DYSO.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3a97ede5",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Let's read a sample audio using librosa\n",
    "import librosa\n",
    "audio_file_path='D:/dataset/healthy/1-a_l.wav'\n",
    "librosa_audio_data,librosa_sample_rate=librosa.load(audio_file_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1d33bca9",
   "metadata": {},
   "outputs": [],
   "source": [
    "audio_file_path='D:/dataset/pathological/561-a_lhl.wav'\n",
    "libosa_audio_data,librosa_sample_rate=librosa.load(audio_file_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e5db838e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.01416103  0.00679945  0.05568034 ...  0.10350126  0.11146182\n",
      "  0.08813323]\n",
      "[0.31210828 0.47397637 0.38048622 ... 0.22966844 0.26631647 0.        ]\n"
     ]
    }
   ],
   "source": [
    "print(librosa_audio_data)\n",
    "print(libosa_audio_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "74002af8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x28f0aa28d30>]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Lets plot the librosa audio data\n",
    "import matplotlib.pyplot as plt\n",
    "# Original audio with 1 channel \n",
    "plt.figure(figsize=(12, 4))\n",
    "# plt.plot(librosa_audio_data)\n",
    "plt.plot(libosa_audio_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c974d6cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x28f0b7b51b0>]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(libosa_audio_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9f39cfee",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c7afe93b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#### Extracting MFCC's For every audio file\n",
    "import pandas as pd\n",
    "import os\n",
    "import librosa\n",
    "import pandas as pd\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "93e655d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\pydub\\utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work\n",
      "  warn(\"Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work\", RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                  file_path  duration         label\n",
      "1864    D:\\dataset\\pathological\\925-i_n.wav   1.96770  pathological\n",
      "1865    D:\\dataset\\pathological\\925-u_h.wav   1.17874  pathological\n",
      "1866    D:\\dataset\\pathological\\925-u_l.wav   1.59148  pathological\n",
      "1867  D:\\dataset\\pathological\\925-u_lhl.wav   1.14544  pathological\n",
      "1868    D:\\dataset\\pathological\\925-u_n.wav   1.67996  pathological\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from pydub import AudioSegment\n",
    "import os\n",
    "\n",
    "\n",
    "data_dir = 'D:\\dataset'\n",
    "\n",
    "def extract_features(file_path):\n",
    "    audio = AudioSegment.from_file(file_path)\n",
    "    duration = audio.duration_seconds\n",
    "    \n",
    "    \n",
    "    return {\n",
    "        'file_path': file_path,\n",
    "        'duration': duration,\n",
    "    }\n",
    "\n",
    "features_list = []\n",
    "\n",
    "for label in ['healthy', 'pathological']:\n",
    "    label_dir = os.path.join(data_dir, label)\n",
    "    \n",
    "    for filename in os.listdir(label_dir):\n",
    "        if filename.endswith('.wav'):\n",
    "            file_path = os.path.join(label_dir, filename)\n",
    "            \n",
    "            features = extract_features(file_path)\n",
    "            \n",
    "            features['label'] = label\n",
    "            \n",
    "            features_list.append(features)\n",
    "\n",
    "df = pd.DataFrame(features_list)\n",
    "\n",
    "print(df.tail())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cd43217b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DataFrame has been successfully saved to dysphonia.csv\n"
     ]
    }
   ],
   "source": [
    "\n",
    "csv_file_path = 'dysphonia.csv'\n",
    "\n",
    "df.to_csv(csv_file_path, index=False)\n",
    "\n",
    "print(f\"DataFrame has been successfully saved to {csv_file_path}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "bc0053cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>file_path</th>\n",
       "      <th>duration</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>D:\\dataset\\healthy\\1-a_h.wav</td>\n",
       "      <td>1.99352</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>D:\\dataset\\healthy\\1-a_l.wav</td>\n",
       "      <td>1.93368</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>D:\\dataset\\healthy\\1-a_lhl.wav</td>\n",
       "      <td>1.42704</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>D:\\dataset\\healthy\\1-a_n.wav</td>\n",
       "      <td>1.94836</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>D:\\dataset\\healthy\\1-i_h.wav</td>\n",
       "      <td>0.62324</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        file_path  duration    label\n",
       "0    D:\\dataset\\healthy\\1-a_h.wav   1.99352  healthy\n",
       "1    D:\\dataset\\healthy\\1-a_l.wav   1.93368  healthy\n",
       "2  D:\\dataset\\healthy\\1-a_lhl.wav   1.42704  healthy\n",
       "3    D:\\dataset\\healthy\\1-a_n.wav   1.94836  healthy\n",
       "4    D:\\dataset\\healthy\\1-i_h.wav   0.62324  healthy"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data=pd.read_csv(\"dysphonia.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d11fa5d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>file_path</th>\n",
       "      <th>duration</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1864</th>\n",
       "      <td>D:\\dataset\\pathological\\925-i_n.wav</td>\n",
       "      <td>1.96770</td>\n",
       "      <td>pathological</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1865</th>\n",
       "      <td>D:\\dataset\\pathological\\925-u_h.wav</td>\n",
       "      <td>1.17874</td>\n",
       "      <td>pathological</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1866</th>\n",
       "      <td>D:\\dataset\\pathological\\925-u_l.wav</td>\n",
       "      <td>1.59148</td>\n",
       "      <td>pathological</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1867</th>\n",
       "      <td>D:\\dataset\\pathological\\925-u_lhl.wav</td>\n",
       "      <td>1.14544</td>\n",
       "      <td>pathological</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1868</th>\n",
       "      <td>D:\\dataset\\pathological\\925-u_n.wav</td>\n",
       "      <td>1.67996</td>\n",
       "      <td>pathological</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  file_path  duration         label\n",
       "1864    D:\\dataset\\pathological\\925-i_n.wav   1.96770  pathological\n",
       "1865    D:\\dataset\\pathological\\925-u_h.wav   1.17874  pathological\n",
       "1866    D:\\dataset\\pathological\\925-u_l.wav   1.59148  pathological\n",
       "1867  D:\\dataset\\pathological\\925-u_lhl.wav   1.14544  pathological\n",
       "1868    D:\\dataset\\pathological\\925-u_n.wav   1.67996  pathological"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3dfc232a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "0it [00:00, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "1869it [00:15, 118.93it/s]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "import resampy  \n",
    "import librosa\n",
    "\n",
    "extracted_features = []\n",
    "for index_num, row in tqdm(data.iterrows()):\n",
    "    file_name=row['file_path']\n",
    "    final_class_labels = row[\"label\"]\n",
    "    audio, sample_rate = librosa.load(file_name) \n",
    "    mfccs_features = librosa.feature.mfcc(y=audio, sr=sample_rate, n_mfcc=40)\n",
    "    mfccs_scaled_features = np.mean(mfccs_features.T,axis=0)\n",
    "    extracted_features.append([mfccs_scaled_features, final_class_labels])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "05cf8e62",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feature</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[-234.28789, 47.39589, -34.445663, -36.55056, ...</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[-215.07884, 102.238014, -35.83801, -17.622807...</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[-201.52502, 106.307175, -40.84559, -39.983242...</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[-226.80856, 66.601944, -27.219215, -42.401924...</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[-311.3803, 38.118633, -21.941334, 53.864014, ...</td>\n",
       "      <td>healthy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             feature    label\n",
       "0  [-234.28789, 47.39589, -34.445663, -36.55056, ...  healthy\n",
       "1  [-215.07884, 102.238014, -35.83801, -17.622807...  healthy\n",
       "2  [-201.52502, 106.307175, -40.84559, -39.983242...  healthy\n",
       "3  [-226.80856, 66.601944, -27.219215, -42.401924...  healthy\n",
       "4  [-311.3803, 38.118633, -21.941334, 53.864014, ...  healthy"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "extracted_features_df=pd.DataFrame(extracted_features,columns=['feature','label'])\n",
    "extracted_features_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0c6452d0",
   "metadata": {},
   "outputs": [],
   "source": [
    "X=np.array(extracted_features_df['feature'].tolist())\n",
    "y=np.array(extracted_features_df[\"label\"].tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "531ae435",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from keras.utils import to_categorical\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "labelencoder=LabelEncoder()\n",
    "y=to_categorical(labelencoder.fit_transform(y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b2ed7224",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       ...,\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "9b0d9be6",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Train Test Split\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "202b0051",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       ...,\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "1e73e35e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1495, 40)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "be6a1a84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(374, 40)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "10c88d93",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1495, 2)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "3b6b7f29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(374, 2)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "004010db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.15.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "54a40ac3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense,Dropout,Activation,Flatten\n",
    "from keras.optimizers import Adam\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "21aa00a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "num_labels=y.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4f9888f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "e14394a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\backend.py:873: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model=Sequential()\n",
    "###first layer\n",
    "model.add(Dense(100,input_shape=(40,)))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "###second layer\n",
    "model.add(Dense(200))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "###third layer\n",
    "model.add(Dense(100))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "###final layer\n",
    "model.add(Dense(num_labels))\n",
    "model.add(Activation('softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "0ada25c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 100)               4100      \n",
      "                                                                 \n",
      " activation (Activation)     (None, 100)               0         \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 100)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 200)               20200     \n",
      "                                                                 \n",
      " activation_1 (Activation)   (None, 200)               0         \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 200)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 100)               20100     \n",
      "                                                                 \n",
      " activation_2 (Activation)   (None, 100)               0         \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 100)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 2)                 202       \n",
      "                                                                 \n",
      " activation_3 (Activation)   (None, 2)                 0         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 44602 (174.23 KB)\n",
      "Trainable params: 44602 (174.23 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "0be6591d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\optimizers\\__init__.py:309: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model.compile(loss='categorical_crossentropy',metrics=['accuracy'],optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "810b1fe9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\utils\\tf_utils.py:492: The name tf.ragged.RaggedTensorValue is deprecated. Please use tf.compat.v1.ragged.RaggedTensorValue instead.\n",
      "\n",
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\engine\\base_layer_utils.py:384: The name tf.executing_eagerly_outside_functions is deprecated. Please use tf.compat.v1.executing_eagerly_outside_functions instead.\n",
      "\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 16.2129 - accuracy: 0.5120 \n",
      "Epoch 1: val_loss improved from inf to 1.45814, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 2s 12ms/step - loss: 15.3158 - accuracy: 0.5084 - val_loss: 1.4581 - val_accuracy: 0.4947\n",
      "Epoch 2/100\n",
      "24/47 [==============>...............] - ETA: 0s - loss: 7.0143 - accuracy: 0.5378"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\engine\\training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
      "  saving_api.save_model(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 2: val_loss improved from 1.45814 to 0.91365, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 6.1388 - accuracy: 0.5224 - val_loss: 0.9136 - val_accuracy: 0.5455\n",
      "Epoch 3/100\n",
      "45/47 [===========================>..] - ETA: 0s - loss: 3.3390 - accuracy: 0.5194\n",
      "Epoch 3: val_loss improved from 0.91365 to 0.70179, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 3.2890 - accuracy: 0.5217 - val_loss: 0.7018 - val_accuracy: 0.5642\n",
      "Epoch 4/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 2.0921 - accuracy: 0.5221\n",
      "Epoch 4: val_loss improved from 0.70179 to 0.68293, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 2.0246 - accuracy: 0.5271 - val_loss: 0.6829 - val_accuracy: 0.5936\n",
      "Epoch 5/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 1.6919 - accuracy: 0.5133\n",
      "Epoch 5: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 1.6447 - accuracy: 0.5130 - val_loss: 0.6843 - val_accuracy: 0.6150\n",
      "Epoch 6/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 1.1769 - accuracy: 0.5361\n",
      "Epoch 6: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 1.1596 - accuracy: 0.5278 - val_loss: 0.6869 - val_accuracy: 0.5561\n",
      "Epoch 7/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 1.0713 - accuracy: 0.5059\n",
      "Epoch 7: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 1.0622 - accuracy: 0.5050 - val_loss: 0.6915 - val_accuracy: 0.5668\n",
      "Epoch 8/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.9085 - accuracy: 0.5134\n",
      "Epoch 8: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.9118 - accuracy: 0.5184 - val_loss: 0.6871 - val_accuracy: 0.6176\n",
      "Epoch 9/100\n",
      "44/47 [===========================>..] - ETA: 0s - loss: 0.8642 - accuracy: 0.5014\n",
      "Epoch 9: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.8673 - accuracy: 0.5010 - val_loss: 0.6901 - val_accuracy: 0.5668\n",
      "Epoch 10/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.8077 - accuracy: 0.5078\n",
      "Epoch 10: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7983 - accuracy: 0.5097 - val_loss: 0.6918 - val_accuracy: 0.5187\n",
      "Epoch 11/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.7435 - accuracy: 0.5335\n",
      "Epoch 11: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7442 - accuracy: 0.5284 - val_loss: 0.6902 - val_accuracy: 0.6176\n",
      "Epoch 12/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.7458 - accuracy: 0.5164\n",
      "Epoch 12: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7433 - accuracy: 0.5197 - val_loss: 0.6890 - val_accuracy: 0.5401\n",
      "Epoch 13/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.7531 - accuracy: 0.5267\n",
      "Epoch 13: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7445 - accuracy: 0.5298 - val_loss: 0.6907 - val_accuracy: 0.5374\n",
      "Epoch 14/100\n",
      "26/47 [===============>..............] - ETA: 0s - loss: 0.7538 - accuracy: 0.5012\n",
      "Epoch 14: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.7391 - accuracy: 0.5050 - val_loss: 0.6885 - val_accuracy: 0.5428\n",
      "Epoch 15/100\n",
      "45/47 [===========================>..] - ETA: 0s - loss: 0.7224 - accuracy: 0.5271\n",
      "Epoch 15: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7239 - accuracy: 0.5278 - val_loss: 0.6904 - val_accuracy: 0.5722\n",
      "Epoch 16/100\n",
      "26/47 [===============>..............] - ETA: 0s - loss: 0.7351 - accuracy: 0.5288\n",
      "Epoch 16: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.7296 - accuracy: 0.5030 - val_loss: 0.6899 - val_accuracy: 0.5508\n",
      "Epoch 17/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.7127 - accuracy: 0.5134\n",
      "Epoch 17: val_loss did not improve from 0.68293\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7093 - accuracy: 0.5171 - val_loss: 0.6870 - val_accuracy: 0.5374\n",
      "Epoch 18/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.7234 - accuracy: 0.5168\n",
      "Epoch 18: val_loss improved from 0.68293 to 0.67763, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.7253 - accuracy: 0.5197 - val_loss: 0.6776 - val_accuracy: 0.6578\n",
      "Epoch 19/100\n",
      "36/47 [=====================>........] - ETA: 0s - loss: 0.7089 - accuracy: 0.5156\n",
      "Epoch 19: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.7116 - accuracy: 0.5151 - val_loss: 0.6887 - val_accuracy: 0.6043\n",
      "Epoch 20/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.7066 - accuracy: 0.5337\n",
      "Epoch 20: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7047 - accuracy: 0.5365 - val_loss: 0.6905 - val_accuracy: 0.6203\n",
      "Epoch 21/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.7044 - accuracy: 0.5168\n",
      "Epoch 21: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 6ms/step - loss: 0.7017 - accuracy: 0.5197 - val_loss: 0.6835 - val_accuracy: 0.6310\n",
      "Epoch 22/100\n",
      "46/47 [============================>.] - ETA: 0s - loss: 0.7006 - accuracy: 0.5374\n",
      "Epoch 22: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 7ms/step - loss: 0.7002 - accuracy: 0.5371 - val_loss: 0.6870 - val_accuracy: 0.6364\n",
      "Epoch 23/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.6913 - accuracy: 0.5357\n",
      "Epoch 23: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6927 - accuracy: 0.5344 - val_loss: 0.6877 - val_accuracy: 0.5802\n",
      "Epoch 24/100\n",
      "45/47 [===========================>..] - ETA: 0s - loss: 0.6995 - accuracy: 0.5424\n",
      "Epoch 24: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6975 - accuracy: 0.5438 - val_loss: 0.6872 - val_accuracy: 0.6230\n",
      "Epoch 25/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.7072 - accuracy: 0.5403\n",
      "Epoch 25: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7037 - accuracy: 0.5385 - val_loss: 0.6862 - val_accuracy: 0.6310\n",
      "Epoch 26/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.7116 - accuracy: 0.5270\n",
      "Epoch 26: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.7067 - accuracy: 0.5298 - val_loss: 0.6834 - val_accuracy: 0.6283\n",
      "Epoch 27/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.7046 - accuracy: 0.5524\n",
      "Epoch 27: val_loss did not improve from 0.67763\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.7017 - accuracy: 0.5525 - val_loss: 0.6804 - val_accuracy: 0.6471\n",
      "Epoch 28/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6774 - accuracy: 0.5699\n",
      "Epoch 28: val_loss improved from 0.67763 to 0.67624, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6802 - accuracy: 0.5645 - val_loss: 0.6762 - val_accuracy: 0.6364\n",
      "Epoch 29/100\n",
      "34/47 [====================>.........] - ETA: 0s - loss: 0.6942 - accuracy: 0.5579\n",
      "Epoch 29: val_loss did not improve from 0.67624\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6977 - accuracy: 0.5498 - val_loss: 0.6814 - val_accuracy: 0.6283\n",
      "Epoch 30/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6847 - accuracy: 0.5559\n",
      "Epoch 30: val_loss improved from 0.67624 to 0.66876, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6810 - accuracy: 0.5625 - val_loss: 0.6688 - val_accuracy: 0.6390\n",
      "Epoch 31/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.6842 - accuracy: 0.5497\n",
      "Epoch 31: val_loss improved from 0.66876 to 0.66806, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6833 - accuracy: 0.5545 - val_loss: 0.6681 - val_accuracy: 0.6390\n",
      "Epoch 32/100\n",
      "25/47 [==============>...............] - ETA: 0s - loss: 0.6762 - accuracy: 0.5850\n",
      "Epoch 32: val_loss did not improve from 0.66806\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6826 - accuracy: 0.5699 - val_loss: 0.6797 - val_accuracy: 0.5829\n",
      "Epoch 33/100\n",
      "44/47 [===========================>..] - ETA: 0s - loss: 0.6897 - accuracy: 0.5497\n",
      "Epoch 33: val_loss did not improve from 0.66806\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6896 - accuracy: 0.5492 - val_loss: 0.6770 - val_accuracy: 0.6230\n",
      "Epoch 34/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6627 - accuracy: 0.5766\n",
      "Epoch 34: val_loss did not improve from 0.66806\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6680 - accuracy: 0.5719 - val_loss: 0.6704 - val_accuracy: 0.6230\n",
      "Epoch 35/100\n",
      "45/47 [===========================>..] - ETA: 0s - loss: 0.6845 - accuracy: 0.5750\n",
      "Epoch 35: val_loss improved from 0.66806 to 0.66075, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6829 - accuracy: 0.5753 - val_loss: 0.6607 - val_accuracy: 0.6283\n",
      "Epoch 36/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.6807 - accuracy: 0.5841\n",
      "Epoch 36: val_loss did not improve from 0.66075\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6845 - accuracy: 0.5786 - val_loss: 0.6751 - val_accuracy: 0.6043\n",
      "Epoch 37/100\n",
      "28/47 [================>.............] - ETA: 0s - loss: 0.6616 - accuracy: 0.5815\n",
      "Epoch 37: val_loss did not improve from 0.66075\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6583 - accuracy: 0.5833 - val_loss: 0.6618 - val_accuracy: 0.6257\n",
      "Epoch 38/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6572 - accuracy: 0.6039\n",
      "Epoch 38: val_loss improved from 0.66075 to 0.64014, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6563 - accuracy: 0.5967 - val_loss: 0.6401 - val_accuracy: 0.6390\n",
      "Epoch 39/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.6576 - accuracy: 0.5930\n",
      "Epoch 39: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6572 - accuracy: 0.5960 - val_loss: 0.6613 - val_accuracy: 0.6016\n",
      "Epoch 40/100\n",
      "32/47 [===================>..........] - ETA: 0s - loss: 0.6542 - accuracy: 0.5938\n",
      "Epoch 40: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6587 - accuracy: 0.5906 - val_loss: 0.6516 - val_accuracy: 0.6230\n",
      "Epoch 41/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.6653 - accuracy: 0.5963\n",
      "Epoch 41: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6670 - accuracy: 0.5920 - val_loss: 0.6532 - val_accuracy: 0.6123\n",
      "Epoch 42/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.6533 - accuracy: 0.6037\n",
      "Epoch 42: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6523 - accuracy: 0.6020 - val_loss: 0.6493 - val_accuracy: 0.6016\n",
      "Epoch 43/100\n",
      "27/47 [================>.............] - ETA: 0s - loss: 0.6617 - accuracy: 0.5787\n",
      "Epoch 43: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.6583 - accuracy: 0.5926 - val_loss: 0.6463 - val_accuracy: 0.6176\n",
      "Epoch 44/100\n",
      "24/47 [==============>...............] - ETA: 0s - loss: 0.6406 - accuracy: 0.6185\n",
      "Epoch 44: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.6409 - accuracy: 0.6281 - val_loss: 0.6431 - val_accuracy: 0.6230\n",
      "Epoch 45/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.6484 - accuracy: 0.6106\n",
      "Epoch 45: val_loss did not improve from 0.64014\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6516 - accuracy: 0.6007 - val_loss: 0.6476 - val_accuracy: 0.6176\n",
      "Epoch 46/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.6397 - accuracy: 0.6128\n",
      "Epoch 46: val_loss improved from 0.64014 to 0.63917, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6343 - accuracy: 0.6147 - val_loss: 0.6392 - val_accuracy: 0.6310\n",
      "Epoch 47/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.6366 - accuracy: 0.6123\n",
      "Epoch 47: val_loss did not improve from 0.63917\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6458 - accuracy: 0.6074 - val_loss: 0.6538 - val_accuracy: 0.6016\n",
      "Epoch 48/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6513 - accuracy: 0.6086\n",
      "Epoch 48: val_loss did not improve from 0.63917\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6492 - accuracy: 0.6120 - val_loss: 0.6402 - val_accuracy: 0.6230\n",
      "Epoch 49/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6407 - accuracy: 0.6180\n",
      "Epoch 49: val_loss did not improve from 0.63917\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6452 - accuracy: 0.6154 - val_loss: 0.6442 - val_accuracy: 0.6230\n",
      "Epoch 50/100\n",
      "46/47 [============================>.] - ETA: 0s - loss: 0.6487 - accuracy: 0.6209\n",
      "Epoch 50: val_loss improved from 0.63917 to 0.62555, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 6ms/step - loss: 0.6488 - accuracy: 0.6214 - val_loss: 0.6256 - val_accuracy: 0.6417\n",
      "Epoch 51/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6391 - accuracy: 0.6357\n",
      "Epoch 51: val_loss did not improve from 0.62555\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6429 - accuracy: 0.6314 - val_loss: 0.6354 - val_accuracy: 0.6310\n",
      "Epoch 52/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.6399 - accuracy: 0.6233\n",
      "Epoch 52: val_loss improved from 0.62555 to 0.62404, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6345 - accuracy: 0.6381 - val_loss: 0.6240 - val_accuracy: 0.6471\n",
      "Epoch 53/100\n",
      "24/47 [==============>...............] - ETA: 0s - loss: 0.6509 - accuracy: 0.6068\n",
      "Epoch 53: val_loss improved from 0.62404 to 0.61646, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6322 - accuracy: 0.6201 - val_loss: 0.6165 - val_accuracy: 0.6337\n",
      "Epoch 54/100\n",
      "27/47 [================>.............] - ETA: 0s - loss: 0.6467 - accuracy: 0.6157\n",
      "Epoch 54: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.6334 - accuracy: 0.6361 - val_loss: 0.6190 - val_accuracy: 0.6444\n",
      "Epoch 55/100\n",
      "35/47 [=====================>........] - ETA: 0s - loss: 0.6464 - accuracy: 0.6143\n",
      "Epoch 55: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6405 - accuracy: 0.6207 - val_loss: 0.6222 - val_accuracy: 0.6497\n",
      "Epoch 56/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6273 - accuracy: 0.6225\n",
      "Epoch 56: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6249 - accuracy: 0.6247 - val_loss: 0.6298 - val_accuracy: 0.6257\n",
      "Epoch 57/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6367 - accuracy: 0.6299\n",
      "Epoch 57: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6366 - accuracy: 0.6261 - val_loss: 0.6298 - val_accuracy: 0.6417\n",
      "Epoch 58/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6330 - accuracy: 0.6308\n",
      "Epoch 58: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6279 - accuracy: 0.6341 - val_loss: 0.6261 - val_accuracy: 0.6471\n",
      "Epoch 59/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.6239 - accuracy: 0.6303\n",
      "Epoch 59: val_loss did not improve from 0.61646\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6253 - accuracy: 0.6288 - val_loss: 0.6228 - val_accuracy: 0.6444\n",
      "Epoch 60/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6081 - accuracy: 0.6570\n",
      "Epoch 60: val_loss improved from 0.61646 to 0.60606, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6070 - accuracy: 0.6555 - val_loss: 0.6061 - val_accuracy: 0.6551\n",
      "Epoch 61/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.6361 - accuracy: 0.6373\n",
      "Epoch 61: val_loss did not improve from 0.60606\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6399 - accuracy: 0.6274 - val_loss: 0.6397 - val_accuracy: 0.6203\n",
      "Epoch 62/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6270 - accuracy: 0.6219\n",
      "Epoch 62: val_loss did not improve from 0.60606\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6270 - accuracy: 0.6241 - val_loss: 0.6087 - val_accuracy: 0.6551\n",
      "Epoch 63/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.6192 - accuracy: 0.6398\n",
      "Epoch 63: val_loss improved from 0.60606 to 0.60485, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6194 - accuracy: 0.6388 - val_loss: 0.6049 - val_accuracy: 0.6551\n",
      "Epoch 64/100\n",
      "24/47 [==============>...............] - ETA: 0s - loss: 0.6056 - accuracy: 0.6484\n",
      "Epoch 64: val_loss did not improve from 0.60485\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.6114 - accuracy: 0.6395 - val_loss: 0.6076 - val_accuracy: 0.6578\n",
      "Epoch 65/100\n",
      "25/47 [==============>...............] - ETA: 0s - loss: 0.6088 - accuracy: 0.6450\n",
      "Epoch 65: val_loss improved from 0.60485 to 0.60121, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6064 - accuracy: 0.6408 - val_loss: 0.6012 - val_accuracy: 0.6711\n",
      "Epoch 66/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.5955 - accuracy: 0.6627\n",
      "Epoch 66: val_loss improved from 0.60121 to 0.60048, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.6008 - accuracy: 0.6575 - val_loss: 0.6005 - val_accuracy: 0.6738\n",
      "Epoch 67/100\n",
      "41/47 [=========================>....] - ETA: 0s - loss: 0.6077 - accuracy: 0.6540\n",
      "Epoch 67: val_loss did not improve from 0.60048\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.6076 - accuracy: 0.6582 - val_loss: 0.6117 - val_accuracy: 0.6524\n",
      "Epoch 68/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.5973 - accuracy: 0.6530\n",
      "Epoch 68: val_loss improved from 0.60048 to 0.58826, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5928 - accuracy: 0.6595 - val_loss: 0.5883 - val_accuracy: 0.6738\n",
      "Epoch 69/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.5876 - accuracy: 0.6765\n",
      "Epoch 69: val_loss did not improve from 0.58826\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5997 - accuracy: 0.6676 - val_loss: 0.5936 - val_accuracy: 0.6738\n",
      "Epoch 70/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.5777 - accuracy: 0.6755\n",
      "Epoch 70: val_loss did not improve from 0.58826\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5854 - accuracy: 0.6642 - val_loss: 0.6066 - val_accuracy: 0.6524\n",
      "Epoch 71/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.5796 - accuracy: 0.6727\n",
      "Epoch 71: val_loss did not improve from 0.58826\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5810 - accuracy: 0.6736 - val_loss: 0.5997 - val_accuracy: 0.6658\n",
      "Epoch 72/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.5918 - accuracy: 0.6664\n",
      "Epoch 72: val_loss did not improve from 0.58826\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5942 - accuracy: 0.6635 - val_loss: 0.5938 - val_accuracy: 0.6711\n",
      "Epoch 73/100\n",
      "43/47 [==========================>...] - ETA: 0s - loss: 0.5770 - accuracy: 0.6810\n",
      "Epoch 73: val_loss did not improve from 0.58826\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5818 - accuracy: 0.6789 - val_loss: 0.5916 - val_accuracy: 0.6658\n",
      "Epoch 74/100\n",
      "47/47 [==============================] - ETA: 0s - loss: 0.5876 - accuracy: 0.6676\n",
      "Epoch 74: val_loss improved from 0.58826 to 0.58513, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5876 - accuracy: 0.6676 - val_loss: 0.5851 - val_accuracy: 0.6765\n",
      "Epoch 75/100\n",
      "47/47 [==============================] - ETA: 0s - loss: 0.5798 - accuracy: 0.6883\n",
      "Epoch 75: val_loss did not improve from 0.58513\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.5798 - accuracy: 0.6883 - val_loss: 0.5901 - val_accuracy: 0.6684\n",
      "Epoch 76/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.5656 - accuracy: 0.6719\n",
      "Epoch 76: val_loss did not improve from 0.58513\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5623 - accuracy: 0.6823 - val_loss: 0.5853 - val_accuracy: 0.6818\n",
      "Epoch 77/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.5797 - accuracy: 0.6834\n",
      "Epoch 77: val_loss did not improve from 0.58513\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5786 - accuracy: 0.6789 - val_loss: 0.5892 - val_accuracy: 0.6872\n",
      "Epoch 78/100\n",
      "35/47 [=====================>........] - ETA: 0s - loss: 0.5852 - accuracy: 0.6679\n",
      "Epoch 78: val_loss did not improve from 0.58513\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5846 - accuracy: 0.6709 - val_loss: 0.5869 - val_accuracy: 0.6845\n",
      "Epoch 79/100\n",
      "43/47 [==========================>...] - ETA: 0s - loss: 0.5718 - accuracy: 0.6897\n",
      "Epoch 79: val_loss improved from 0.58513 to 0.57421, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5717 - accuracy: 0.6863 - val_loss: 0.5742 - val_accuracy: 0.6872\n",
      "Epoch 80/100\n",
      "43/47 [==========================>...] - ETA: 0s - loss: 0.5774 - accuracy: 0.6897\n",
      "Epoch 80: val_loss improved from 0.57421 to 0.57297, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5712 - accuracy: 0.6930 - val_loss: 0.5730 - val_accuracy: 0.6845\n",
      "Epoch 81/100\n",
      "38/47 [=======================>......] - ETA: 0s - loss: 0.5718 - accuracy: 0.6859\n",
      "Epoch 81: val_loss improved from 0.57297 to 0.57057, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5658 - accuracy: 0.6890 - val_loss: 0.5706 - val_accuracy: 0.6979\n",
      "Epoch 82/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.5510 - accuracy: 0.7227\n",
      "Epoch 82: val_loss did not improve from 0.57057\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5504 - accuracy: 0.7211 - val_loss: 0.5744 - val_accuracy: 0.6765\n",
      "Epoch 83/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.5727 - accuracy: 0.6898\n",
      "Epoch 83: val_loss improved from 0.57057 to 0.56359, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5693 - accuracy: 0.6890 - val_loss: 0.5636 - val_accuracy: 0.7193\n",
      "Epoch 84/100\n",
      "26/47 [===============>..............] - ETA: 0s - loss: 0.5720 - accuracy: 0.6875\n",
      "Epoch 84: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.5514 - accuracy: 0.7030 - val_loss: 0.5671 - val_accuracy: 0.6979\n",
      "Epoch 85/100\n",
      "27/47 [================>.............] - ETA: 0s - loss: 0.5578 - accuracy: 0.7141\n",
      "Epoch 85: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.5654 - accuracy: 0.6970 - val_loss: 0.5686 - val_accuracy: 0.6979\n",
      "Epoch 86/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.5436 - accuracy: 0.7131\n",
      "Epoch 86: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5471 - accuracy: 0.7077 - val_loss: 0.5677 - val_accuracy: 0.6925\n",
      "Epoch 87/100\n",
      "43/47 [==========================>...] - ETA: 0s - loss: 0.5608 - accuracy: 0.7049\n",
      "Epoch 87: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5568 - accuracy: 0.7064 - val_loss: 0.5669 - val_accuracy: 0.7086\n",
      "Epoch 88/100\n",
      "35/47 [=====================>........] - ETA: 0s - loss: 0.5625 - accuracy: 0.6884\n",
      "Epoch 88: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5541 - accuracy: 0.6970 - val_loss: 0.5788 - val_accuracy: 0.7059\n",
      "Epoch 89/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.5619 - accuracy: 0.7002\n",
      "Epoch 89: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5564 - accuracy: 0.7077 - val_loss: 0.5766 - val_accuracy: 0.6765\n",
      "Epoch 90/100\n",
      "43/47 [==========================>...] - ETA: 0s - loss: 0.5348 - accuracy: 0.7151\n",
      "Epoch 90: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5451 - accuracy: 0.7084 - val_loss: 0.5691 - val_accuracy: 0.7139\n",
      "Epoch 91/100\n",
      "47/47 [==============================] - ETA: 0s - loss: 0.5523 - accuracy: 0.7077\n",
      "Epoch 91: val_loss did not improve from 0.56359\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5523 - accuracy: 0.7077 - val_loss: 0.5769 - val_accuracy: 0.6711\n",
      "Epoch 92/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.5270 - accuracy: 0.7172\n",
      "Epoch 92: val_loss improved from 0.56359 to 0.56312, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5202 - accuracy: 0.7298 - val_loss: 0.5631 - val_accuracy: 0.7166\n",
      "Epoch 93/100\n",
      "37/47 [======================>.......] - ETA: 0s - loss: 0.5190 - accuracy: 0.7247\n",
      "Epoch 93: val_loss improved from 0.56312 to 0.55734, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5175 - accuracy: 0.7304 - val_loss: 0.5573 - val_accuracy: 0.7086\n",
      "Epoch 94/100\n",
      "26/47 [===============>..............] - ETA: 0s - loss: 0.4934 - accuracy: 0.7428\n",
      "Epoch 94: val_loss did not improve from 0.55734\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.5160 - accuracy: 0.7338 - val_loss: 0.5580 - val_accuracy: 0.6979\n",
      "Epoch 95/100\n",
      "24/47 [==============>...............] - ETA: 0s - loss: 0.5302 - accuracy: 0.7214\n",
      "Epoch 95: val_loss did not improve from 0.55734\n",
      "47/47 [==============================] - 0s 3ms/step - loss: 0.5154 - accuracy: 0.7331 - val_loss: 0.5604 - val_accuracy: 0.7246\n",
      "Epoch 96/100\n",
      "45/47 [===========================>..] - ETA: 0s - loss: 0.5316 - accuracy: 0.7250\n",
      "Epoch 96: val_loss did not improve from 0.55734\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.5312 - accuracy: 0.7217 - val_loss: 0.5654 - val_accuracy: 0.7246\n",
      "Epoch 97/100\n",
      "42/47 [=========================>....] - ETA: 0s - loss: 0.5176 - accuracy: 0.7284\n",
      "Epoch 97: val_loss improved from 0.55734 to 0.55251, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.5167 - accuracy: 0.7318 - val_loss: 0.5525 - val_accuracy: 0.7219\n",
      "Epoch 98/100\n",
      "34/47 [====================>.........] - ETA: 0s - loss: 0.5012 - accuracy: 0.7353\n",
      "Epoch 98: val_loss did not improve from 0.55251\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.4947 - accuracy: 0.7344 - val_loss: 0.5552 - val_accuracy: 0.7326\n",
      "Epoch 99/100\n",
      "39/47 [=======================>......] - ETA: 0s - loss: 0.4860 - accuracy: 0.7468\n",
      "Epoch 99: val_loss did not improve from 0.55251\n",
      "47/47 [==============================] - 0s 4ms/step - loss: 0.4932 - accuracy: 0.7512 - val_loss: 0.5674 - val_accuracy: 0.7299\n",
      "Epoch 100/100\n",
      "40/47 [========================>.....] - ETA: 0s - loss: 0.4731 - accuracy: 0.7648\n",
      "Epoch 100: val_loss improved from 0.55251 to 0.54279, saving model to saved_models\\audio_classification.hdf5\n",
      "47/47 [==============================] - 0s 5ms/step - loss: 0.4745 - accuracy: 0.7599 - val_loss: 0.5428 - val_accuracy: 0.7112\n",
      "Training completed in time:  0:00:22.296960\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from datetime import datetime \n",
    "\n",
    "num_epochs = 100\n",
    "num_batch_size = 32\n",
    "\n",
    "checkpointer = ModelCheckpoint(filepath='saved_models/audio_classification.hdf5', \n",
    "                               verbose=1, save_best_only=True)\n",
    "start = datetime.now()\n",
    "\n",
    "model.fit(X_train, y_train, batch_size=num_batch_size, epochs=num_epochs, validation_data=(X_test, y_test), callbacks=[checkpointer], verbose=1)\n",
    "\n",
    "\n",
    "duration = datetime.now() - start\n",
    "print(\"Training completed in time: \", duration)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "8aea0a5a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.5427889823913574, 0.7112299203872681]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ta=model.evaluate(X_test,y_test,verbose=0)\n",
    "ta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "fbb4844d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 125ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.24761268, 0.75238734]], dtype=float32)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def features_ex(file):\n",
    "    audio, sample_rate = librosa.load(file) \n",
    "    mfccs_features = librosa.feature.mfcc(y=audio, sr=sample_rate, n_mfcc=40)\n",
    "    mfccs_scaled_features = np.mean(mfccs_features.T,axis=0)\n",
    "    \n",
    "    return mfccs_scaled_features\n",
    "\n",
    "fp=\"D:/dataset/pathological/742-u_lhl.wav\"\n",
    "pf=features_ex(fp)\n",
    "pf=pf.reshape(1,-1)\n",
    "model.predict(pf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "37d7a86b",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('dysphonia.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "3fd9c843",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\Users\\DEEPAK\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\backend.py:6642: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from keras import layers, models\n",
    "\n",
    "model = models.Sequential()\n",
    "model.add(layers.Conv1D(32, 3, activation='relu', input_shape=(40, 1)))\n",
    "model.add(layers.MaxPooling1D(2))\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dense(256, activation='relu'))\n",
    "model.add(layers.Dropout(0.5))\n",
    "model.add(layers.Dense(2, activation='sigmoid'))\n",
    "\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "3dfaa068",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "38/38 [==============================] - 1s 12ms/step - loss: 3.0440 - accuracy: 0.5594 - val_loss: 0.9045 - val_accuracy: 0.6622\n",
      "Epoch 2/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.8594 - accuracy: 0.6288 - val_loss: 0.6194 - val_accuracy: 0.6388\n",
      "Epoch 3/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.6484 - accuracy: 0.6396 - val_loss: 0.5653 - val_accuracy: 0.7057\n",
      "Epoch 4/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.6056 - accuracy: 0.6731 - val_loss: 0.5557 - val_accuracy: 0.7124\n",
      "Epoch 5/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.5831 - accuracy: 0.7115 - val_loss: 0.5514 - val_accuracy: 0.7090\n",
      "Epoch 6/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.5815 - accuracy: 0.6998 - val_loss: 0.5550 - val_accuracy: 0.7124\n",
      "Epoch 7/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.5511 - accuracy: 0.7216 - val_loss: 0.5410 - val_accuracy: 0.7358\n",
      "Epoch 8/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.5390 - accuracy: 0.7291 - val_loss: 0.5310 - val_accuracy: 0.7157\n",
      "Epoch 9/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.5494 - accuracy: 0.7199 - val_loss: 0.5328 - val_accuracy: 0.7391\n",
      "Epoch 10/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.5289 - accuracy: 0.7358 - val_loss: 0.5225 - val_accuracy: 0.7425\n",
      "Epoch 11/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.5131 - accuracy: 0.7517 - val_loss: 0.5167 - val_accuracy: 0.7525\n",
      "Epoch 12/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4995 - accuracy: 0.7609 - val_loss: 0.4864 - val_accuracy: 0.7458\n",
      "Epoch 13/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4972 - accuracy: 0.7617 - val_loss: 0.5582 - val_accuracy: 0.7090\n",
      "Epoch 14/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.4877 - accuracy: 0.7567 - val_loss: 0.5008 - val_accuracy: 0.7458\n",
      "Epoch 15/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.4971 - accuracy: 0.7492 - val_loss: 0.4986 - val_accuracy: 0.7793\n",
      "Epoch 16/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.4682 - accuracy: 0.7851 - val_loss: 0.4830 - val_accuracy: 0.7592\n",
      "Epoch 17/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4583 - accuracy: 0.7793 - val_loss: 0.4848 - val_accuracy: 0.7659\n",
      "Epoch 18/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4495 - accuracy: 0.7784 - val_loss: 0.4939 - val_accuracy: 0.7759\n",
      "Epoch 19/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4422 - accuracy: 0.7935 - val_loss: 0.4732 - val_accuracy: 0.7625\n",
      "Epoch 20/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4251 - accuracy: 0.7968 - val_loss: 0.4827 - val_accuracy: 0.7692\n",
      "Epoch 21/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4346 - accuracy: 0.7893 - val_loss: 0.4968 - val_accuracy: 0.7625\n",
      "Epoch 22/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4088 - accuracy: 0.8161 - val_loss: 0.4843 - val_accuracy: 0.7726\n",
      "Epoch 23/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4103 - accuracy: 0.8102 - val_loss: 0.4780 - val_accuracy: 0.7826\n",
      "Epoch 24/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.3910 - accuracy: 0.8211 - val_loss: 0.4799 - val_accuracy: 0.7793\n",
      "Epoch 25/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.4024 - accuracy: 0.8119 - val_loss: 0.5030 - val_accuracy: 0.7391\n",
      "Epoch 26/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3817 - accuracy: 0.8202 - val_loss: 0.4760 - val_accuracy: 0.7893\n",
      "Epoch 27/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3670 - accuracy: 0.8311 - val_loss: 0.4624 - val_accuracy: 0.7893\n",
      "Epoch 28/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3995 - accuracy: 0.8144 - val_loss: 0.5191 - val_accuracy: 0.7492\n",
      "Epoch 29/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.4120 - accuracy: 0.8119 - val_loss: 0.4755 - val_accuracy: 0.7726\n",
      "Epoch 30/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3704 - accuracy: 0.8319 - val_loss: 0.4847 - val_accuracy: 0.7826\n",
      "Epoch 31/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3902 - accuracy: 0.8052 - val_loss: 0.4610 - val_accuracy: 0.7926\n",
      "Epoch 32/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3634 - accuracy: 0.8386 - val_loss: 0.5386 - val_accuracy: 0.7759\n",
      "Epoch 33/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.3589 - accuracy: 0.8453 - val_loss: 0.4795 - val_accuracy: 0.7759\n",
      "Epoch 34/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.3564 - accuracy: 0.8370 - val_loss: 0.5120 - val_accuracy: 0.7659\n",
      "Epoch 35/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3386 - accuracy: 0.8370 - val_loss: 0.5259 - val_accuracy: 0.7625\n",
      "Epoch 36/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.3174 - accuracy: 0.8671 - val_loss: 0.4970 - val_accuracy: 0.7960\n",
      "Epoch 37/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3286 - accuracy: 0.8503 - val_loss: 0.4908 - val_accuracy: 0.7659\n",
      "Epoch 38/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3242 - accuracy: 0.8554 - val_loss: 0.4719 - val_accuracy: 0.8060\n",
      "Epoch 39/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3131 - accuracy: 0.8595 - val_loss: 0.5002 - val_accuracy: 0.7893\n",
      "Epoch 40/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3044 - accuracy: 0.8612 - val_loss: 0.4969 - val_accuracy: 0.7860\n",
      "Epoch 41/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.3252 - accuracy: 0.8562 - val_loss: 0.4888 - val_accuracy: 0.7860\n",
      "Epoch 42/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.3231 - accuracy: 0.8537 - val_loss: 0.4738 - val_accuracy: 0.8060\n",
      "Epoch 43/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2834 - accuracy: 0.8662 - val_loss: 0.5194 - val_accuracy: 0.7893\n",
      "Epoch 44/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2766 - accuracy: 0.8704 - val_loss: 0.5132 - val_accuracy: 0.7893\n",
      "Epoch 45/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2667 - accuracy: 0.8737 - val_loss: 0.4749 - val_accuracy: 0.8194\n",
      "Epoch 46/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2978 - accuracy: 0.8637 - val_loss: 0.4786 - val_accuracy: 0.8094\n",
      "Epoch 47/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3070 - accuracy: 0.8620 - val_loss: 0.4544 - val_accuracy: 0.8194\n",
      "Epoch 48/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2736 - accuracy: 0.8729 - val_loss: 0.4799 - val_accuracy: 0.7960\n",
      "Epoch 49/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2645 - accuracy: 0.8829 - val_loss: 0.4835 - val_accuracy: 0.8261\n",
      "Epoch 50/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2730 - accuracy: 0.8671 - val_loss: 0.4583 - val_accuracy: 0.8127\n",
      "Epoch 51/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2631 - accuracy: 0.8863 - val_loss: 0.5345 - val_accuracy: 0.7960\n",
      "Epoch 52/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2677 - accuracy: 0.8846 - val_loss: 0.5231 - val_accuracy: 0.7893\n",
      "Epoch 53/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2764 - accuracy: 0.8712 - val_loss: 0.5187 - val_accuracy: 0.8094\n",
      "Epoch 54/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2454 - accuracy: 0.8863 - val_loss: 0.5254 - val_accuracy: 0.7960\n",
      "Epoch 55/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2765 - accuracy: 0.8804 - val_loss: 0.4760 - val_accuracy: 0.8395\n",
      "Epoch 56/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2630 - accuracy: 0.8796 - val_loss: 0.5140 - val_accuracy: 0.7926\n",
      "Epoch 57/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2391 - accuracy: 0.8838 - val_loss: 0.5013 - val_accuracy: 0.8127\n",
      "Epoch 58/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2273 - accuracy: 0.8938 - val_loss: 0.5253 - val_accuracy: 0.8027\n",
      "Epoch 59/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2473 - accuracy: 0.8896 - val_loss: 0.4922 - val_accuracy: 0.7993\n",
      "Epoch 60/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2511 - accuracy: 0.8821 - val_loss: 0.5230 - val_accuracy: 0.8328\n",
      "Epoch 61/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2124 - accuracy: 0.9080 - val_loss: 0.5296 - val_accuracy: 0.8094\n",
      "Epoch 62/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2460 - accuracy: 0.8880 - val_loss: 0.4966 - val_accuracy: 0.8294\n",
      "Epoch 63/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2308 - accuracy: 0.8938 - val_loss: 0.5114 - val_accuracy: 0.8161\n",
      "Epoch 64/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2492 - accuracy: 0.8788 - val_loss: 0.5199 - val_accuracy: 0.8060\n",
      "Epoch 65/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2276 - accuracy: 0.8988 - val_loss: 0.5211 - val_accuracy: 0.7993\n",
      "Epoch 66/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2730 - accuracy: 0.8821 - val_loss: 0.5340 - val_accuracy: 0.8094\n",
      "Epoch 67/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2696 - accuracy: 0.8737 - val_loss: 0.4977 - val_accuracy: 0.8094\n",
      "Epoch 68/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2495 - accuracy: 0.8813 - val_loss: 0.4921 - val_accuracy: 0.8161\n",
      "Epoch 69/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1873 - accuracy: 0.9189 - val_loss: 0.5348 - val_accuracy: 0.8161\n",
      "Epoch 70/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.1914 - accuracy: 0.9189 - val_loss: 0.5430 - val_accuracy: 0.8127\n",
      "Epoch 71/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2377 - accuracy: 0.8938 - val_loss: 0.5092 - val_accuracy: 0.8294\n",
      "Epoch 72/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.2210 - accuracy: 0.8997 - val_loss: 0.4847 - val_accuracy: 0.8495\n",
      "Epoch 73/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2251 - accuracy: 0.8946 - val_loss: 0.4828 - val_accuracy: 0.8127\n",
      "Epoch 74/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.2057 - accuracy: 0.9038 - val_loss: 0.5590 - val_accuracy: 0.8194\n",
      "Epoch 75/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1758 - accuracy: 0.9264 - val_loss: 0.5697 - val_accuracy: 0.8127\n",
      "Epoch 76/100\n",
      "38/38 [==============================] - 0s 8ms/step - loss: 0.1971 - accuracy: 0.9197 - val_loss: 0.5440 - val_accuracy: 0.8027\n",
      "Epoch 77/100\n",
      "38/38 [==============================] - 0s 11ms/step - loss: 0.1788 - accuracy: 0.9264 - val_loss: 0.6171 - val_accuracy: 0.8027\n",
      "Epoch 78/100\n",
      "38/38 [==============================] - 0s 10ms/step - loss: 0.1563 - accuracy: 0.9273 - val_loss: 0.6104 - val_accuracy: 0.8060\n",
      "Epoch 79/100\n",
      "38/38 [==============================] - 0s 10ms/step - loss: 0.1918 - accuracy: 0.9072 - val_loss: 0.5470 - val_accuracy: 0.8261\n",
      "Epoch 80/100\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.1750 - accuracy: 0.9231 - val_loss: 0.6068 - val_accuracy: 0.8127\n",
      "Epoch 81/100\n",
      "38/38 [==============================] - 0s 11ms/step - loss: 0.1580 - accuracy: 0.9273 - val_loss: 0.6505 - val_accuracy: 0.7993\n",
      "Epoch 82/100\n",
      "38/38 [==============================] - 0s 11ms/step - loss: 0.1952 - accuracy: 0.9156 - val_loss: 0.6026 - val_accuracy: 0.8328\n",
      "Epoch 83/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1768 - accuracy: 0.9214 - val_loss: 0.5661 - val_accuracy: 0.8127\n",
      "Epoch 84/100\n",
      "38/38 [==============================] - 0s 8ms/step - loss: 0.1849 - accuracy: 0.9189 - val_loss: 0.5835 - val_accuracy: 0.8227\n",
      "Epoch 85/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1759 - accuracy: 0.9197 - val_loss: 0.5643 - val_accuracy: 0.8161\n",
      "Epoch 86/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1786 - accuracy: 0.9164 - val_loss: 0.5643 - val_accuracy: 0.8161\n",
      "Epoch 87/100\n",
      "38/38 [==============================] - 0s 8ms/step - loss: 0.1839 - accuracy: 0.9164 - val_loss: 0.6508 - val_accuracy: 0.8127\n",
      "Epoch 88/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1888 - accuracy: 0.9172 - val_loss: 0.5427 - val_accuracy: 0.8027\n",
      "Epoch 89/100\n",
      "38/38 [==============================] - 0s 9ms/step - loss: 0.1893 - accuracy: 0.9156 - val_loss: 0.5481 - val_accuracy: 0.8127\n",
      "Epoch 90/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.2037 - accuracy: 0.9139 - val_loss: 0.5505 - val_accuracy: 0.8361\n",
      "Epoch 91/100\n",
      "38/38 [==============================] - 0s 8ms/step - loss: 0.1741 - accuracy: 0.9289 - val_loss: 0.5689 - val_accuracy: 0.8495\n",
      "Epoch 92/100\n",
      "38/38 [==============================] - 0s 9ms/step - loss: 0.1848 - accuracy: 0.9156 - val_loss: 0.6208 - val_accuracy: 0.8328\n",
      "Epoch 93/100\n",
      "38/38 [==============================] - 0s 8ms/step - loss: 0.1705 - accuracy: 0.9356 - val_loss: 0.5844 - val_accuracy: 0.8127\n",
      "Epoch 94/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1684 - accuracy: 0.9139 - val_loss: 0.5661 - val_accuracy: 0.8294\n",
      "Epoch 95/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1726 - accuracy: 0.9281 - val_loss: 0.5960 - val_accuracy: 0.8227\n",
      "Epoch 96/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1636 - accuracy: 0.9348 - val_loss: 0.6396 - val_accuracy: 0.8060\n",
      "Epoch 97/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1641 - accuracy: 0.9256 - val_loss: 0.5449 - val_accuracy: 0.8261\n",
      "Epoch 98/100\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 0.1583 - accuracy: 0.9323 - val_loss: 0.6591 - val_accuracy: 0.8027\n",
      "Epoch 99/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1696 - accuracy: 0.9281 - val_loss: 0.5971 - val_accuracy: 0.8294\n",
      "Epoch 100/100\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.1831 - accuracy: 0.9189 - val_loss: 0.5593 - val_accuracy: 0.8227\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "    X_train, y_train,\n",
    "    epochs=100,  # Adjust as needed\n",
    "    batch_size=32,  # Adjust as needed\n",
    "    validation_split=0.2,  # Adjust validation split as needed\n",
    "    verbose=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "6324367a",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save(\"dysphoniacnn.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "03abe1b5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 94ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.01522202, 0.9868507 ]], dtype=float32)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def features_ex(file):\n",
    "    audio, sample_rate = librosa.load(file) \n",
    "    mfccs_features = librosa.feature.mfcc(y=audio, sr=sample_rate, n_mfcc=40)\n",
    "    mfccs_scaled_features = np.mean(mfccs_features.T,axis=0)\n",
    "    \n",
    "    return mfccs_scaled_features\n",
    "\n",
    "fp=\"D:\\dataset\\pathological\\925-u_n.wav\"\n",
    "pf=features_ex(fp)\n",
    "pf=pf.reshape(1,-1)\n",
    "model.predict(pf)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}