5.GV.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "8.GV_final",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# GLOBAL VARIANCE"
      ],
      "metadata": {
        "id": "335ldQTzaMMM"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Loading all the required libraries. \n",
        "\n"
      ],
      "metadata": {
        "id": "QA8uF2riaSpg"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cbjN4Qt0gulm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f6785f26-e863-48c4-88fd-ea749b97d8f7"
      },
      "source": [
        "!pip3 install pysptk"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting pysptk\n",
            "  Downloading pysptk-0.1.20.tar.gz (410 kB)\n",
            "\u001b[?25l\r\u001b[K     |▉                               | 10 kB 22.0 MB/s eta 0:00:01\r\u001b[K     |█▋                              | 20 kB 22.2 MB/s eta 0:00:01\r\u001b[K     |██▍                             | 30 kB 12.4 MB/s eta 0:00:01\r\u001b[K     |███▏                            | 40 kB 10.4 MB/s eta 0:00:01\r\u001b[K     |████                            | 51 kB 4.6 MB/s eta 0:00:01\r\u001b[K     |████▉                           | 61 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████▋                          | 71 kB 5.8 MB/s eta 0:00:01\r\u001b[K     |██████▍                         | 81 kB 6.0 MB/s eta 0:00:01\r\u001b[K     |███████▏                        | 92 kB 6.7 MB/s eta 0:00:01\r\u001b[K     |████████                        | 102 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████▉                       | 112 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████▋                      | 122 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████▍                     | 133 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████▏                    | 143 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████                    | 153 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████▉                   | 163 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████████▋                  | 174 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████████▍                 | 184 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████████▏                | 194 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████                | 204 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████▊               | 215 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████████████▋              | 225 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████████████▍             | 235 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████████████▏            | 245 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████            | 256 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████▊           | 266 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▋          | 276 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▍         | 286 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▏        | 296 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████        | 307 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▊       | 317 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▋      | 327 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▍     | 337 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▏    | 348 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████    | 358 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▊   | 368 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▋  | 378 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▍ | 389 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▏| 399 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 409 kB 5.4 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 410 kB 5.4 MB/s \n",
            "\u001b[?25h  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "    Preparing wheel metadata ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.7/dist-packages (from pysptk) (4.4.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from pysptk) (1.15.0)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from pysptk) (1.4.1)\n",
            "Requirement already satisfied: cython>=0.28.0 in /usr/local/lib/python3.7/dist-packages (from pysptk) (0.29.28)\n",
            "Requirement already satisfied: numpy>=1.13.3 in /usr/local/lib/python3.7/dist-packages (from scipy->pysptk) (1.21.5)\n",
            "Building wheels for collected packages: pysptk\n",
            "  Building wheel for pysptk (PEP 517) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pysptk: filename=pysptk-0.1.20-cp37-cp37m-linux_x86_64.whl size=929941 sha256=2f910d51b369a466f0056feaa1f884cab3ae0132ffd7c745fe1e3a83dd2cb092\n",
            "  Stored in directory: /root/.cache/pip/wheels/dd/37/3f/7583b00bf73e73650aad5d268ee46b95dbd6dffbd237e1bf58\n",
            "Successfully built pysptk\n",
            "Installing collected packages: pysptk\n",
            "Successfully installed pysptk-0.1.20\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x75MBp3_g82f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b5cc37c1-64e4-4273-b282-f365102e9ec3"
      },
      "source": [
        "!pip3 install pyworld"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting pyworld\n",
            "  Downloading pyworld-0.3.0.tar.gz (212 kB)\n",
            "\u001b[?25l\r\u001b[K     |█▌                              | 10 kB 20.7 MB/s eta 0:00:01\r\u001b[K     |███                             | 20 kB 12.3 MB/s eta 0:00:01\r\u001b[K     |████▋                           | 30 kB 9.6 MB/s eta 0:00:01\r\u001b[K     |██████▏                         | 40 kB 8.4 MB/s eta 0:00:01\r\u001b[K     |███████▊                        | 51 kB 4.5 MB/s eta 0:00:01\r\u001b[K     |█████████▎                      | 61 kB 5.3 MB/s eta 0:00:01\r\u001b[K     |██████████▉                     | 71 kB 5.7 MB/s eta 0:00:01\r\u001b[K     |████████████▍                   | 81 kB 4.2 MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 92 kB 4.6 MB/s eta 0:00:01\r\u001b[K     |███████████████▌                | 102 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |█████████████████               | 112 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |██████████████████▌             | 122 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |████████████████████            | 133 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▋          | 143 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▏        | 153 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▊       | 163 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▎     | 174 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▉    | 184 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▍  | 194 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████ | 204 kB 5.1 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 212 kB 5.1 MB/s \n",
            "\u001b[?25h  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "    Preparing wheel metadata ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: cython>=0.24.0 in /usr/local/lib/python3.7/dist-packages (from pyworld) (0.29.28)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from pyworld) (1.21.5)\n",
            "Building wheels for collected packages: pyworld\n",
            "  Building wheel for pyworld (PEP 517) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for pyworld: filename=pyworld-0.3.0-cp37-cp37m-linux_x86_64.whl size=609504 sha256=97a466ab771ec1030bfaa1e3d16440a9cbd6ba72ebb37499d6069de8f2e916df\n",
            "  Stored in directory: /root/.cache/pip/wheels/e7/7c/11/c775fffa0e1e7b05a6604b4323408a77f80fb4ab304d96b5c6\n",
            "Successfully built pyworld\n",
            "Installing collected packages: pyworld\n",
            "Successfully installed pyworld-0.3.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rObLqwVmgifu"
      },
      "source": [
        "import torch\n",
        "from scipy.io import wavfile\n",
        "import pysptk\n",
        "from pysptk.synthesis import Synthesizer, MLSADF\n",
        "import pyworld\n",
        "from os.path import join, basename\n",
        "import librosa\n",
        "import librosa.display\n",
        "import IPython\n",
        "from IPython.display import Audio\n",
        "import glob\n",
        "import os\n",
        "import numpy as np\n",
        "from os.path import join, basename\n",
        "import collections"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Loading Data \n",
        "* Upload your data in your google drive \n",
        "* Run the cell below , and your drive will be accessible from this notebook. This will ask for your permission.\n",
        "* After granting the permission , you can access any folder or file in your drive from this notebook environment.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "wUf2L9vkaueI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "eYUo9bm0BSYp",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "036f9702-c333-4637-d103-69ce0a8c3e85"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def getpaths():  \n",
        "  \"\"\"\n",
        "  This function takes input from the user , i.e. , the number of datasets and the path of the datasets\n",
        "  \n",
        "  \"\"\"\n",
        "  print(\"Enter No. of Datasets :\")\n",
        "  n=int(input())\n",
        "  paths=[]\n",
        "  for i in range(n):\n",
        "    if i ==0:\n",
        "      paths.append(input(\"Enter Original path : \"))\n",
        "    else:\n",
        "      paths.append(input(f\"Enter path{i} : \"))\n",
        "  return paths"
      ],
      "metadata": {
        "id": "k7saSLpbya79"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Enter the number of datasets and the paths :\n",
        "* Enter a numrical value , like 2/4 with respect to the number of datasets you want to visualize.\n",
        "* Enter the path of the original data first.\n",
        "* Then you can put the path of the other datasets. \n",
        "\n",
        "### Note : The datasets should contain same number of wav files with the same name of each wav file."
      ],
      "metadata": {
        "id": "ikLiTEGnb12A"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "paths=getpaths()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PZdyMEtGz7zk",
        "outputId": "eb0893f3-cc0e-4e4d-d686-9bde69b0b32c"
      },
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter No. of Datasets :\n",
            "2\n",
            "Enter Original path : /content/drive/MyDrive/Male-to-Male-Voice/SM1\n",
            "Enter path1 : /content/drive/MyDrive/Male-to-Male-Voice/converted_SM2_Speech_in_SM1_Voice\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%pylab inline\n",
        "from matplotlib import pyplot as plt\n",
        "rcParams[\"figure.figsize\"] = (16,5)"
      ],
      "metadata": {
        "id": "P21lTC7q89o2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "61bb12b8-92bd-457d-9cb6-b8ea4c65282b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Populating the interactive namespace from numpy and matplotlib\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Funtion to compute the static feature - \n",
        "* Takes wav file path as input.\n",
        "* Returns the Global variance as output."
      ],
      "metadata": {
        "id": "-PMK79QHcy_p"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9MESR2TJqcNY"
      },
      "source": [
        "def compute_static_features(wav):\n",
        "    fs,x=wavfile.read(wav)\n",
        "    x = x.astype(np.float64)\n",
        "    f0, timeaxis = pyworld.dio(x, fs, frame_period=5.0)\n",
        "    f0 = pyworld.stonemask(x, f0, timeaxis, fs)\n",
        "    spectrogram = pyworld.cheaptrick(x, f0, timeaxis, fs)\n",
        "    aperiodicity = pyworld.d4c(x, f0, timeaxis, fs)\n",
        "    alpha = pysptk.util.mcepalpha(fs)\n",
        "    mc = pysptk.sp2mc(spectrogram, order=24, alpha=alpha) \n",
        "    c0, mc = mc[:, 0], mc[:, 1:]\n",
        "    #print(shape(mc))\n",
        "    gv=np.var(mc,axis=0)\n",
        "    #print(gv)\n",
        "    return gv  #    [x1,x2, ......x24]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Function to calculate RMSE value "
      ],
      "metadata": {
        "id": "j7i9PwnJdVPQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def calc_rmse(x,y):\n",
        "    min=np.sqrt(((x - y) ** 2).mean())\n",
        "    return min"
      ],
      "metadata": {
        "id": "8r-laznddBK3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Function to visualize GV:\n",
        "* Takes the paths as input.\n",
        "* Takes user input for Labels and markers for the plot, with respect to each dataset.\n",
        "* Calculates the GV of all the datasets.\n",
        "* Compare the values and shows the best plot from the dataset."
      ],
      "metadata": {
        "id": "8NpTy8Ozdav3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def vis_gv(paths):\n",
        "    print(\"Enter the Labels and Markers for each label [few common markers - x , + , o , * , > , < , D , H , X , v , ^ ]:\")\n",
        "    labels=[]\n",
        "    marker=[]\n",
        "    n=len(paths)\n",
        "    for i in range(n):\n",
        "      if i ==0:\n",
        "        labels.append(input(\"Enter label for Original path : \"))\n",
        "        marker.append(input(\"Enter marker for Original path : \"))\n",
        "      else:\n",
        "        labels.append(input(f\"Enter label for path_{i} : \"))\n",
        "        marker.append(input(f\"ENter marker for path_{i} : \"))\n",
        "\n",
        "\n",
        "    gv_dict = collections.defaultdict(list)\n",
        "    file_paths={}\n",
        "    for i in range(n):\n",
        "      path=glob.glob(paths[i]+'/*')\n",
        "      file_paths[f'path_{i+1}'] =sorted(path)\n",
        "      for wav in file_paths[f'path_{i+1}']:\n",
        "        gv_dict[f'path_{i+1}'].append(compute_static_features(wav))\n",
        "\n",
        "\n",
        "    mini=[]\n",
        "\n",
        "    for i in range(len(gv_dict['path_1'])):\n",
        "      j=1\n",
        "      min_dif=0\n",
        "      while (j+1)<=n:\n",
        "    \n",
        "        min_dif+=(calc_rmse(gv_dict['path_1'][i],gv_dict[f'path_{j+1}'][i]))\n",
        "        #print(min_dif,j)\n",
        "        j+=1\n",
        "      #print(min_dif)\n",
        "      mini.append(min_dif)\n",
        "\n",
        "    #print(len(mini))\n",
        "    ind=mini.index(min(mini))\n",
        "    #print(ind)\n",
        "\n",
        "    figure(figsize=(16,6))\n",
        "    j=0\n",
        "    for i in gv_dict.keys():\n",
        "      plt.plot(gv_dict[i][ind], marker=marker[j] ,linewidth=2, label=labels[j])\n",
        "      legend(prop={\"size\": 18})\n",
        "      yscale(\"log\")\n",
        "      #xlim(0,24)\n",
        "      ylabel(\"GV\", fontsize=16)\n",
        "      xlabel(\"Index of Mel-cepstral coefficient\", fontsize=16)\n",
        "      plt.savefig('GV.png')\n",
        "      j+=1"
      ],
      "metadata": {
        "id": "MJuI6YohdEZx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vis_gv(paths)"
      ],
      "metadata": {
        "id": "ZYNiqJ6hk227",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 480
        },
        "outputId": "bb726716-ec47-4d0a-8d16-a800be984ed0"
      },
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter the Labels and Markers for each label [few common markers - x , + , o , * , > , < , D , H , X , v , ^ ]:\n",
            "Enter label for Original path : Original Audio\n",
            "Enter marker for Original path : X\n",
            "Enter label for path_1 : Synthesized Audio\n",
            "ENter marker for path_1 : *\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "ITJ0zQoLiy0W"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}