{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.039681,
     "end_time": "2022-05-13T08:04:34.931983",
     "exception": false,
     "start_time": "2022-05-13T08:04:34.892302",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Example of usage "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.038012,
     "end_time": "2022-05-13T08:04:35.009857",
     "exception": false,
     "start_time": "2022-05-13T08:04:34.971845",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Install library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:35.095343Z",
     "iopub.status.busy": "2022-05-13T08:04:35.094511Z",
     "iopub.status.idle": "2022-05-13T08:04:46.535492Z",
     "shell.execute_reply": "2022-05-13T08:04:46.534817Z",
     "shell.execute_reply.started": "2022-05-13T07:51:56.319388Z"
    },
    "papermill": {
     "duration": 11.48471,
     "end_time": "2022-05-13T08:04:46.535614",
     "exception": false,
     "start_time": "2022-05-13T08:04:35.050904",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting praat-textgrids\n",
      "  Downloading praat_textgrids-1.4.0-py3-none-any.whl (25 kB)\n",
      "Installing collected packages: praat-textgrids\n",
      "Successfully installed praat-textgrids-1.4.0\n"
     ]
    }
   ],
   "source": [
    "# !pip install praat-textgrids"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040382,
     "end_time": "2022-05-13T08:04:46.617320",
     "exception": false,
     "start_time": "2022-05-13T08:04:46.576938",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Load data from kaggle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:46.702989Z",
     "iopub.status.busy": "2022-05-13T08:04:46.701906Z",
     "iopub.status.idle": "2022-05-13T08:04:46.706623Z",
     "shell.execute_reply": "2022-05-13T08:04:46.707280Z",
     "shell.execute_reply.started": "2022-05-13T07:52:06.686417Z"
    },
    "papermill": {
     "duration": 0.049187,
     "end_time": "2022-05-13T08:04:46.707426",
     "exception": false,
     "start_time": "2022-05-13T08:04:46.658239",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# This Python 3 environment comes with many helpful analytics libraries installed\n",
    "# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n",
    "# For example, here's several helpful packages to load\n",
    "\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "\n",
    "# Input data files are available in the read-only \"../input/\" directory\n",
    "# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040399,
     "end_time": "2022-05-13T08:04:46.788727",
     "exception": false,
     "start_time": "2022-05-13T08:04:46.748328",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Define the function extracting the ground truth labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040318,
     "end_time": "2022-05-13T08:04:46.869676",
     "exception": false,
     "start_time": "2022-05-13T08:04:46.829358",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "It's important for this kind of tasks to perform short time analysis on the signal, so it needs to assign the lables (SPEECH/NONSPEECH) to very little portions of the signal. I decide to split the data into portion of 30 milliseconds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:46.962749Z",
     "iopub.status.busy": "2022-05-13T08:04:46.961682Z",
     "iopub.status.idle": "2022-05-13T08:04:46.975714Z",
     "shell.execute_reply": "2022-05-13T08:04:46.976361Z",
     "shell.execute_reply.started": "2022-05-13T07:52:06.693467Z"
    },
    "papermill": {
     "duration": 0.059975,
     "end_time": "2022-05-13T08:04:46.976532",
     "exception": false,
     "start_time": "2022-05-13T08:04:46.916557",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import textgrids\n",
    "\n",
    "FRAME_DURATION = 20 # 30 msec\n",
    "OVERLAP_RATE = 0.0 # frames don't overlap\n",
    "\n",
    "def readFile(path : str, frame_len_ms = 20 ):\n",
    "    '''\n",
    "    Read the file and return the list of SPEECH/NONSPEECH labels for each frame\n",
    "    '''\n",
    "        \n",
    "    labeled_list  = []\n",
    "    grid = textgrids.TextGrid(path)\n",
    "\n",
    "    for interval in grid['silences']:\n",
    "        label = int(interval.text)\n",
    "\n",
    "        dur = interval.dur\n",
    "        dur_msec = dur * 1000 # sec -> msec\n",
    "        num_frames = int(round(dur_msec /frame_len_ms)) # the audio is divided into 30 msec frames\n",
    "        print(dur_msec)\n",
    "        for i in range(num_frames):\n",
    "            \n",
    "            labeled_list.append(label)\n",
    "\n",
    "    return labeled_list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040297,
     "end_time": "2022-05-13T08:04:47.058246",
     "exception": false,
     "start_time": "2022-05-13T08:04:47.017949",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Load a file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:47.144357Z",
     "iopub.status.busy": "2022-05-13T08:04:47.143349Z",
     "iopub.status.idle": "2022-05-13T08:04:57.883297Z",
     "shell.execute_reply": "2022-05-13T08:04:57.882316Z",
     "shell.execute_reply.started": "2022-05-13T07:52:06.711722Z"
    },
    "papermill": {
     "duration": 10.783827,
     "end_time": "2022-05-13T08:04:57.883437",
     "exception": false,
     "start_time": "2022-05-13T08:04:47.099610",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "170.40362811791388\n",
      "215.99999999999997\n",
      "136.0\n",
      "1296.0\n",
      "167.99999999999991\n",
      "120.00000000000011\n",
      "152.00000000000014\n",
      "199.99999999999974\n",
      "466.4036281179138\n"
     ]
    }
   ],
   "source": [
    "\n",
    "import librosa\n",
    "\n",
    "root ='/Female/TIMIT/SA2'\n",
    "annotation_path = \"Annotation/Female/TMIT/SI2220.TextGrid\"\n",
    "audio_path = \"Audio/Female/TMIT/SI2220.wav\"\n",
    "\n",
    "# read annotaion\n",
    "label_list = readFile(annotation_path)\n",
    "\n",
    "# read wav file\n",
    "data, fs = librosa.load(audio_path, sr = 16_000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16000"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(147, (46797,), 146.240625)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(label_list), data.shape, data.shape[0] /16 / FRAME_DURATION"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.042802,
     "end_time": "2022-05-13T08:04:57.969432",
     "exception": false,
     "start_time": "2022-05-13T08:04:57.926630",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Plot signal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.043151,
     "end_time": "2022-05-13T08:04:58.056024",
     "exception": false,
     "start_time": "2022-05-13T08:04:58.012873",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Preparing the variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:58.151654Z",
     "iopub.status.busy": "2022-05-13T08:04:58.150961Z",
     "iopub.status.idle": "2022-05-13T08:04:58.154631Z",
     "shell.execute_reply": "2022-05-13T08:04:58.153991Z",
     "shell.execute_reply.started": "2022-05-13T07:52:16.026565Z"
    },
    "papermill": {
     "duration": 0.054958,
     "end_time": "2022-05-13T08:04:58.154746",
     "exception": false,
     "start_time": "2022-05-13T08:04:58.099788",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# define time axis\n",
    "Ns = len(data)  # number of sample\n",
    "Ts = 1 / fs  # sampling period\n",
    "t = np.arange(Ns) * 1000 * Ts  # time axis\n",
    "\n",
    "\n",
    "shift = 1 - OVERLAP_RATE\n",
    "frame_length = int(np.floor(FRAME_DURATION * fs / 1000)) # frame length in sample\n",
    "frame_shift = round(frame_length * shift)# frame shift in sample\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:58.250902Z",
     "iopub.status.busy": "2022-05-13T08:04:58.250199Z",
     "iopub.status.idle": "2022-05-13T08:04:58.829681Z",
     "shell.execute_reply": "2022-05-13T08:04:58.828681Z",
     "shell.execute_reply.started": "2022-05-13T07:52:16.039612Z"
    },
    "papermill": {
     "duration": 0.631293,
     "end_time": "2022-05-13T08:04:58.829805",
     "exception": false,
     "start_time": "2022-05-13T08:04:58.198512",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "figure = plt.Figure(figsize=(10, 7), dpi=85)\n",
    "plt.plot(t, data)\n",
    "\n",
    "for i, frame_labeled in enumerate(label_list):\n",
    "    idx = i * frame_shift\n",
    "    if (frame_labeled == 1):\n",
    "        plt.axvspan(xmin= t[idx], xmax=t[idx + frame_length-1], ymin=-1000, ymax=1000, alpha=0.4, zorder=-100, facecolor='g', label='Speech')\n",
    "\n",
    "plt.title(\"Ground truth labels\")\n",
    "plt.legend(['Signal', 'Speech'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:58.923831Z",
     "iopub.status.busy": "2022-05-13T08:04:58.922763Z",
     "iopub.status.idle": "2022-05-13T08:04:58.930309Z",
     "shell.execute_reply": "2022-05-13T08:04:58.929625Z",
     "shell.execute_reply.started": "2022-05-13T07:52:16.641220Z"
    },
    "papermill": {
     "duration": 0.056337,
     "end_time": "2022-05-13T08:04:58.930437",
     "exception": false,
     "start_time": "2022-05-13T08:04:58.874100",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n",
      "64492\n",
      "64492\n"
     ]
    }
   ],
   "source": [
    "print(len(label_list))\n",
    "print(len(t))\n",
    "print(len(data))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.044144,
     "end_time": "2022-05-13T08:04:59.018840",
     "exception": false,
     "start_time": "2022-05-13T08:04:58.974696",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "The green parts indicates the frames where an human speech is detected."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.043908,
     "end_time": "2022-05-13T08:04:59.107062",
     "exception": false,
     "start_time": "2022-05-13T08:04:59.063154",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Voice Activity Detection:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.044314,
     "end_time": "2022-05-13T08:04:59.195759",
     "exception": false,
     "start_time": "2022-05-13T08:04:59.151445",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Preparation:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.044184,
     "end_time": "2022-05-13T08:04:59.284427",
     "exception": false,
     "start_time": "2022-05-13T08:04:59.240243",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Libraries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:04:59.385013Z",
     "iopub.status.busy": "2022-05-13T08:04:59.384003Z",
     "iopub.status.idle": "2022-05-13T08:05:09.062392Z",
     "shell.execute_reply": "2022-05-13T08:05:09.063042Z",
     "shell.execute_reply.started": "2022-05-13T07:52:16.648088Z"
    },
    "papermill": {
     "duration": 9.734754,
     "end_time": "2022-05-13T08:05:09.063238",
     "exception": false,
     "start_time": "2022-05-13T08:04:59.328484",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting python_speech_features\r\n",
      "  Downloading python_speech_features-0.6.tar.gz (5.6 kB)\r\n",
      "Building wheels for collected packages: python-speech-features\r\n",
      "  Building wheel for python-speech-features (setup.py) ... \u001b[?25l-\b \b\\\b \bdone\r\n",
      "\u001b[?25h  Created wheel for python-speech-features: filename=python_speech_features-0.6-py3-none-any.whl size=5888 sha256=8ed4e6ea245fec3d7e978b02a6ed1aa23dea1884e92ab77a51e2b88613982903\r\n",
      "  Stored in directory: /root/.cache/pip/wheels/b0/0e/94/28cd6afa3cd5998a63eef99fe31777acd7d758f59cf24839eb\r\n",
      "Successfully built python-speech-features\r\n",
      "Installing collected packages: python-speech-features\r\n",
      "Successfully installed python-speech-features-0.6\r\n",
      "\u001b[33mWARNING: You are using pip version 20.2.4; however, version 22.1 is available.\r\n",
      "You should consider upgrading via the '/opt/conda/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\r\n"
     ]
    }
   ],
   "source": [
    "!pip install python_speech_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:09.169402Z",
     "iopub.status.busy": "2022-05-13T08:05:09.168384Z",
     "iopub.status.idle": "2022-05-13T08:05:15.198563Z",
     "shell.execute_reply": "2022-05-13T08:05:15.197858Z",
     "shell.execute_reply.started": "2022-05-13T07:52:25.056106Z"
    },
    "papermill": {
     "duration": 6.085504,
     "end_time": "2022-05-13T08:05:15.198681",
     "exception": false,
     "start_time": "2022-05-13T08:05:09.113177",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os # Working with directories\n",
    "import python_speech_features # For exctracting features for deep learning\n",
    "from tqdm import tqdm # Progress meter\n",
    "from sklearn import model_selection, preprocessing, metrics # Preparation data\n",
    "from tensorflow.keras import models, layers\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.047604,
     "end_time": "2022-05-13T08:05:15.294604",
     "exception": false,
     "start_time": "2022-05-13T08:05:15.247000",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Load all files:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.047212,
     "end_time": "2022-05-13T08:05:15.389813",
     "exception": false,
     "start_time": "2022-05-13T08:05:15.342601",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's redefine funtion that has already been implemented above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:15.493715Z",
     "iopub.status.busy": "2022-05-13T08:05:15.493023Z",
     "iopub.status.idle": "2022-05-13T08:05:15.496513Z",
     "shell.execute_reply": "2022-05-13T08:05:15.495828Z",
     "shell.execute_reply.started": "2022-05-13T07:52:31.106449Z"
    },
    "papermill": {
     "duration": 0.058893,
     "end_time": "2022-05-13T08:05:15.496636",
     "exception": false,
     "start_time": "2022-05-13T08:05:15.437743",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Function for reading labels from .TextGrig file:\n",
    "def readLabels(path, sample_rate):\n",
    "    '''\n",
    "    Read the file and return the list of SPEECH/NONSPEECH labels for each frame\n",
    "    '''\n",
    "        \n",
    "    labeled_list  = []\n",
    "    grid = textgrids.TextGrid(path)\n",
    "\n",
    "    for interval in grid['silences']:\n",
    "        label = int(interval.text)\n",
    "\n",
    "        dur = interval.dur\n",
    "        dur_samples = int(np.round(dur * sample_rate)) # sec -> num of samples\n",
    "        \n",
    "        for i in range(dur_samples):\n",
    "            labeled_list.append(label)\n",
    "\n",
    "    return labeled_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:15.599250Z",
     "iopub.status.busy": "2022-05-13T08:05:15.598540Z",
     "iopub.status.idle": "2022-05-13T08:05:15.600823Z",
     "shell.execute_reply": "2022-05-13T08:05:15.601317Z",
     "shell.execute_reply.started": "2022-05-13T07:52:31.114449Z"
    },
    "papermill": {
     "duration": 0.056964,
     "end_time": "2022-05-13T08:05:15.601484",
     "exception": false,
     "start_time": "2022-05-13T08:05:15.544520",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Function for getting all files in directories and sub-directories with definite extension:\n",
    "def getFiles(path, extension):\n",
    "    list_paths = list()\n",
    "    for root, dirs, files in os.walk(path):\n",
    "        for file in files:\n",
    "            if(file.endswith(extension)):\n",
    "                list_paths.append(os.path.join(root, file))\n",
    "    return list_paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:15.701870Z",
     "iopub.status.busy": "2022-05-13T08:05:15.701158Z",
     "iopub.status.idle": "2022-05-13T08:05:16.116961Z",
     "shell.execute_reply": "2022-05-13T08:05:16.116393Z",
     "shell.execute_reply.started": "2022-05-13T07:52:31.126692Z"
    },
    "papermill": {
     "duration": 0.46775,
     "end_time": "2022-05-13T08:05:16.117089",
     "exception": false,
     "start_time": "2022-05-13T08:05:15.649339",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of files: 719\n"
     ]
    }
   ],
   "source": [
    "annotation_path = '/kaggle/input/speech-activity-detection-datasets/Data/Annotation/'\n",
    "annotation_extension = '.TextGrid'\n",
    "audio_path = '/kaggle/input/speech-activity-detection-datasets/Data/Audio/'\n",
    "audio_extension = '.wav'\n",
    "\n",
    "annotation_files = getFiles(path=annotation_path, extension=annotation_extension)\n",
    "audio_files = getFiles(path=audio_path, extension=audio_extension)\n",
    "annotation_files = sorted(annotation_files)\n",
    "audio_files = sorted(audio_files)\n",
    "\n",
    "print('Number of files:', len(audio_files))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.047839,
     "end_time": "2022-05-13T08:05:16.213382",
     "exception": false,
     "start_time": "2022-05-13T08:05:16.165543",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Processing files:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.04747,
     "end_time": "2022-05-13T08:05:16.308736",
     "exception": false,
     "start_time": "2022-05-13T08:05:16.261266",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Algorithm based on a article \"A HYBRID CNN-BILSTM VOICE ACTIVITY DETECTOR\"\n",
    "\n",
    "In preprocessing part we need extract Mel filter bank energies from signal as a features. Features are extracted every 10 ms using a 25 ms window. We will use 32 Mel log energies and the log energy of frame. And after extracting, we need form sequences of 32 × 32 spectrogram imagesas input features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:16.411250Z",
     "iopub.status.busy": "2022-05-13T08:05:16.410529Z",
     "iopub.status.idle": "2022-05-13T08:05:16.412935Z",
     "shell.execute_reply": "2022-05-13T08:05:16.413506Z",
     "shell.execute_reply.started": "2022-05-13T07:52:31.389992Z"
    },
    "papermill": {
     "duration": 0.056498,
     "end_time": "2022-05-13T08:05:16.413654",
     "exception": false,
     "start_time": "2022-05-13T08:05:16.357156",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Set params for model:\n",
    "preemphasis_coef = 0.97 # Coefficient for pre-processing filter\n",
    "frame_length = 0.025 # Window length in sec\n",
    "frame_step = 0.01 # Length of step in sec\n",
    "num_nfft = 512 # Point for FFT\n",
    "num_features = 32 # Number of Mel filters\n",
    "n_frames = 32 # Number of frames for uniting in image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:05:16.514498Z",
     "iopub.status.busy": "2022-05-13T08:05:16.513743Z",
     "iopub.status.idle": "2022-05-13T08:08:12.307632Z",
     "shell.execute_reply": "2022-05-13T08:08:12.306909Z",
     "shell.execute_reply.started": "2022-05-13T07:52:31.396614Z"
    },
    "papermill": {
     "duration": 175.844992,
     "end_time": "2022-05-13T08:08:12.307760",
     "exception": false,
     "start_time": "2022-05-13T08:05:16.462768",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 719/719 [02:55<00:00,  4.09it/s]\n"
     ]
    }
   ],
   "source": [
    "# Extraction features for each file:\n",
    "dataset = list()\n",
    "for i in tqdm(range(len(audio_files))):\n",
    "    sig, sample_rate = librosa.load(audio_files[i])\n",
    "    markers = readLabels(path=annotation_files[i], sample_rate=sample_rate)\n",
    "    \n",
    "    # Extract logfbank features:\n",
    "    features_logfbank = python_speech_features.base.logfbank(signal=sig, samplerate=sample_rate, winlen=frame_length, winstep=frame_step, nfilt=num_features, \n",
    "                                                                        nfft=num_nfft, lowfreq=0, highfreq=None, preemph=preemphasis_coef)\n",
    "    \n",
    "    # Reshape labels for each group of features:\n",
    "    markers_of_frames = python_speech_features.sigproc.framesig(sig=markers, frame_len=frame_length * sample_rate, frame_step=frame_step * sample_rate, \n",
    "                                                                winfunc=np.ones)\n",
    "    \n",
    "    # For every frame calc label:\n",
    "    marker_per_frame = np.zeros(markers_of_frames.shape[0])\n",
    "    marker_per_frame = np.array([1 if np.sum(markers_of_frames[j], axis=0) > markers_of_frames.shape[0] / 2 else 0 for j in range(markers_of_frames.shape[0])])\n",
    "    \n",
    "    spectrogram_image = np.zeros((n_frames, n_frames))\n",
    "    for j in range(int(np.floor(features_logfbank.shape[0] / n_frames))):\n",
    "        spectrogram_image = features_logfbank[j * n_frames:(j + 1) * n_frames]\n",
    "        label_spectrogram_image = 1 if np.sum(marker_per_frame[j * n_frames:(j + 1) * n_frames]) > n_frames / 2 else 0\n",
    "        dataset.append((label_spectrogram_image, spectrogram_image))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:12.841691Z",
     "iopub.status.busy": "2022-05-13T08:08:12.840371Z",
     "iopub.status.idle": "2022-05-13T08:08:12.845355Z",
     "shell.execute_reply": "2022-05-13T08:08:12.844557Z",
     "shell.execute_reply.started": "2022-05-13T07:55:12.722281Z"
    },
    "papermill": {
     "duration": 0.278,
     "end_time": "2022-05-13T08:08:12.845487",
     "exception": false,
     "start_time": "2022-05-13T08:08:12.567487",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(262, 551)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "markers_of_frames.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-12-17T08:36:32.237686Z",
     "iopub.status.busy": "2021-12-17T08:36:32.237225Z",
     "iopub.status.idle": "2021-12-17T08:36:32.244073Z",
     "shell.execute_reply": "2021-12-17T08:36:32.24307Z",
     "shell.execute_reply.started": "2021-12-17T08:36:32.237645Z"
    },
    "papermill": {
     "duration": 0.269002,
     "end_time": "2022-05-13T08:08:13.382406",
     "exception": false,
     "start_time": "2022-05-13T08:08:13.113404",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Build and train the model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.267675,
     "end_time": "2022-05-13T08:08:13.912401",
     "exception": false,
     "start_time": "2022-05-13T08:08:13.644726",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We will use model shown in the picture below"
   ]
  },
  {
   "attachments": {
    "0bcd572e-6e05-497c-bc62-672dd3fdeaa3.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.303722,
     "end_time": "2022-05-13T08:08:14.469362",
     "exception": false,
     "start_time": "2022-05-13T08:08:14.165640",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "![image.png](attachment:0bcd572e-6e05-497c-bc62-672dd3fdeaa3.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.265886,
     "end_time": "2022-05-13T08:08:14.985759",
     "exception": false,
     "start_time": "2022-05-13T08:08:14.719873",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Preparation data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:15.505814Z",
     "iopub.status.busy": "2022-05-13T08:08:15.505008Z",
     "iopub.status.idle": "2022-05-13T08:08:15.786287Z",
     "shell.execute_reply": "2022-05-13T08:08:15.785710Z",
     "shell.execute_reply.started": "2022-05-13T07:55:12.733025Z"
    },
    "papermill": {
     "duration": 0.535144,
     "end_time": "2022-05-13T08:08:15.786412",
     "exception": false,
     "start_time": "2022-05-13T08:08:15.251268",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6342, 32, 32) (3125, 32, 32) (6342,) (3125,)\n"
     ]
    }
   ],
   "source": [
    "# Split dataset on train and test:\n",
    "X = list()\n",
    "y = list()\n",
    "for i in range(len(dataset)):\n",
    "    X.append(dataset[i][1])\n",
    "    y.append(dataset[i][0])\n",
    "    \n",
    "X = np.array(X)\n",
    "y = np.array(y)\n",
    "\n",
    "X_train, X_test, y_train, y_test = model_selection.train_test_split(X, y, test_size=0.33, shuffle=True, random_state=1)\n",
    "\n",
    "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)\n",
    "\n",
    "# Reshaping for scaling:\n",
    "X_train = X_train.reshape(X_train.shape[0], X_train.shape[1] * X_train.shape[2])\n",
    "X_test = X_test.reshape(X_test.shape[0], X_test.shape[1] * X_test.shape[2])\n",
    "    \n",
    "# Scale data:\n",
    "scaler = preprocessing.StandardScaler()\n",
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)\n",
    "\n",
    "# And reshape back:\n",
    "X_train = X_train.reshape(X_train.shape[0], n_frames, n_frames)\n",
    "X_test = X_test.reshape(X_test.shape[0], n_frames, n_frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:16.292716Z",
     "iopub.status.busy": "2022-05-13T08:08:16.292034Z",
     "iopub.status.idle": "2022-05-13T08:08:16.300436Z",
     "shell.execute_reply": "2022-05-13T08:08:16.300969Z",
     "shell.execute_reply.started": "2022-05-13T07:55:13.006525Z"
    },
    "papermill": {
     "duration": 0.263772,
     "end_time": "2022-05-13T08:08:16.301132",
     "exception": false,
     "start_time": "2022-05-13T08:08:16.037360",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Encoding label:\n",
    "y_train = pd.get_dummies(y_train)\n",
    "y_test = pd.get_dummies(y_test)\n",
    "y_train = np.array(y_train)\n",
    "y_test = np.array(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:16.817891Z",
     "iopub.status.busy": "2022-05-13T08:08:16.816822Z",
     "iopub.status.idle": "2022-05-13T08:08:16.820352Z",
     "shell.execute_reply": "2022-05-13T08:08:16.819680Z",
     "shell.execute_reply.started": "2022-05-13T07:55:13.022597Z"
    },
    "papermill": {
     "duration": 0.270093,
     "end_time": "2022-05-13T08:08:16.820485",
     "exception": false,
     "start_time": "2022-05-13T08:08:16.550392",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Reshape data for convolution layer:\n",
    "stride = int(15)\n",
    "\n",
    "X_train_reshaped = X_train[:int(np.floor(X_train.shape[0] / stride) * stride)]\n",
    "X_test_reshaped = X_test[:int(np.floor(X_test.shape[0] / stride) * stride)]\n",
    "\n",
    "y_train_reshaped = y_train[:int(np.floor(X_train.shape[0] / stride) * stride)]\n",
    "y_test_reshaped = y_test[:int(np.floor(X_test.shape[0] / stride) * stride)]\n",
    "\n",
    "X_train_reshaped = X_train_reshaped.reshape((int(X_train_reshaped.shape[0] / stride), stride, n_frames, n_frames, 1))\n",
    "X_test_reshaped = X_test_reshaped.reshape((int(X_test_reshaped.shape[0] / stride), stride, n_frames, n_frames, 1))\n",
    "\n",
    "y_train_reshaped = y_train_reshaped.reshape((int(y_train_reshaped.shape[0] / stride), stride, y_train[-1].shape[0]))\n",
    "y_test_reshaped = y_test_reshaped.reshape((int(y_test_reshaped.shape[0] / stride), stride, y_test[-1].shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:17.356535Z",
     "iopub.status.busy": "2022-05-13T08:08:17.355692Z",
     "iopub.status.idle": "2022-05-13T08:08:17.358414Z",
     "shell.execute_reply": "2022-05-13T08:08:17.358925Z",
     "shell.execute_reply.started": "2022-05-13T07:55:13.038309Z"
    },
    "papermill": {
     "duration": 0.276313,
     "end_time": "2022-05-13T08:08:17.359071",
     "exception": false,
     "start_time": "2022-05-13T08:08:17.082758",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(422, 15, 32, 32, 1)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_reshaped.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.249199,
     "end_time": "2022-05-13T08:08:17.860469",
     "exception": false,
     "start_time": "2022-05-13T08:08:17.611270",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Build model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:18.366791Z",
     "iopub.status.busy": "2022-05-13T08:08:18.366042Z",
     "iopub.status.idle": "2022-05-13T08:08:19.217207Z",
     "shell.execute_reply": "2022-05-13T08:08:19.217763Z",
     "shell.execute_reply.started": "2022-05-13T07:55:13.054195Z"
    },
    "papermill": {
     "duration": 1.103662,
     "end_time": "2022-05-13T08:08:19.217957",
     "exception": false,
     "start_time": "2022-05-13T08:08:18.114295",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "model = models.Sequential()\n",
    "model.add(layers.TimeDistributed(layers.Conv2D(64, (5, 5), activation='elu'), input_shape=(stride, n_frames, n_frames, 1)))\n",
    "model.add(layers.TimeDistributed(layers.MaxPooling2D((2, 2))))\n",
    "model.add(layers.TimeDistributed(layers.Conv2D(128, (3, 3), activation='elu')))\n",
    "model.add(layers.TimeDistributed(layers.MaxPooling2D((2, 2))))\n",
    "model.add(layers.TimeDistributed(layers.Conv2D(128, (3, 3), activation='elu')))\n",
    "model.add(layers.TimeDistributed(layers.MaxPooling2D((2, 2))))\n",
    "model.add(layers.TimeDistributed(layers.Flatten()))\n",
    "model.add(layers.TimeDistributed(layers.Dense(64, activation='elu')))\n",
    "model.add(layers.Dropout(0.5))\n",
    "model.add(layers.Bidirectional(layers.LSTM(128, return_sequences=True)))\n",
    "model.add(layers.Dropout(0.5))\n",
    "model.add(layers.TimeDistributed(layers.Dense(y_train[-1].shape[0], activation='softmax')))\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:19.760513Z",
     "iopub.status.busy": "2022-05-13T08:08:19.759784Z",
     "iopub.status.idle": "2022-05-13T08:08:20.513580Z",
     "shell.execute_reply": "2022-05-13T08:08:20.514180Z",
     "shell.execute_reply.started": "2022-05-13T07:55:13.820378Z"
    },
    "papermill": {
     "duration": 1.027048,
     "end_time": "2022-05-13T08:08:20.514339",
     "exception": false,
     "start_time": "2022-05-13T08:08:19.487291",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "plot_model(model, 'model.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:08:21.062075Z",
     "iopub.status.busy": "2022-05-13T08:08:21.061266Z",
     "iopub.status.idle": "2022-05-13T08:13:12.783017Z",
     "shell.execute_reply": "2022-05-13T08:13:12.781826Z",
     "shell.execute_reply.started": "2022-05-13T07:55:14.479038Z"
    },
    "papermill": {
     "duration": 292.000009,
     "end_time": "2022-05-13T08:13:12.783387",
     "exception": false,
     "start_time": "2022-05-13T08:08:20.783378",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "14/14 [==============================] - 11s 789ms/step - loss: 0.4674 - accuracy: 0.7883 - val_loss: 0.3314 - val_accuracy: 0.8641\n",
      "Epoch 2/25\n",
      "14/14 [==============================] - 11s 757ms/step - loss: 0.3640 - accuracy: 0.8504 - val_loss: 0.3198 - val_accuracy: 0.8673\n",
      "Epoch 3/25\n",
      "14/14 [==============================] - 10s 737ms/step - loss: 0.3313 - accuracy: 0.8607 - val_loss: 0.3003 - val_accuracy: 0.8731\n",
      "Epoch 4/25\n",
      "14/14 [==============================] - 10s 725ms/step - loss: 0.3161 - accuracy: 0.8668 - val_loss: 0.2875 - val_accuracy: 0.8744\n",
      "Epoch 5/25\n",
      "14/14 [==============================] - 10s 750ms/step - loss: 0.2956 - accuracy: 0.8706 - val_loss: 0.2719 - val_accuracy: 0.8782\n",
      "Epoch 6/25\n",
      "14/14 [==============================] - 10s 734ms/step - loss: 0.2941 - accuracy: 0.8720 - val_loss: 0.2700 - val_accuracy: 0.8760\n",
      "Epoch 7/25\n",
      "14/14 [==============================] - 10s 722ms/step - loss: 0.2888 - accuracy: 0.8722 - val_loss: 0.2878 - val_accuracy: 0.8667\n",
      "Epoch 8/25\n",
      "14/14 [==============================] - 11s 762ms/step - loss: 0.2847 - accuracy: 0.8727 - val_loss: 0.2605 - val_accuracy: 0.8696\n",
      "Epoch 9/25\n",
      "14/14 [==============================] - 10s 728ms/step - loss: 0.2721 - accuracy: 0.8788 - val_loss: 0.2530 - val_accuracy: 0.8837\n",
      "Epoch 10/25\n",
      "14/14 [==============================] - 10s 730ms/step - loss: 0.2626 - accuracy: 0.8823 - val_loss: 0.2530 - val_accuracy: 0.8808\n",
      "Epoch 11/25\n",
      "14/14 [==============================] - 11s 762ms/step - loss: 0.2596 - accuracy: 0.8790 - val_loss: 0.2385 - val_accuracy: 0.8859\n",
      "Epoch 12/25\n",
      "14/14 [==============================] - 10s 728ms/step - loss: 0.2466 - accuracy: 0.8863 - val_loss: 0.2659 - val_accuracy: 0.8869\n",
      "Epoch 13/25\n",
      "14/14 [==============================] - 10s 725ms/step - loss: 0.2564 - accuracy: 0.8844 - val_loss: 0.2491 - val_accuracy: 0.8827\n",
      "Epoch 14/25\n",
      "14/14 [==============================] - 11s 764ms/step - loss: 0.2402 - accuracy: 0.8945 - val_loss: 0.2391 - val_accuracy: 0.8929\n",
      "Epoch 15/25\n",
      "14/14 [==============================] - 10s 735ms/step - loss: 0.2251 - accuracy: 0.9030 - val_loss: 0.2323 - val_accuracy: 0.8933\n",
      "Epoch 16/25\n",
      "14/14 [==============================] - 10s 723ms/step - loss: 0.2175 - accuracy: 0.9046 - val_loss: 0.2339 - val_accuracy: 0.8942\n",
      "Epoch 17/25\n",
      "14/14 [==============================] - 11s 766ms/step - loss: 0.2258 - accuracy: 0.8989 - val_loss: 0.2392 - val_accuracy: 0.8990\n",
      "Epoch 18/25\n",
      "14/14 [==============================] - 10s 724ms/step - loss: 0.2112 - accuracy: 0.9046 - val_loss: 0.2285 - val_accuracy: 0.8978\n",
      "Epoch 19/25\n",
      "14/14 [==============================] - 10s 740ms/step - loss: 0.1957 - accuracy: 0.9142 - val_loss: 0.3079 - val_accuracy: 0.8567\n",
      "Epoch 20/25\n",
      "14/14 [==============================] - 19s 1s/step - loss: 0.2052 - accuracy: 0.9084 - val_loss: 0.2541 - val_accuracy: 0.8792\n",
      "Epoch 21/25\n",
      "14/14 [==============================] - 10s 736ms/step - loss: 0.1944 - accuracy: 0.9145 - val_loss: 0.2364 - val_accuracy: 0.8974\n",
      "Epoch 22/25\n",
      "14/14 [==============================] - 11s 766ms/step - loss: 0.1788 - accuracy: 0.9229 - val_loss: 0.2293 - val_accuracy: 0.9038\n",
      "Epoch 23/25\n",
      "14/14 [==============================] - 10s 736ms/step - loss: 0.1702 - accuracy: 0.9280 - val_loss: 0.2487 - val_accuracy: 0.8949\n",
      "Epoch 24/25\n",
      "14/14 [==============================] - 10s 749ms/step - loss: 0.1686 - accuracy: 0.9248 - val_loss: 0.2541 - val_accuracy: 0.8955\n",
      "Epoch 25/25\n",
      "14/14 [==============================] - 10s 732ms/step - loss: 0.1542 - accuracy: 0.9365 - val_loss: 0.2418 - val_accuracy: 0.9038\n"
     ]
    }
   ],
   "source": [
    "epochs = 25\n",
    "batch_size = 32\n",
    "callbacks = None\n",
    "\n",
    "history = model.fit(\n",
    "    X_train_reshaped, y_train_reshaped,\n",
    "    validation_data=(X_test_reshaped, y_test_reshaped),\n",
    "    epochs=epochs, batch_size=batch_size, callbacks=callbacks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:13.641058Z",
     "iopub.status.busy": "2022-05-13T08:13:13.639994Z",
     "iopub.status.idle": "2022-05-13T08:13:14.074643Z",
     "shell.execute_reply": "2022-05-13T08:13:14.074037Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.245369Z"
    },
    "papermill": {
     "duration": 0.85287,
     "end_time": "2022-05-13T08:13:14.074764",
     "exception": false,
     "start_time": "2022-05-13T08:13:13.221894",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1728x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1728x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_loss(history):\n",
    "    plt.figure(figsize=(24, 6))\n",
    "    plt.plot(history.history['loss'], c =\"darkblue\")\n",
    "    plt.plot(history.history['val_loss'], c =\"crimson\")\n",
    "    plt.legend([\"Train\", \"Validation\"])\n",
    "    plt.title(\"Model loss\")\n",
    "    plt.xlabel(\"Epoch\")\n",
    "    plt.ylabel(\"Loss\")\n",
    "    plt.grid(True, alpha = 0.2)\n",
    "    plt.show()\n",
    "    \n",
    "\n",
    "def plot_accuracy(history):\n",
    "    plt.figure(figsize=(24, 6))\n",
    "    plt.plot(history.history['accuracy'], c =\"darkblue\")\n",
    "    plt.plot(history.history['val_accuracy'], c =\"crimson\")\n",
    "    plt.legend([\"Train\", \"Validation\"])\n",
    "    plt.title(\"Model accuracy\")\n",
    "    plt.xlabel(\"Epoch\")\n",
    "    plt.ylabel(\"Accuracy\")\n",
    "    plt.grid(True, alpha = 0.2)\n",
    "    plt.show() \n",
    "\n",
    "    \n",
    "plot_loss(history)\n",
    "plot_accuracy(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.388295,
     "end_time": "2022-05-13T08:13:14.847247",
     "exception": false,
     "start_time": "2022-05-13T08:13:14.458952",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Validation model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.387791,
     "end_time": "2022-05-13T08:13:15.694824",
     "exception": false,
     "start_time": "2022-05-13T08:13:15.307033",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Let's will get prediction on choosen audio file and will compare with truth labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.394802,
     "end_time": "2022-05-13T08:13:16.498361",
     "exception": false,
     "start_time": "2022-05-13T08:13:16.103559",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Preprocessing input data (extraction features):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:17.291785Z",
     "iopub.status.busy": "2022-05-13T08:13:17.291137Z",
     "iopub.status.idle": "2022-05-13T08:13:17.413913Z",
     "shell.execute_reply": "2022-05-13T08:13:17.414603Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.643073Z"
    },
    "papermill": {
     "duration": 0.520649,
     "end_time": "2022-05-13T08:13:17.414784",
     "exception": false,
     "start_time": "2022-05-13T08:13:16.894135",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "annotation_file = \"../input/speech-activity-detection-datasets/Data/Annotation/Female/TMIT/SA1.TextGrid\"\n",
    "audio_file = \"../input/speech-activity-detection-datasets/Data/Audio/Female/TMIT/SA1.wav\"\n",
    "\n",
    "dataset_valid = list()\n",
    "\n",
    "# Load samples:\n",
    "input_signal, fs = librosa.load(audio_file)\n",
    "\n",
    "# Load labels:\n",
    "truth_labels = readLabels(path=annotation_file, sample_rate=fs)\n",
    "\n",
    "# Extract logfbank features:\n",
    "features_logfbank_valid = python_speech_features.base.logfbank(signal=input_signal, samplerate=fs, winlen=frame_length, winstep=frame_step, nfilt=num_features, \n",
    "                                                               nfft=num_nfft, lowfreq=0, highfreq=None, preemph=preemphasis_coef)\n",
    "\n",
    "# Reshape labels for each group of features:\n",
    "marker_per_frames_truth = python_speech_features.sigproc.framesig(sig=truth_labels, frame_len=frame_length * fs, frame_step=frame_step * fs, \n",
    "                                                                 winfunc=np.ones)\n",
    "\n",
    "# For every frame calc label:\n",
    "marker_per_frame_truth = np.zeros(marker_per_frames_truth.shape[0])\n",
    "marker_per_frame_truth = np.array([1 if np.sum(marker_per_frames_truth[j], axis=0) > marker_per_frames_truth.shape[0] / 2 else 0 \n",
    "                                   for j in range(marker_per_frames_truth.shape[0])])\n",
    "\n",
    "spectrogram_image_valid = np.zeros((n_frames, n_frames))\n",
    "for j in range(int(np.floor(features_logfbank_valid.shape[0] / n_frames))):\n",
    "    spectrogram_image_valid = features_logfbank_valid[j * n_frames:(j + 1) * n_frames]\n",
    "    label_spectrogram_image_valid = 1 if np.sum(marker_per_frame_truth[j * n_frames:(j + 1) * n_frames]) > n_frames / 2 else 0\n",
    "    dataset_valid.append((label_spectrogram_image_valid, spectrogram_image_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:18.201180Z",
     "iopub.status.busy": "2022-05-13T08:13:18.200174Z",
     "iopub.status.idle": "2022-05-13T08:13:18.205690Z",
     "shell.execute_reply": "2022-05-13T08:13:18.206323Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.762695Z"
    },
    "papermill": {
     "duration": 0.398122,
     "end_time": "2022-05-13T08:13:18.206471",
     "exception": false,
     "start_time": "2022-05-13T08:13:17.808349",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "if stride - len(dataset_valid) > 0:\n",
    "    for i in range(stride - len(dataset_valid)):\n",
    "        dataset_valid.append((0, np.zeros((n_frames, n_frames))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:18.988203Z",
     "iopub.status.busy": "2022-05-13T08:13:18.987185Z",
     "iopub.status.idle": "2022-05-13T08:13:18.997272Z",
     "shell.execute_reply": "2022-05-13T08:13:18.997833Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.771094Z"
    },
    "papermill": {
     "duration": 0.402316,
     "end_time": "2022-05-13T08:13:18.997989",
     "exception": false,
     "start_time": "2022-05-13T08:13:18.595673",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Split dataset on train and test:\n",
    "X_valid = list()\n",
    "y_valid = list()\n",
    "for i in range(len(dataset_valid)):\n",
    "    X_valid.append(dataset_valid[i][1])\n",
    "    y_valid.append(dataset_valid[i][0])\n",
    "    \n",
    "X_valid = np.array(X_valid)\n",
    "y_valid = np.array(y_valid)\n",
    "\n",
    "# Reshaping for scaling:\n",
    "X_valid = X_valid.reshape(X_valid.shape[0], X_valid.shape[1] * X_valid.shape[2])\n",
    "    \n",
    "# Scale data:\n",
    "X_valid = scaler.transform(X_valid)\n",
    "\n",
    "# And reshape back:\n",
    "X_valid = X_valid.reshape(X_valid.shape[0], n_frames, n_frames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:19.778583Z",
     "iopub.status.busy": "2022-05-13T08:13:19.777502Z",
     "iopub.status.idle": "2022-05-13T08:13:19.786836Z",
     "shell.execute_reply": "2022-05-13T08:13:19.787384Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.787345Z"
    },
    "papermill": {
     "duration": 0.401646,
     "end_time": "2022-05-13T08:13:19.787540",
     "exception": false,
     "start_time": "2022-05-13T08:13:19.385894",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Encoding label:\n",
    "y_truth = y_valid\n",
    "y_valid = pd.get_dummies(y_valid)\n",
    "y_valid = np.array(y_valid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:20.563666Z",
     "iopub.status.busy": "2022-05-13T08:13:20.562672Z",
     "iopub.status.idle": "2022-05-13T08:13:20.570790Z",
     "shell.execute_reply": "2022-05-13T08:13:20.571366Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.809025Z"
    },
    "papermill": {
     "duration": 0.396233,
     "end_time": "2022-05-13T08:13:20.571575",
     "exception": false,
     "start_time": "2022-05-13T08:13:20.175342",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Reshape data for convolution layer:\n",
    "X_valid_reshaped = X_valid[:int(np.floor(X_valid.shape[0] / stride) * stride)]\n",
    "y_valid_reshaped = y_valid[:int(np.floor(X_valid.shape[0] / stride) * stride)]\n",
    "\n",
    "X_valid_reshaped = X_valid_reshaped.reshape((int(X_valid_reshaped.shape[0] / stride), stride, n_frames, n_frames, 1))\n",
    "y_valid_reshaped = y_valid_reshaped.reshape((int(y_valid_reshaped.shape[0] / stride), stride, y_valid[-1].shape[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:21.420451Z",
     "iopub.status.busy": "2022-05-13T08:13:21.419373Z",
     "iopub.status.idle": "2022-05-13T08:13:22.264988Z",
     "shell.execute_reply": "2022-05-13T08:13:22.264028Z",
     "shell.execute_reply.started": "2022-05-13T07:59:45.822985Z"
    },
    "papermill": {
     "duration": 1.235258,
     "end_time": "2022-05-13T08:13:22.265135",
     "exception": false,
     "start_time": "2022-05-13T08:13:21.029877",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction = model.predict(X_valid_reshaped)\n",
    "\n",
    "predicted_label = np.zeros(prediction.shape[1])\n",
    "predicted_proba = np.zeros(prediction.shape[1])\n",
    "ind = 0\n",
    "for i in range(prediction.shape[1]):\n",
    "    if prediction[0][i][0] >= prediction[0][i][1]:\n",
    "        predicted_label[ind] = 0\n",
    "        predicted_proba[ind] = prediction[0][i][0]\n",
    "    else:\n",
    "        predicted_label[ind] = 1\n",
    "        predicted_proba[ind] = prediction[0][i][1]\n",
    "    ind = ind + 1\n",
    "        \n",
    "predicted_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:23.048674Z",
     "iopub.status.busy": "2022-05-13T08:13:23.047625Z",
     "iopub.status.idle": "2022-05-13T08:13:23.054811Z",
     "shell.execute_reply": "2022-05-13T08:13:23.054131Z",
     "shell.execute_reply.started": "2022-05-13T07:59:46.623988Z"
    },
    "papermill": {
     "duration": 0.402646,
     "end_time": "2022-05-13T08:13:23.054936",
     "exception": false,
     "start_time": "2022-05-13T08:13:22.652290",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.] [0 1 1 1 1 1 1 1 1 0 0 0 0 0 0]\n"
     ]
    }
   ],
   "source": [
    "print(predicted_label, y_truth)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:23.838025Z",
     "iopub.status.busy": "2022-05-13T08:13:23.836940Z",
     "iopub.status.idle": "2022-05-13T08:13:23.847706Z",
     "shell.execute_reply": "2022-05-13T08:13:23.848269Z",
     "shell.execute_reply.started": "2022-05-13T07:59:46.631870Z"
    },
    "papermill": {
     "duration": 0.406107,
     "end_time": "2022-05-13T08:13:23.848464",
     "exception": false,
     "start_time": "2022-05-13T08:13:23.442357",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "predicted_label_widely = np.zeros(predicted_label.shape[0] * n_frames)\n",
    "ind_start = 0\n",
    "ind_stop = n_frames\n",
    "shift_step = n_frames\n",
    "for i in range(predicted_label.shape[0]):\n",
    "    predicted_label_widely[ind_start:ind_stop] = predicted_label[i]\n",
    "    ind_start = ind_start + shift_step\n",
    "    ind_stop = ind_stop + shift_step\n",
    "\n",
    "label_timeseries = np.zeros(input_signal.shape[0])\n",
    "begin = int(0)\n",
    "end = int(frame_length * fs)\n",
    "shift_step = int(frame_step * fs)\n",
    "for i in range(predicted_label_widely.shape[0]):\n",
    "    label_timeseries[begin:end] = predicted_label_widely[i]\n",
    "    begin = begin + shift_step\n",
    "    end = end + shift_step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-05-13T08:13:24.633481Z",
     "iopub.status.busy": "2022-05-13T08:13:24.632440Z",
     "iopub.status.idle": "2022-05-13T08:13:25.059258Z",
     "shell.execute_reply": "2022-05-13T08:13:25.058565Z",
     "shell.execute_reply.started": "2022-05-13T07:59:46.648311Z"
    },
    "papermill": {
     "duration": 0.823386,
     "end_time": "2022-05-13T08:13:25.059380",
     "exception": false,
     "start_time": "2022-05-13T08:13:24.235994",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1728x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# define time axis\n",
    "Ns = len(input_signal)  # number of sample\n",
    "Ts = 1 / fs  # sampling period\n",
    "t = np.arange(Ns) * Ts  # time axis in seconds\n",
    "norm_coef = 1.1 * np.max(input_signal)\n",
    "\n",
    "edge_ind = np.min([input_signal.shape[0], len(truth_labels)])\n",
    "\n",
    "plt.figure(figsize=(24, 6))\n",
    "plt.plot(t[:edge_ind], input_signal[:edge_ind])\n",
    "plt.plot(t[:edge_ind], np.array(truth_labels[:edge_ind]) * norm_coef)\n",
    "plt.plot(t[:edge_ind], label_timeseries[:edge_ind] * norm_coef)\n",
    "\n",
    "plt.title(\"Ground truth labels\")\n",
    "plt.legend(['Signal', 'Speech', 'Predicted'])\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.9.16"
  },
  "papermill": {
   "duration": 535.307432,
   "end_time": "2022-05-13T08:13:25.559897",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2022-05-13T08:04:30.252465",
   "version": "2.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}