, \n",
    "                  us_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have improved the training accuracy of the baseline model, to 94%, by increasing the learning rate and increasing the batch size but this has only served to exascerbate the issue of overfitting with test accuracy not reducing to 65% and the difference between the two plots being dramatic.\n",
    "\n",
    "In an effort to deal with the issue of overfitting, we will introduce some regularization layers and see what effect they have on performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model 3: Adding hidden layers to the baseline model, deepening the network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build a second baseline model\n",
    "def build_model_2(input_shape, \n",
    "                  output_units, \n",
    "                  loss_func='categorical_crossentropy', \n",
    "                  learning_rate=0.0001):\n",
    "    \n",
    "    model_2 = tf.keras.models.Sequential()\n",
    "    model_2.add(tf.keras.layers.InputLayer(input_shape=input_shape))\n",
    "    model_2.add(tf.keras.layers.Flatten())\n",
    "    model_2.add(tf.keras.layers.BatchNormalization())\n",
    "    model_2.add(tf.keras.layers.Dense(256, activation='relu'))\n",
    "#     model_2.add(tf.keras.layers.Dropout(0.3))\n",
    "    model_2.add(tf.keras.layers.Dense(128, activation='relu'))\n",
    "    model_2.add(tf.keras.layers.Dense(64, activation='relu'))\n",
    "    model_2.add(tf.keras.layers.Dense(output_units, activation='softmax'))\n",
    "    \n",
    "    # Set optimizer and learning rate\n",
    "    optimiser = tf.optimizers.Adam(learning_rate=learning_rate)\n",
    "    \n",
    "    # Compile the baseline model\n",
    "    model_2.compile(loss=loss_func,\n",
    "                           optimizer=optimiser,\n",
    "                           metrics=['acc'])\n",
    "\n",
    "    # Print summary for model\n",
    "    model_2.summary()\n",
    "    \n",
    "    return model_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_3\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten_3 (Flatten)          (None, 572)               0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 572)               2288      \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 256)               146688    \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 128)               32896     \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 64)                8256      \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 35)                2275      \n",
      "=================================================================\n",
      "Total params: 192,403\n",
      "Trainable params: 191,259\n",
      "Non-trainable params: 1,144\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc5968a830> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "55/62 [=========================>....] - ETA: 0s - loss: 3.5501 - acc: 0.0506WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc62d970e0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "62/62 [==============================] - 1s 14ms/step - loss: 3.5272 - acc: 0.0563 - val_loss: 3.4481 - val_acc: 0.0862\n",
      "Epoch 2/40\n",
      "62/62 [==============================] - 0s 8ms/step - loss: 3.1335 - acc: 0.1933 - val_loss: 2.9474 - val_acc: 0.2191\n",
      "Epoch 3/40\n",
      "62/62 [==============================] - 1s 9ms/step - loss: 2.7496 - acc: 0.3323 - val_loss: 2.5577 - val_acc: 0.3722\n",
      "Epoch 4/40\n",
      "62/62 [==============================] - 1s 9ms/step - loss: 2.3500 - acc: 0.4401 - val_loss: 2.2210 - val_acc: 0.4635\n",
      "Epoch 5/40\n",
      "62/62 [==============================] - 0s 8ms/step - loss: 1.9996 - acc: 0.5155 - val_loss: 1.9466 - val_acc: 0.5254\n",
      "Epoch 6/40\n",
      "62/62 [==============================] - 1s 9ms/step - loss: 1.7161 - acc: 0.5832 - val_loss: 1.7296 - val_acc: 0.5751\n",
      "Epoch 7/40\n",
      "62/62 [==============================] - 1s 9ms/step - loss: 1.5033 - acc: 0.6291 - val_loss: 1.5744 - val_acc: 0.6095\n",
      "Epoch 8/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 1.3268 - acc: 0.6804 - val_loss: 1.4615 - val_acc: 0.6430\n",
      "Epoch 9/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 1.2031 - acc: 0.7067 - val_loss: 1.3774 - val_acc: 0.6582\n",
      "Epoch 10/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 1.0949 - acc: 0.7301 - val_loss: 1.3159 - val_acc: 0.6724\n",
      "Epoch 11/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 1.0011 - acc: 0.7600 - val_loss: 1.2649 - val_acc: 0.6876\n",
      "Epoch 12/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.9319 - acc: 0.7737 - val_loss: 1.2312 - val_acc: 0.6897\n",
      "Epoch 13/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.8720 - acc: 0.7897 - val_loss: 1.1951 - val_acc: 0.6998\n",
      "Epoch 14/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.8090 - acc: 0.8021 - val_loss: 1.1768 - val_acc: 0.6968\n",
      "Epoch 15/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.7621 - acc: 0.8176 - val_loss: 1.1466 - val_acc: 0.7008\n",
      "Epoch 16/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.7159 - acc: 0.8267 - val_loss: 1.1314 - val_acc: 0.7170\n",
      "Epoch 17/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.6835 - acc: 0.8392 - val_loss: 1.1220 - val_acc: 0.7191\n",
      "Epoch 18/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.6520 - acc: 0.8425 - val_loss: 1.1017 - val_acc: 0.7191\n",
      "Epoch 19/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.6178 - acc: 0.8473 - val_loss: 1.0930 - val_acc: 0.7181\n",
      "Epoch 20/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.5786 - acc: 0.8633 - val_loss: 1.0786 - val_acc: 0.7282\n",
      "Epoch 21/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.5479 - acc: 0.8772 - val_loss: 1.0797 - val_acc: 0.7252\n",
      "Epoch 22/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.5273 - acc: 0.8767 - val_loss: 1.0654 - val_acc: 0.7373\n",
      "Epoch 23/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.4998 - acc: 0.8871 - val_loss: 1.0601 - val_acc: 0.7373\n",
      "Epoch 24/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.4815 - acc: 0.8904 - val_loss: 1.0538 - val_acc: 0.7323\n",
      "Epoch 25/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.4612 - acc: 0.8965 - val_loss: 1.0582 - val_acc: 0.7333\n",
      "Epoch 26/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.4366 - acc: 0.9059 - val_loss: 1.0541 - val_acc: 0.7373\n",
      "Epoch 27/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.4152 - acc: 0.9112 - val_loss: 1.0557 - val_acc: 0.7353\n",
      "Epoch 28/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3986 - acc: 0.9145 - val_loss: 1.0484 - val_acc: 0.7373\n",
      "Epoch 29/40\n",
      "62/62 [==============================] - 0s 7ms/step - loss: 0.3868 - acc: 0.9137 - val_loss: 1.0531 - val_acc: 0.7383\n",
      "Epoch 30/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3667 - acc: 0.9242 - val_loss: 1.0463 - val_acc: 0.7373\n",
      "Epoch 31/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3532 - acc: 0.9224 - val_loss: 1.0558 - val_acc: 0.7353\n",
      "Epoch 32/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3406 - acc: 0.9242 - val_loss: 1.0478 - val_acc: 0.7383\n",
      "Epoch 33/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3217 - acc: 0.9328 - val_loss: 1.0502 - val_acc: 0.7475\n",
      "Epoch 34/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.3070 - acc: 0.9406 - val_loss: 1.0470 - val_acc: 0.7424\n",
      "Epoch 35/40\n",
      "62/62 [==============================] - 0s 5ms/step - loss: 0.2924 - acc: 0.9406 - val_loss: 1.0421 - val_acc: 0.7444\n",
      "Epoch 36/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.2850 - acc: 0.9427 - val_loss: 1.0497 - val_acc: 0.7414\n",
      "Epoch 37/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.2748 - acc: 0.9439 - val_loss: 1.0438 - val_acc: 0.7404\n",
      "Epoch 38/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.2672 - acc: 0.9460 - val_loss: 1.0497 - val_acc: 0.7383\n",
      "Epoch 39/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.2491 - acc: 0.9531 - val_loss: 1.0559 - val_acc: 0.7383\n",
      "Epoch 40/40\n",
      "62/62 [==============================] - 0s 6ms/step - loss: 0.2477 - acc: 0.9503 - val_loss: 1.0460 - val_acc: 0.7404\n"
     ]
    }
   ],
   "source": [
    "# Create baseline model for Ultrasuite dataset\n",
    "us_input_shape = (us_X_train[0].shape)\n",
    "us_output_units = 35\n",
    "us_model_2 = build_model_2(us_input_shape, \n",
    "                           us_output_units, \n",
    "                           learning_rate=0.0001)\n",
    "\n",
    "# Fit model\n",
    "us_epochs = 40\n",
    "us_batch_size = 64\n",
    "us_patience = 3\n",
    "us_results_3 = fit_model(us_model_2, \n",
    "                      us_epochs, \n",
    "                      us_batch_size, \n",
    "                      us_patience, \n",
    "                      us_X_train, \n",
    "                      us_train_y, \n",
    "                      us_X_val, \n",
    "                      us_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(us_results_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[19  0  0 ...  1  1  0]\n",
      " [ 0 34  0 ...  0  2  0]\n",
      " [ 0  0 26 ...  0  0  0]\n",
      " ...\n",
      " [ 1  0  0 ... 16  0  1]\n",
      " [ 0  1  0 ...  0 43  0]\n",
      " [ 0  0  0 ...  1  0 18]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 0.19082011282444\n",
      "Accuracy: 96.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 1.085261344909668\n",
      "Accuracy: 72.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(us_model_2, \n",
    "                  us_X_train, \n",
    "                  us_train_y,\n",
    "                  us_X_test, \n",
    "                  us_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Running the model using Mel Spectrograms instead of MFCCs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A thought exercise more than anything else, just to see if there are any performance gains that can be made by using Mel Spectrograms instead of MFCCs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Datasets loaded...\n"
     ]
    }
   ],
   "source": [
    "# See shared_functions/preprocessing.py for create_train_test function\n",
    "# Create the train, test and val datasets for the Ultrasuite top 35 subset using Mel Spectrograms\n",
    "us_data_path = 'ultrasuite_top35_data_melspec.json'\n",
    "(\n",
    "    X_train_mel, \n",
    "    y_train_mel, \n",
    "    X_val_mel, \n",
    "    y_val_mel, \n",
    "    X_test_mel, \n",
    "    y_test_mel,\n",
    ") = preprocess.create_train_test(us_data_path, 'mel_specs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "flatten_4 (Flatten)          (None, 5632)              0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_4 (Batch (None, 5632)              22528     \n",
      "_________________________________________________________________\n",
      "dense_7 (Dense)              (None, 256)               1442048   \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 128)               32896     \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 64)                8256      \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 35)                2275      \n",
      "=================================================================\n",
      "Total params: 1,508,003\n",
      "Trainable params: 1,496,739\n",
      "Non-trainable params: 11,264\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc5968a950> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "60/62 [============================>.] - ETA: 0s - loss: 3.4671 - acc: 0.0865WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc57518f80> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "62/62 [==============================] - 2s 28ms/step - loss: 3.4620 - acc: 0.0880 - val_loss: 3.7351 - val_acc: 0.0963\n",
      "Epoch 2/40\n",
      "62/62 [==============================] - 2s 25ms/step - loss: 3.1100 - acc: 0.2012 - val_loss: 3.6237 - val_acc: 0.1623\n",
      "Epoch 3/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 2.7843 - acc: 0.3080 - val_loss: 3.5139 - val_acc: 0.2160\n",
      "Epoch 4/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 2.4528 - acc: 0.3917 - val_loss: 3.3831 - val_acc: 0.2556\n",
      "Epoch 5/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 2.1439 - acc: 0.4670 - val_loss: 3.2965 - val_acc: 0.2789\n",
      "Epoch 6/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.9055 - acc: 0.5264 - val_loss: 3.2257 - val_acc: 0.2708\n",
      "Epoch 7/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 1.7198 - acc: 0.5680 - val_loss: 3.1958 - val_acc: 0.2698\n",
      "Epoch 8/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.5940 - acc: 0.6010 - val_loss: 3.1567 - val_acc: 0.2769\n",
      "Epoch 9/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 1.4724 - acc: 0.6279 - val_loss: 3.1539 - val_acc: 0.3124\n",
      "Epoch 10/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.3905 - acc: 0.6350 - val_loss: 3.1377 - val_acc: 0.3306\n",
      "Epoch 11/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.2777 - acc: 0.6745 - val_loss: 3.1758 - val_acc: 0.3164\n",
      "Epoch 12/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.2230 - acc: 0.6842 - val_loss: 3.2217 - val_acc: 0.3083\n",
      "Epoch 13/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 1.1639 - acc: 0.7032 - val_loss: 3.2602 - val_acc: 0.3022\n",
      "Epoch 14/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 1.1209 - acc: 0.7123 - val_loss: 3.2949 - val_acc: 0.2809\n",
      "Epoch 15/40\n",
      "62/62 [==============================] - 1s 24ms/step - loss: 1.0745 - acc: 0.7291 - val_loss: 3.3318 - val_acc: 0.2982\n",
      "Epoch 16/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 1.0090 - acc: 0.7466 - val_loss: 3.3124 - val_acc: 0.3002\n",
      "Epoch 17/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.9990 - acc: 0.7440 - val_loss: 3.3178 - val_acc: 0.3002\n",
      "Epoch 18/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.9617 - acc: 0.7501 - val_loss: 3.2790 - val_acc: 0.3053\n",
      "Epoch 19/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.9240 - acc: 0.7643 - val_loss: 3.3495 - val_acc: 0.2931\n",
      "Epoch 20/40\n",
      "62/62 [==============================] - 1s 24ms/step - loss: 0.8890 - acc: 0.7770 - val_loss: 3.3075 - val_acc: 0.2941\n",
      "Epoch 21/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 0.8662 - acc: 0.7796 - val_loss: 3.3758 - val_acc: 0.2850\n",
      "Epoch 22/40\n",
      "62/62 [==============================] - 1s 24ms/step - loss: 0.8357 - acc: 0.7905 - val_loss: 3.3370 - val_acc: 0.3124\n",
      "Epoch 23/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.8373 - acc: 0.7905 - val_loss: 3.3239 - val_acc: 0.3124\n",
      "Epoch 24/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 0.7985 - acc: 0.7991 - val_loss: 3.3491 - val_acc: 0.3215\n",
      "Epoch 25/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.7879 - acc: 0.7978 - val_loss: 3.3441 - val_acc: 0.3205\n",
      "Epoch 26/40\n",
      "62/62 [==============================] - 1s 20ms/step - loss: 0.7650 - acc: 0.8070 - val_loss: 3.4099 - val_acc: 0.3002\n",
      "Epoch 27/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.7573 - acc: 0.8110 - val_loss: 3.4577 - val_acc: 0.3215\n",
      "Epoch 28/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.7224 - acc: 0.8242 - val_loss: 3.4225 - val_acc: 0.3174\n",
      "Epoch 29/40\n",
      "62/62 [==============================] - 1s 23ms/step - loss: 0.7122 - acc: 0.8247 - val_loss: 3.4836 - val_acc: 0.2951\n",
      "Epoch 30/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.7024 - acc: 0.8265 - val_loss: 3.4519 - val_acc: 0.3114\n",
      "Epoch 31/40\n",
      "62/62 [==============================] - 1s 24ms/step - loss: 0.6780 - acc: 0.8318 - val_loss: 3.5031 - val_acc: 0.2982\n",
      "Epoch 32/40\n",
      "62/62 [==============================] - 1s 21ms/step - loss: 0.6721 - acc: 0.8336 - val_loss: 3.5957 - val_acc: 0.2972\n",
      "Epoch 33/40\n",
      "62/62 [==============================] - 1s 21ms/step - loss: 0.6723 - acc: 0.8311 - val_loss: 3.5796 - val_acc: 0.2992\n",
      "Epoch 34/40\n",
      "62/62 [==============================] - 1s 20ms/step - loss: 0.6445 - acc: 0.8369 - val_loss: 3.5385 - val_acc: 0.3083\n",
      "Epoch 35/40\n",
      "62/62 [==============================] - 1s 21ms/step - loss: 0.6333 - acc: 0.8412 - val_loss: 3.5832 - val_acc: 0.2911\n",
      "Epoch 36/40\n",
      "62/62 [==============================] - 1s 21ms/step - loss: 0.6290 - acc: 0.8397 - val_loss: 3.6739 - val_acc: 0.2870\n",
      "Epoch 37/40\n",
      "62/62 [==============================] - 2s 25ms/step - loss: 0.6315 - acc: 0.8432 - val_loss: 3.7004 - val_acc: 0.2779\n",
      "Epoch 38/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.6181 - acc: 0.8480 - val_loss: 3.7068 - val_acc: 0.2819\n",
      "Epoch 39/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.6002 - acc: 0.8488 - val_loss: 3.6283 - val_acc: 0.3124\n",
      "Epoch 40/40\n",
      "62/62 [==============================] - 1s 22ms/step - loss: 0.5956 - acc: 0.8529 - val_loss: 3.7373 - val_acc: 0.3043\n"
     ]
    }
   ],
   "source": [
    "# Create model using Mel Spectrograms instead of MFCCs\n",
    "us_mel_input_shape = (X_train_mel.shape[1], X_train_mel.shape[2], 1)\n",
    "us_mel_output_units = 35\n",
    "us_mel_model = build_model_2(us_mel_input_shape, \n",
    "                             us_mel_output_units, \n",
    "                             learning_rate=0.0001)\n",
    "\n",
    "# One-hot encode Speech Commands labels\n",
    "mel_train_y = reformat_y(y_train_mel)\n",
    "mel_test_y = reformat_y(y_test_mel)\n",
    "mel_val_y = reformat_y(y_val_mel)\n",
    "\n",
    "# Fit model\n",
    "us_mel_epochs = 40\n",
    "us_mel_batch_size = 64\n",
    "us_mel_patience = 3\n",
    "us_mel_results = fit_model(us_mel_model, \n",
    "                           us_mel_epochs, \n",
    "                           us_mel_batch_size, \n",
    "                           us_mel_patience, \n",
    "                           X_train_mel, \n",
    "                           mel_train_y, \n",
    "                           X_val_mel, \n",
    "                           mel_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(us_mel_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[ 6  0  0 ...  3  2  0]\n",
      " [ 1 18  0 ...  0  6  0]\n",
      " [ 0  1  3 ...  1  9  0]\n",
      " ...\n",
      " [ 0  0  0 ...  5  7  0]\n",
      " [ 0  3  0 ...  0 40  0]\n",
      " [ 3  1  0 ...  0  3  4]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 1.8458300828933716\n",
      "Accuracy: 53.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 3.5853798389434814\n",
      "Accuracy: 30.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(us_mel_model, \n",
    "                  X_train_mel, \n",
    "                  mel_train_y,\n",
    "                  X_test_mel, \n",
    "                  mel_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model 4: Convolutional Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build a CNN model\n",
    "\n",
    "def build_cnn_model(input_shape, \n",
    "                    output_units, \n",
    "                    loss='categorical_crossentropy', \n",
    "                    learning_rate=0.0001):\n",
    "\n",
    "    cnn_model = tf.keras.models.Sequential()\n",
    "\n",
    "    # 1st convolutional layer\n",
    "    cnn_model.add(tf.keras.layers.Conv2D(32, (3, 3), \n",
    "                                         activation='relu', \n",
    "                                         input_shape=input_shape,\n",
    "                                         kernel_regularizer=tf.keras.regularizers.l2(0.005)))\n",
    "    cnn_model.add(tf.keras.layers.BatchNormalization())\n",
    "    cnn_model.add(tf.keras.layers.MaxPooling2D((2, 2), \n",
    "                                               strides=(2,2), \n",
    "                                               padding='same'))\n",
    "\n",
    "    # 2nd convolutional layer\n",
    "    cnn_model.add(tf.keras.layers.Conv2D(32, (4, 4), \n",
    "                                     activation='relu',\n",
    "                                     kernel_regularizer=tf.keras.regularizers.l2(0.005)))\n",
    "    cnn_model.add(tf.keras.layers.BatchNormalization())\n",
    "    cnn_model.add(tf.keras.layers.MaxPooling2D((3, 3), \n",
    "                                               strides=(2,2), \n",
    "                                               padding='same'))\n",
    "\n",
    "    # 3rd convolutional layer\n",
    "    cnn_model.add(tf.keras.layers.Conv2D(64, (2, 2), \n",
    "                                         activation='relu',\n",
    "                                         kernel_regularizer=tf.keras.regularizers.l2(0.005)))\n",
    "    cnn_model.add(tf.keras.layers.BatchNormalization())\n",
    "    cnn_model.add(tf.keras.layers.MaxPooling2D((2, 2), \n",
    "                                               strides=(2,2), \n",
    "                                               padding='same'))\n",
    "\n",
    "    # Flatten output and feed into dense layer\n",
    "    cnn_model.add(tf.keras.layers.Flatten())\n",
    "    cnn_model.add(tf.keras.layers.Dense(32, activation='relu'))\n",
    "    cnn_model.add(tf.keras.layers.Dropout(0.3))\n",
    "\n",
    "    # Softmax output layer\n",
    "    cnn_model.add(tf.keras.layers.Dense(output_units, activation='softmax'))\n",
    "\n",
    "    optimiser = tf.optimizers.Adam(learning_rate=learning_rate)\n",
    "\n",
    "    # Compile model\n",
    "    cnn_model.compile(optimizer=optimiser,\n",
    "                  loss=loss,\n",
    "                  metrics=['acc'])\n",
    "\n",
    "    # Print summary for model\n",
    "    cnn_model.summary()\n",
    "\n",
    "    return cnn_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### CNN model for the Speech Commands dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_5\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 42, 11, 32)        320       \n",
      "_________________________________________________________________\n",
      "batch_normalization_5 (Batch (None, 42, 11, 32)        128       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 21, 6, 32)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 18, 3, 32)         16416     \n",
      "_________________________________________________________________\n",
      "batch_normalization_6 (Batch (None, 18, 3, 32)         128       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 9, 2, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 8, 1, 64)          8256      \n",
      "_________________________________________________________________\n",
      "batch_normalization_7 (Batch (None, 8, 1, 64)          256       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 4, 1, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_5 (Flatten)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_11 (Dense)             (None, 32)                8224      \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_12 (Dense)             (None, 35)                1155      \n",
      "=================================================================\n",
      "Total params: 34,883\n",
      "Trainable params: 34,627\n",
      "Non-trainable params: 256\n",
      "_________________________________________________________________\n",
      "Epoch 1/40\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc553c7c20> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "953/954 [============================>.] - ETA: 0s - loss: 3.7092 - acc: 0.1230WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc576b45f0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "954/954 [==============================] - 50s 52ms/step - loss: 3.7088 - acc: 0.1231 - val_loss: 3.1140 - val_acc: 0.2747\n",
      "Epoch 2/40\n",
      "954/954 [==============================] - 49s 52ms/step - loss: 2.9439 - acc: 0.2828 - val_loss: 2.3898 - val_acc: 0.4656\n",
      "Epoch 3/40\n",
      "954/954 [==============================] - 43s 46ms/step - loss: 2.3825 - acc: 0.4127 - val_loss: 1.8690 - val_acc: 0.5807\n",
      "Epoch 4/40\n",
      "954/954 [==============================] - 49s 52ms/step - loss: 2.0034 - acc: 0.4999 - val_loss: 1.5207 - val_acc: 0.6578\n",
      "Epoch 5/40\n",
      "954/954 [==============================] - 52s 54ms/step - loss: 1.7362 - acc: 0.5636 - val_loss: 1.3251 - val_acc: 0.7037\n",
      "Epoch 6/40\n",
      "954/954 [==============================] - 39s 41ms/step - loss: 1.5507 - acc: 0.6118 - val_loss: 1.1748 - val_acc: 0.7375\n",
      "Epoch 7/40\n",
      "954/954 [==============================] - 46s 48ms/step - loss: 1.3938 - acc: 0.6487 - val_loss: 1.0468 - val_acc: 0.7629\n",
      "Epoch 8/40\n",
      "954/954 [==============================] - 46s 48ms/step - loss: 1.2888 - acc: 0.6780 - val_loss: 0.9546 - val_acc: 0.7851\n",
      "Epoch 9/40\n",
      "954/954 [==============================] - 41s 43ms/step - loss: 1.2031 - acc: 0.7013 - val_loss: 0.9116 - val_acc: 0.7922\n",
      "Epoch 10/40\n",
      "954/954 [==============================] - 52s 54ms/step - loss: 1.1226 - acc: 0.7192 - val_loss: 0.8309 - val_acc: 0.8134\n",
      "Epoch 11/40\n",
      "954/954 [==============================] - 47s 50ms/step - loss: 1.0612 - acc: 0.7351 - val_loss: 0.8033 - val_acc: 0.8200\n",
      "Epoch 12/40\n",
      "954/954 [==============================] - 48s 51ms/step - loss: 1.0055 - acc: 0.7482 - val_loss: 0.7577 - val_acc: 0.8308\n",
      "Epoch 13/40\n",
      "954/954 [==============================] - 47s 49ms/step - loss: 0.9544 - acc: 0.7624 - val_loss: 0.7389 - val_acc: 0.8285\n",
      "Epoch 14/40\n",
      "954/954 [==============================] - 59s 62ms/step - loss: 0.9134 - acc: 0.7719 - val_loss: 0.7114 - val_acc: 0.8408\n",
      "Epoch 15/40\n",
      "954/954 [==============================] - 58s 61ms/step - loss: 0.8735 - acc: 0.7834 - val_loss: 0.6873 - val_acc: 0.8433\n",
      "Epoch 16/40\n",
      "954/954 [==============================] - 58s 61ms/step - loss: 0.8398 - acc: 0.7920 - val_loss: 0.6529 - val_acc: 0.8519\n",
      "Epoch 17/40\n",
      "954/954 [==============================] - 52s 55ms/step - loss: 0.8160 - acc: 0.7979 - val_loss: 0.6353 - val_acc: 0.8546\n",
      "Epoch 18/40\n",
      "954/954 [==============================] - 45s 47ms/step - loss: 0.7844 - acc: 0.8050 - val_loss: 0.5976 - val_acc: 0.8626\n",
      "Epoch 19/40\n",
      "954/954 [==============================] - 50s 52ms/step - loss: 0.7572 - acc: 0.8128 - val_loss: 0.5983 - val_acc: 0.8614\n",
      "Epoch 20/40\n",
      "954/954 [==============================] - 51s 53ms/step - loss: 0.7457 - acc: 0.8146 - val_loss: 0.5827 - val_acc: 0.8647\n",
      "Epoch 21/40\n",
      "954/954 [==============================] - 52s 54ms/step - loss: 0.7195 - acc: 0.8235 - val_loss: 0.5566 - val_acc: 0.8702\n",
      "Epoch 22/40\n",
      "954/954 [==============================] - 51s 53ms/step - loss: 0.6977 - acc: 0.8274 - val_loss: 0.5586 - val_acc: 0.8701\n",
      "Epoch 23/40\n",
      "954/954 [==============================] - 46s 48ms/step - loss: 0.6811 - acc: 0.8303 - val_loss: 0.5513 - val_acc: 0.8730\n",
      "Epoch 24/40\n",
      "954/954 [==============================] - 52s 54ms/step - loss: 0.6649 - acc: 0.8354 - val_loss: 0.5259 - val_acc: 0.8810\n",
      "Epoch 25/40\n",
      "954/954 [==============================] - 45s 47ms/step - loss: 0.6495 - acc: 0.8382 - val_loss: 0.5197 - val_acc: 0.8830\n",
      "Epoch 26/40\n",
      "954/954 [==============================] - 53s 56ms/step - loss: 0.6374 - acc: 0.8413 - val_loss: 0.5070 - val_acc: 0.8855\n",
      "Epoch 27/40\n",
      "954/954 [==============================] - 52s 55ms/step - loss: 0.6248 - acc: 0.8460 - val_loss: 0.5123 - val_acc: 0.8823\n",
      "Epoch 28/40\n",
      "954/954 [==============================] - 55s 57ms/step - loss: 0.6080 - acc: 0.8501 - val_loss: 0.5023 - val_acc: 0.8806\n",
      "Epoch 29/40\n",
      "954/954 [==============================] - 60s 63ms/step - loss: 0.5989 - acc: 0.8523 - val_loss: 0.4880 - val_acc: 0.8884\n",
      "Epoch 30/40\n",
      "954/954 [==============================] - 40s 42ms/step - loss: 0.5856 - acc: 0.8542 - val_loss: 0.4854 - val_acc: 0.8863\n",
      "Epoch 31/40\n",
      "954/954 [==============================] - 56s 59ms/step - loss: 0.5817 - acc: 0.8554 - val_loss: 0.4848 - val_acc: 0.8884\n",
      "Epoch 32/40\n",
      "954/954 [==============================] - 47s 49ms/step - loss: 0.5657 - acc: 0.8591 - val_loss: 0.4732 - val_acc: 0.8889\n",
      "Epoch 33/40\n",
      "954/954 [==============================] - 52s 55ms/step - loss: 0.5618 - acc: 0.8585 - val_loss: 0.4695 - val_acc: 0.8905\n",
      "Epoch 34/40\n",
      "954/954 [==============================] - 41s 42ms/step - loss: 0.5492 - acc: 0.8621 - val_loss: 0.4653 - val_acc: 0.8937\n",
      "Epoch 35/40\n",
      "954/954 [==============================] - 43s 45ms/step - loss: 0.5357 - acc: 0.8660 - val_loss: 0.4607 - val_acc: 0.8901\n",
      "Epoch 36/40\n",
      "954/954 [==============================] - 56s 59ms/step - loss: 0.5325 - acc: 0.8665 - val_loss: 0.4723 - val_acc: 0.8895\n",
      "Epoch 37/40\n",
      "954/954 [==============================] - 53s 55ms/step - loss: 0.5239 - acc: 0.8689 - val_loss: 0.4454 - val_acc: 0.8965\n",
      "Epoch 38/40\n",
      "954/954 [==============================] - 57s 60ms/step - loss: 0.5163 - acc: 0.8697 - val_loss: 0.4475 - val_acc: 0.8952\n",
      "Epoch 39/40\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "954/954 [==============================] - 44s 46ms/step - loss: 0.5080 - acc: 0.8732 - val_loss: 0.4398 - val_acc: 0.8973\n",
      "Epoch 40/40\n",
      "954/954 [==============================] - 67s 70ms/step - loss: 0.5010 - acc: 0.8752 - val_loss: 0.4468 - val_acc: 0.8958\n"
     ]
    }
   ],
   "source": [
    "# Create CNN model using Speech Commands MFCCs\n",
    "sc_cnn_input_shape = (sc_X_train.shape[1], sc_X_train.shape[2], 1)\n",
    "sc_output_units = 35\n",
    "sc_cnn_model = build_cnn_model(sc_cnn_input_shape, \n",
    "                               sc_output_units, \n",
    "                               learning_rate=0.0001)\n",
    "\n",
    " \n",
    "# Fit model to Speech Commands data\n",
    "sc_epochs = 40\n",
    "sc_batch_size = 64\n",
    "sc_patience = 3\n",
    "sc_cnn_results = fit_model(sc_cnn_model, \n",
    "                      sc_epochs, \n",
    "                      sc_batch_size, \n",
    "                      sc_patience, \n",
    "                      sc_X_train, \n",
    "                      sc_train_y, \n",
    "                      sc_X_val, \n",
    "                      sc_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(sc_cnn_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[613   3   0 ...   0   1   2]\n",
      " [  5 629   0 ...   0   2   1]\n",
      " [  0   1 339 ...   0   4   0]\n",
      " ...\n",
      " [  0   0   0 ... 225   0  46]\n",
      " [  1   0   1 ...   1 604   2]\n",
      " [  0   0   3 ...  34   4 602]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 0.32391873002052307\n",
      "Accuracy: 93.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 0.4416976571083069\n",
      "Accuracy: 90.0%\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model performance\n",
    "model_performance(sc_cnn_model, \n",
    "                  sc_X_train, \n",
    "                  sc_train_y,\n",
    "                  sc_X_test, \n",
    "                  sc_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### CNN model for the Ultrasuite dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_3 (Conv2D)            (None, 42, 11, 32)        320       \n",
      "_________________________________________________________________\n",
      "batch_normalization_8 (Batch (None, 42, 11, 32)        128       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 21, 6, 32)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 18, 3, 32)         16416     \n",
      "_________________________________________________________________\n",
      "batch_normalization_9 (Batch (None, 18, 3, 32)         128       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 9, 2, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 8, 1, 64)          8256      \n",
      "_________________________________________________________________\n",
      "batch_normalization_10 (Batc (None, 8, 1, 64)          256       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 4, 1, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_6 (Flatten)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_13 (Dense)             (None, 32)                8224      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_14 (Dense)             (None, 35)                1155      \n",
      "=================================================================\n",
      "Total params: 34,883\n",
      "Trainable params: 34,627\n",
      "Non-trainable params: 256\n",
      "_________________________________________________________________\n",
      "Epoch 1/50\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7ffc606efa70> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "246/247 [============================>.] - ETA: 0s - loss: 4.0910 - acc: 0.0587WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7ffc60e384d0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "247/247 [==============================] - 6s 21ms/step - loss: 4.0908 - acc: 0.0586 - val_loss: 3.7927 - val_acc: 0.1004\n",
      "Epoch 2/50\n",
      "247/247 [==============================] - 5s 18ms/step - loss: 3.7403 - acc: 0.0974 - val_loss: 3.5976 - val_acc: 0.1521\n",
      "Epoch 3/50\n",
      "247/247 [==============================] - 4s 17ms/step - loss: 3.5408 - acc: 0.1535 - val_loss: 3.3845 - val_acc: 0.2252\n",
      "Epoch 4/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 3.3737 - acc: 0.1951 - val_loss: 3.1844 - val_acc: 0.3053\n",
      "Epoch 5/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 3.2035 - acc: 0.2425 - val_loss: 2.9921 - val_acc: 0.3519\n",
      "Epoch 6/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 3.0242 - acc: 0.2867 - val_loss: 2.8223 - val_acc: 0.3742\n",
      "Epoch 7/50\n",
      "247/247 [==============================] - 6s 23ms/step - loss: 2.8464 - acc: 0.3262 - val_loss: 2.6281 - val_acc: 0.4341\n",
      "Epoch 8/50\n",
      "247/247 [==============================] - 4s 17ms/step - loss: 2.7237 - acc: 0.3620 - val_loss: 2.4881 - val_acc: 0.4787\n",
      "Epoch 9/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 2.5954 - acc: 0.3841 - val_loss: 2.3533 - val_acc: 0.5213\n",
      "Epoch 10/50\n",
      "247/247 [==============================] - 5s 18ms/step - loss: 2.4762 - acc: 0.4196 - val_loss: 2.2484 - val_acc: 0.5446\n",
      "Epoch 11/50\n",
      "247/247 [==============================] - 6s 23ms/step - loss: 2.3950 - acc: 0.4305 - val_loss: 2.1548 - val_acc: 0.5517\n",
      "Epoch 12/50\n",
      "247/247 [==============================] - 6s 23ms/step - loss: 2.2847 - acc: 0.4701 - val_loss: 2.0918 - val_acc: 0.5842\n",
      "Epoch 13/50\n",
      "247/247 [==============================] - 6s 23ms/step - loss: 2.2267 - acc: 0.4924 - val_loss: 2.0116 - val_acc: 0.5984\n",
      "Epoch 14/50\n",
      "247/247 [==============================] - 6s 23ms/step - loss: 2.1600 - acc: 0.4982 - val_loss: 1.9578 - val_acc: 0.6014\n",
      "Epoch 15/50\n",
      "247/247 [==============================] - 4s 17ms/step - loss: 2.0876 - acc: 0.5195 - val_loss: 1.9197 - val_acc: 0.6095\n",
      "Epoch 16/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 2.0407 - acc: 0.5190 - val_loss: 1.8734 - val_acc: 0.6318\n",
      "Epoch 17/50\n",
      "247/247 [==============================] - 5s 18ms/step - loss: 1.9878 - acc: 0.5302 - val_loss: 1.8275 - val_acc: 0.6329\n",
      "Epoch 18/50\n",
      "247/247 [==============================] - 5s 19ms/step - loss: 1.9344 - acc: 0.5512 - val_loss: 1.7812 - val_acc: 0.6420\n",
      "Epoch 19/50\n",
      "247/247 [==============================] - 5s 18ms/step - loss: 1.8902 - acc: 0.5538 - val_loss: 1.7528 - val_acc: 0.6430\n",
      "Epoch 20/50\n",
      "247/247 [==============================] - 4s 18ms/step - loss: 1.8324 - acc: 0.5708 - val_loss: 1.7003 - val_acc: 0.6420\n",
      "Epoch 21/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.7940 - acc: 0.5840 - val_loss: 1.7111 - val_acc: 0.6582\n",
      "Epoch 22/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.7602 - acc: 0.5923 - val_loss: 1.6437 - val_acc: 0.6694\n",
      "Epoch 23/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.7274 - acc: 0.6012 - val_loss: 1.6252 - val_acc: 0.6704\n",
      "Epoch 24/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.6829 - acc: 0.6101 - val_loss: 1.6119 - val_acc: 0.6795\n",
      "Epoch 25/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.6464 - acc: 0.6253 - val_loss: 1.5987 - val_acc: 0.6643\n",
      "Epoch 26/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.6449 - acc: 0.6240 - val_loss: 1.5735 - val_acc: 0.6876\n",
      "Epoch 27/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.5907 - acc: 0.6312 - val_loss: 1.5419 - val_acc: 0.6937\n",
      "Epoch 28/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.5673 - acc: 0.6474 - val_loss: 1.5426 - val_acc: 0.6856\n",
      "Epoch 29/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.5410 - acc: 0.6398 - val_loss: 1.5068 - val_acc: 0.6927\n",
      "Epoch 30/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.4843 - acc: 0.6568 - val_loss: 1.4887 - val_acc: 0.6968\n",
      "Epoch 31/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.4563 - acc: 0.6626 - val_loss: 1.4859 - val_acc: 0.6947\n",
      "Epoch 32/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.4650 - acc: 0.6682 - val_loss: 1.4875 - val_acc: 0.6927\n",
      "Epoch 33/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.4224 - acc: 0.6692 - val_loss: 1.4581 - val_acc: 0.7049\n",
      "Epoch 34/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.4209 - acc: 0.6700 - val_loss: 1.4327 - val_acc: 0.7181\n",
      "Epoch 35/50\n",
      "247/247 [==============================] - 4s 17ms/step - loss: 1.3814 - acc: 0.6763 - val_loss: 1.4364 - val_acc: 0.7059\n",
      "Epoch 36/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.3704 - acc: 0.6811 - val_loss: 1.4012 - val_acc: 0.7079\n",
      "Epoch 37/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.3538 - acc: 0.6903 - val_loss: 1.3988 - val_acc: 0.7160\n",
      "Epoch 38/50\n",
      "247/247 [==============================] - 5s 21ms/step - loss: 1.3115 - acc: 0.7035 - val_loss: 1.3917 - val_acc: 0.7181\n",
      "Epoch 39/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "247/247 [==============================] - 6s 23ms/step - loss: 1.3139 - acc: 0.6936 - val_loss: 1.3899 - val_acc: 0.7110\n",
      "Epoch 40/50\n",
      "247/247 [==============================] - 5s 21ms/step - loss: 1.2923 - acc: 0.7055 - val_loss: 1.3968 - val_acc: 0.7160\n",
      "Epoch 41/50\n",
      "247/247 [==============================] - 5s 21ms/step - loss: 1.2689 - acc: 0.7057 - val_loss: 1.3584 - val_acc: 0.7302\n",
      "Epoch 42/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.2618 - acc: 0.7103 - val_loss: 1.3751 - val_acc: 0.7201\n",
      "Epoch 43/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.2337 - acc: 0.7139 - val_loss: 1.3636 - val_acc: 0.7221\n",
      "Epoch 44/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.2307 - acc: 0.7166 - val_loss: 1.3521 - val_acc: 0.7241\n",
      "Epoch 45/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.2308 - acc: 0.7050 - val_loss: 1.3500 - val_acc: 0.7262\n",
      "Epoch 46/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.2048 - acc: 0.7237 - val_loss: 1.3460 - val_acc: 0.7231\n",
      "Epoch 47/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.1851 - acc: 0.7349 - val_loss: 1.3209 - val_acc: 0.7323\n",
      "Epoch 48/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.1572 - acc: 0.7392 - val_loss: 1.3335 - val_acc: 0.7272\n",
      "Epoch 49/50\n",
      "247/247 [==============================] - 4s 15ms/step - loss: 1.1328 - acc: 0.7438 - val_loss: 1.3180 - val_acc: 0.7363\n",
      "Epoch 50/50\n",
      "247/247 [==============================] - 4s 16ms/step - loss: 1.1325 - acc: 0.7425 - val_loss: 1.3151 - val_acc: 0.7252\n"
     ]
    }
   ],
   "source": [
    "# Create CNN model using Ultrasuite MFCCs\n",
    "us_cnn_input_shape = (us_X_train.shape[1], us_X_train.shape[2], 1)\n",
    "us_output_units = 35\n",
    "us_cnn_model = build_cnn_model(us_cnn_input_shape, \n",
    "                               us_output_units, \n",
    "                               learning_rate=0.0001)\n",
    "\n",
    " \n",
    "# Fit model to Ultrasuite data\n",
    "us_epochs = 50\n",
    "us_batch_size = 16\n",
    "us_patience = 3\n",
    "us_cnn_results = fit_model(us_cnn_model, \n",
    "                      us_epochs, \n",
    "                      us_batch_size, \n",
    "                      us_patience, \n",
    "                      us_X_train, \n",
    "                      us_train_y, \n",
    "                      us_X_val, \n",
    "                      us_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualise the training and validation loss and accuracy across epochs\n",
    "visualise_results(us_cnn_results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_model(us_cnn_model, 'final_model.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final Model Performance Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONFUSION MATRIX ------------------\n",
      "[[17  0  0 ...  0  0  1]\n",
      " [ 0 35  0 ...  1  0  1]\n",
      " [ 0  0 29 ...  0  1  0]\n",
      " ...\n",
      " [ 1  0  0 ... 15  1  0]\n",
      " [ 0  1  2 ...  0 43  0]\n",
      " [ 1  0  0 ...  0  0 16]]\n",
      "\n",
      "TRAIN METRICS ----------------------\n",
      "Loss: 0.7186428308486938\n",
      "Accuracy: 89.0%\n",
      "\n",
      "TEST METRICS ----------------------\n",
      "Loss: 1.3699027299880981\n",
      "Accuracy: 72.0%\n"
     ]
    }
   ],
   "source": [
    "model_performance(us_cnn_model, \n",
    "                  us_X_train, \n",
    "                  us_train_y,\n",
    "                  us_X_test, \n",
    "                  us_test_y\n",
    "                 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "        crab       0.74      0.61      0.67        28\n",
      "     fishing       0.80      0.81      0.80        43\n",
      "        gown       0.78      0.76      0.77        38\n",
      "         say       0.74      0.73      0.73        62\n",
      "         boy       0.96      0.81      0.88        27\n",
      "      bridge       0.89      0.82      0.85        49\n",
      "          in       0.59      0.53      0.56        45\n",
      "       zebra       0.62      0.69      0.66        29\n",
      "       times       0.60      0.58      0.59        31\n",
      "       sheep       0.76      0.61      0.68        31\n",
      "       tiger       0.65      0.56      0.60        27\n",
      "       teeth       0.56      0.68      0.61        22\n",
      "        pack       0.88      0.68      0.77        22\n",
      "  helicopter       0.85      0.94      0.89        47\n",
      "         got       0.67      0.57      0.62        49\n",
      "         ear       0.81      0.74      0.77        34\n",
      "         cab       0.69      0.75      0.72        24\n",
      "       thank       0.74      0.72      0.73        39\n",
      "         tie       0.55      0.40      0.46        15\n",
      "      gloves       0.79      0.88      0.83        42\n",
      "    umbrella       0.80      0.89      0.85        37\n",
      "      spider       0.79      0.79      0.79        43\n",
      "         was       0.47      0.78      0.59        36\n",
      "       watch       0.79      0.82      0.80        40\n",
      "      monkey       0.75      0.65      0.70        23\n",
      "    scissors       0.79      0.84      0.82        58\n",
      "      yellow       0.81      0.71      0.75        41\n",
      "        frog       0.75      0.50      0.60        30\n",
      "          or       0.39      0.78      0.52        23\n",
      "          on       0.58      0.48      0.53        31\n",
      "         ken       0.62      0.74      0.68        27\n",
      "      school       0.74      0.61      0.67        33\n",
      "        five       0.83      0.48      0.61        31\n",
      "    elephant       0.88      0.84      0.86        51\n",
      "        four       0.43      0.64      0.52        25\n",
      "\n",
      "    accuracy                           0.72      1233\n",
      "   macro avg       0.72      0.70      0.70      1233\n",
      "weighted avg       0.73      0.72      0.72      1233\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Import necessary libraries for classification report\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "preds = us_cnn_model.predict(us_X_test)\n",
    "\n",
    "us_keywords = [\n",
    "    'crab',\n",
    "    'fishing',\n",
    "    'gown',\n",
    "    'say',\n",
    "    'boy',\n",
    "    'bridge',\n",
    "    'in',\n",
    "    'zebra',\n",
    "    'times',\n",
    "    'sheep',\n",
    "    'tiger',\n",
    "    'teeth',\n",
    "    'pack',\n",
    "    'helicopter',\n",
    "    'got',\n",
    "    'ear',\n",
    "    'cab',\n",
    "    'thank',\n",
    "    'tie',\n",
    "    'gloves',\n",
    "    'umbrella',\n",
    "    'spider',\n",
    "    'was',\n",
    "    'watch',\n",
    "    'monkey',\n",
    "    'scissors',\n",
    "    'yellow',\n",
    "    'frog',\n",
    "    'or',\n",
    "    'on',\n",
    "    'ken',\n",
    "    'school',\n",
    "    'five',\n",
    "    'elephant',\n",
    "    'four']\n",
    "\n",
    "# Print the classification report\n",
    "print(classification_report(us_test_y.argmax(axis=1), \n",
    "                            preds.argmax(axis=1), \n",
    "                            target_names=us_keywords))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1152 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use Seaborn to make the confusion matrix more visually presentable\n",
    "cm = confusion_matrix(us_test_y.argmax(axis=1), \n",
    "                      np.round(preds.argmax(axis=1)))\n",
    "\n",
    "plt.figure(figsize=(20,16))\n",
    "ax = plt.subplot()\n",
    "sns.heatmap(cm, annot=True, ax=ax, fmt='g', cmap='Blues')\n",
    "\n",
    "ax.set_title('Confusion Matrix', fontsize=22, pad=30)\n",
    "ax.set_xlabel('Predicted Label', fontsize=18)\n",
    "ax.set_ylabel('Actual Label', fontsize=18)\n",
    "ax.xaxis.set_ticklabels(us_keywords, rotation=90, fontsize=12)\n",
    "ax.yaxis.set_ticklabels(us_keywords, rotation=0, fontsize=12)\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making a prediction on an unseen audio sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Audio sample: Umbrella\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "                <audio  controls=\"controls\" >\n",
       "                    <source src=\"data:audio/wav;base64,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\" type=\"audio/wav\" />\n",
       "                    Your browser does not support the audio element.\n",
       "                </audio>\n",
       "              "
      ],
      "text/plain": [
       "<IPython.lib.display.Audio object>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the audio sample and preview\n",
    "target_sample = 'audio/martha-umbrella.wav'\n",
    "target_label = 'Umbrella'\n",
    "audio_sample, sr = librosa.load(target_sample)\n",
    "print('Audio sample:', target_label)\n",
    "ipd.Audio(audio_sample, rate=sr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the waveform for the specific audio sample\n",
    "plt.figure(figsize=(15, 5))\n",
    "plt.title(target_label, fontsize=18, pad=20)\n",
    "librosa.display.waveplot(audio_sample, sr, alpha=0.8)\n",
    "plt.xlabel('Time (seconds)', fontsize=14, labelpad=20)\n",
    "plt.ylabel('Amplitude', fontsize=14, labelpad=20)\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Martha says... umbrella !\n"
     ]
    }
   ],
   "source": [
    "# Run inference on the unseen audio file\n",
    "mfccs = librosa.feature.mfcc(audio_sample, \n",
    "                             sr,\n",
    "                             n_mfcc=13, \n",
    "                             n_fft=2048, \n",
    "                             hop_length=512)\n",
    "mfccs = mfccs.T\n",
    "mfccs = mfccs[np.newaxis, ..., np.newaxis]\n",
    "\n",
    "prediction = us_cnn_model.predict(mfccs)\n",
    "predicted_index = np.argmax(prediction)\n",
    "\n",
    "predicted_keyword = us_keywords[predicted_index]\n",
    "print('Martha says...',predicted_keyword, '!')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Overall Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Initial models did not generalise well and tended to overfit the training data. Subsequent changes to parameters significantly improved the training accuracy but, again were overfitting the training data.\n",
    "\n",
    "Whilst it does not have a high accuracy score, this final model using a Convolutional Neural Network produced the best results classifying over 70% of the “unseen” audio samples whilst minimising the overfitting to training data.  Looking at the classification report, it is also clear that the model performs better for more identifiable words such as `scissors` or `umbrella`.\n",
    "\n",
    "The very nature of speech sound disorders mean that a model that has been simply trained on audio samples of \"typical\" speech will generally be more accurate than one that has been trained on audio samples of \"atypical\" speech as demonstrated above with the model comparison between the Speech Commands and Ultrasuite datasets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Recommendations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/google-recorder-app.png\" title=\"Google Recorder App Screenshot\" width=\"25%\" style=\"float: right; padding-left: 30px;\"/>\n",
    "\n",
    "It is envisaged that the model would be used in the form of a mobile app, similar to that of Google Recorder (see image), that could be activated by the parent in order to capture the child speaking.\n",
    "\n",
    "The app would also enable the parent to crop or isolate the audio sample and would provide a side-by-side prediction or transcript of what the child was saying.\n",
    "\n",
    "This could even be provided as a setting within the Google recorder app itself that allowed the user, in this case the parent, to switch between the standard speech recognition model and our final model specifically trained for children with speech disorders.\n",
    "\n",
    "Given this eventual usage of this model, it is arguable the app would be more useful if it suggested three potential words in order of likelihood, giving the parent options of what the child might be trying to communicate.\n",
    "\n",
    "The accuracy of the final model is suitable enough to go ahead with a soft launch of the app to a controlled group of parents, in part for testing but also as a way of collecting additional data to improve the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Future Work"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Future work to improve the model could include:\n",
    "\n",
    "* Utilise other model architectures and enable them to accept longer audio samples as these are more representative of atypical speech patterns.\n",
    "\n",
    "* Source additional data in the form of further audio samples potentially even using the app as a means for gathering additional samples and improving the model.\n",
    "\n",
    "* Use data augmentation when training the model, in particular the [MixSpeech](https://arxiv.org/abs/2102.12664) method that could take a weighted combination of mel-spectrograms and MFCCs in order to improve model performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr size=\"1\" />\n",
    "<small>\n",
    "<strong>Sources / Code adapted from:</strong><br/>\n",
    "    * <a href=\"https://www.oreilly.com/library/view/hands-on-machine-learning/9781492032632/\" target=\"_new\">Hands-On Machine Learning with Scikit-Learn, Keras & Tensorflow - Aurélien Géron</a><br/>\n",
    "    * <a href=\"https://github.com/musikalkemist/Deep-Learning-Audio-Application-From-Design-to-Deployment\" target=\"_new\">Deep Learning Audio Application from Design to Deployment - Valerio Velardo - The Sound of AI</a><br/>\n",
    "</small>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:capstone-env] *",
   "language": "python",
   "name": "conda-env-capstone-env-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  },
  "toc": {
   "base_numbering": "4",
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}