speech to text neural network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ffcad52a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "import pickle\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as ticker\n",
    "from IPython.display import HTML\n",
    "import numpy as np\n",
    "from sklearn import preprocessing\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.io import wavfile as wav\n",
    "import scipy\n",
    "from python_speech_features import mfcc\n",
    "from python_speech_features import logfbank"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c25306a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Flatten, Activation, Dropout, LSTM\n",
    "from keras.layers import Conv2D, MaxPooling2D, GlobalMaxPooling2D\n",
    "from tensorflow.keras.optimizers import SGD\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "fc161eca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(rate,sig1) = wav.read('C:/Users/ACER/spoken_digit_dataset/free-spoken-digit-dataset-master/recordings/0_jackson_0.wav')\n",
    "N = sig1.shape[0]\n",
    "L = N / rate\n",
    "f, ax = plt.subplots()\n",
    "ax.plot(np.arange(N) / rate, sig1)\n",
    "ax.set_xlabel('Time [s]')\n",
    "ax.set_ylabel('Amplitude [Base-ten signed range (per sample)]');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "bc82bec6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x171f2aa7fa0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mfcc_feat = mfcc(sig1,rate,nfft=512)\n",
    "plt.imshow(mfcc_feat[0:20,:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "4d7120ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "soundfile = os.listdir('C:/Users/ACER/spoken_digit_dataset/free-spoken-digit-dataset-master/recordings')\n",
    "data=[]\n",
    "for i in soundfile[:2000]:\n",
    "    (rate,sig) = wav.read('C:/Users/ACER/spoken_digit_dataset/free-spoken-digit-dataset-master/recordings/'+i)\n",
    "    data.append(sig)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "360253ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "5f25bd0b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(soundfile[:2000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "dc855aa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#set the independent var\n",
    "size = 48\n",
    "X=[]\n",
    "for i in range(len(data)):\n",
    "    mfcc_feat = mfcc(data[i],rate,nfft=512)\n",
    "    mfcc_feat = np.resize(mfcc_feat, (size,13))\n",
    "    X.append(mfcc_feat)\n",
    "X = np.array(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "470a1732",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2000"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "2b52bbc7",
   "metadata": {},
   "outputs": [],
   "source": [
    "#set the target label\n",
    "y = [i[0] for i in soundfile[:2000]]\n",
    "Y = pd.get_dummies(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "b5aec2f4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0  1  2  3  4  5  6\n",
      "0  1  0  0  0  0  0  0\n",
      "1  1  0  0  0  0  0  0\n",
      "2  1  0  0  0  0  0  0\n"
     ]
    }
   ],
   "source": [
    "print(Y[0:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "4b90aed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = np.array(Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "73463b9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2000, 48, 13)\n"
     ]
    }
   ],
   "source": [
    "print(X.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "38bcbb89",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "a8d61f4b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Convolution layers\n",
    "model.add(Conv2D(8, (3, 3), activation='relu', input_shape=(size, 13,1)))\n",
    "model.add(Conv2D(8, (3, 3), activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(Dropout(0.1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "95c269cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_8 (Conv2D)           (None, 46, 11, 8)         80        \n",
      "                                                                 \n",
      " conv2d_9 (Conv2D)           (None, 44, 9, 8)          584       \n",
      "                                                                 \n",
      " max_pooling2d_4 (MaxPooling  (None, 22, 4, 8)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout_12 (Dropout)        (None, 22, 4, 8)          0         \n",
      "                                                                 \n",
      " flatten_4 (Flatten)         (None, 704)               0         \n",
      "                                                                 \n",
      " dense_12 (Dense)            (None, 64)                45120     \n",
      "                                                                 \n",
      " dropout_13 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_12 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_13 (Dense)            (None, 64)                4160      \n",
      "                                                                 \n",
      " dropout_14 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_13 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_14 (Dense)            (None, 10)                650       \n",
      "                                                                 \n",
      " activation_14 (Activation)  (None, 10)                0         \n",
      "                                                                 \n",
      " flatten_5 (Flatten)         (None, 10)                0         \n",
      "                                                                 \n",
      " dense_15 (Dense)            (None, 64)                704       \n",
      "                                                                 \n",
      " dropout_15 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_15 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_16 (Dense)            (None, 64)                4160      \n",
      "                                                                 \n",
      " dropout_16 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_16 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_17 (Dense)            (None, 3)                 195       \n",
      "                                                                 \n",
      " activation_17 (Activation)  (None, 3)                 0         \n",
      "                                                                 \n",
      " flatten_6 (Flatten)         (None, 3)                 0         \n",
      "                                                                 \n",
      " dense_18 (Dense)            (None, 64)                256       \n",
      "                                                                 \n",
      " dropout_17 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_18 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_19 (Dense)            (None, 64)                4160      \n",
      "                                                                 \n",
      " dropout_18 (Dropout)        (None, 64)                0         \n",
      "                                                                 \n",
      " activation_19 (Activation)  (None, 64)                0         \n",
      "                                                                 \n",
      " dense_20 (Dense)            (None, 10)                650       \n",
      "                                                                 \n",
      " activation_20 (Activation)  (None, 10)                0         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 60,719\n",
      "Trainable params: 60,719\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#Flattening\n",
    "model.add(Flatten(input_shape=(size, 13,1)))\n",
    "\n",
    "#1st fully connected Neural Network hidden-layer\n",
    "model.add(Dense(64))\n",
    "model.add(Dropout(0.16))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "#2nd fully connected Neural Network hidden-layer\n",
    "model.add(Dense(64))\n",
    "model.add(Dropout(0.12))\n",
    "model.add(Activation('relu'))\n",
    "\n",
    "#Output layer\n",
    "model.add(Dense(10))\n",
    "model.add(Activation('softmax'))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "950532c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\optimizer_v2\\gradient_descent.py:102: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
      "  super(SGD, self).__init__(name, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "model.compile(loss='binary_crossentropy',\n",
    "optimizer=sgd,\n",
    "metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "9db8f83a",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.25)\n",
    "x_train = x_train.reshape(-1, size, 13, 1)\n",
    "x_test = x_test.reshape(-1, size, 13, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "b5a5f0dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/18\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "in user code:\n\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 878, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 867, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 860, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 809, in train_step\n        loss = self.compiled_loss(\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\compile_utils.py\", line 201, in __call__\n        loss_value = loss_obj(y_t, y_p, sample_weight=sw)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 141, in __call__\n        losses = call_fn(y_true, y_pred)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 245, in call  **\n        return ag_fn(y_true, y_pred, **self._fn_kwargs)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 1664, in categorical_crossentropy\n        return backend.categorical_crossentropy(\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\backend.py\", line 4994, in categorical_crossentropy\n        target.shape.assert_is_compatible_with(output.shape)\n\n    ValueError: Shapes (None, 7) and (None, 3) are incompatible\n",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-133-9921959aee41>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m history = model.fit(\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mx_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0my_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[0mepochs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m18\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m32\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\keras\\utils\\traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     65\u001b[0m     \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint: disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 67\u001b[1;33m       \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     68\u001b[0m     \u001b[1;32mfinally\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     69\u001b[0m       \u001b[1;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\tensorflow\\python\\framework\\func_graph.py\u001b[0m in \u001b[0;36mautograph_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m   1127\u001b[0m           \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint:disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1128\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"ag_error_metadata\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1129\u001b[1;33m               \u001b[1;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1130\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1131\u001b[0m               \u001b[1;32mraise\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: in user code:\n\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 878, in train_function  *\n        return step_function(self, iterator)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 867, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 860, in run_step  **\n        outputs = model.train_step(data)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\training.py\", line 809, in train_step\n        loss = self.compiled_loss(\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\engine\\compile_utils.py\", line 201, in __call__\n        loss_value = loss_obj(y_t, y_p, sample_weight=sw)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 141, in __call__\n        losses = call_fn(y_true, y_pred)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 245, in call  **\n        return ag_fn(y_true, y_pred, **self._fn_kwargs)\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\losses.py\", line 1664, in categorical_crossentropy\n        return backend.categorical_crossentropy(\n    File \"C:\\Users\\ACER\\anaconda3\\lib\\site-packages\\keras\\backend.py\", line 4994, in categorical_crossentropy\n        target.shape.assert_is_compatible_with(output.shape)\n\n    ValueError: Shapes (None, 7) and (None, 3) are incompatible\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "x_train,\n",
    "y_train,\n",
    "epochs=18,\n",
    "batch_size=32,\n",
    "validation_split=0.2,\n",
    "shuffle=True\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5ab00589",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}