plotMfcc.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import librosa\n",
    "import numpy as np\n",
    "import tqdm\n",
    "import subprocess\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import IPython.display as ipd\n",
    "import librosa.display\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_dim_1 = 20\n",
    "feature_dim_2 = 15\n",
    "channel = 1\n",
    "epochs = 100\n",
    "batch_size = 100\n",
    "verbose = 1\n",
    "num_classes = 19 #categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wav2mfcc(file_path, max_pad_len=feature_dim_2):\n",
    "    wave, sr = librosa.load(file_path, mono=True, sr=None)\n",
    "    wave = wave[::3]\n",
    "    mfcc = librosa.feature.mfcc(wave, sr=16000)\n",
    "    pad_width = max_pad_len - mfcc.shape[1]\n",
    "    mfcc = np.pad(mfcc, pad_width=((0, 0), (0, pad_width)), mode='constant')\n",
    "    #print(mfcc.shape)\n",
    "    return mfcc\n",
    "\n",
    "def calculate_MFCC(file_path):\n",
    "    samples, sample_rate = librosa.load(file_path)\n",
    "    S = librosa.feature.melspectrogram(samples, sr=16000, n_mels=128)\n",
    "    log_S = librosa.power_to_db(S, ref=np.max)\n",
    "    mfcc = librosa.feature.mfcc(S=log_S, n_mfcc=21)\n",
    "    return librosa.feature.delta(_mfcc, order=2)\n",
    "\n",
    "def calculate_Mel_LogSpectrum_Mfcc(file_path,sampleRate = 16000):\n",
    "    max_pad_len = 15\n",
    "    samples, sample_rate = librosa.load(file_path)\n",
    "    S = librosa.feature.melspectrogram(samples, sr=16000, n_mels=128)\n",
    "    # Convert to log scale (dB). We'll use the peak power (max) as reference.\n",
    "    log_S = librosa.power_to_db(S, ref=np.max)\n",
    "    mfcc = librosa.feature.mfcc(S=log_S, n_mfcc=21)\n",
    "    return S, log_S, mfcc\n",
    "    #print(\"mfcc.shape[1] is {}\".format(mfcc.))\n",
    "    #pad_width = max_pad_len - mfcc.shape[1]\n",
    "    #mfcc = np.pad(mfcc, pad_width=((0, 0), (0, pad_width)), mode='constant')\n",
    "    \n",
    "    \n",
    "def Draw_MelSpectrogram_LogS_MFCC(wavfilePath,samplerate=16000):\n",
    "    _melspectrum, _log_s, _mfcc =  calculate_Mel_LogSpectrum_Mfcc(wavfilePath)\n",
    "    delta2_mfcc = librosa.feature.delta(_mfcc, order=2)\n",
    "    print(\"delta2_mfcc shape is {}\".format(delta2_mfcc.shape))\n",
    "    #Draw MelSpectrogram\n",
    "    plt.figure(figsize=(12, 4))\n",
    "    librosa.display.specshow(_log_s, sr=samplerate, x_axis='time', y_axis='mel')\n",
    "    plt.title('Mel power spectrogram ')\n",
    "    plt.colorbar(format='%+02.0f dB')\n",
    "    plt.tight_layout()\n",
    "    #Draw MFCC\n",
    "    plt.figure(figsize=(12, 4))\n",
    "    librosa.display.specshow(delta2_mfcc)\n",
    "    plt.ylabel('MFCC coeffs')\n",
    "    plt.xlabel('Time')\n",
    "    plt.title('MFCC')\n",
    "    plt.colorbar()\n",
    "    plt.tight_layout()\n",
    "    print(\"mfcc shape is : {}\".format(_mfcc.shape))\n",
    "    \n",
    "def drawNpy(npyFile):\n",
    "    img_array = np.load(npyFile)\n",
    "    plt.imshow(img_array)\n",
    "    librosa.display.specshow(img_array)\n",
    "    plt.ylabel('MFCC coeffs')\n",
    "    plt.xlabel('Time')\n",
    "    plt.title('MFCC')\n",
    "    plt.colorbar()\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/scipy/signal/_arraytools.py:45: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  b = a[a_slice]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "delta2_mfcc shape is (21, 44)\n",
      "mfcc shape is : (21, 44)\n",
      "delta2_mfcc shape is (21, 44)\n",
      "mfcc shape is : (21, 44)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sun_test_file_path1 = \"../../DyAudio_Corpus/DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_0_rec.wav\"\n",
    "home_sun_test_file_path1 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_0_rec.wav\"\n",
    "sun_test_file_path2 = \"../../DyAudio_Corpus/DysSpeech_Corpora//DrSun/filtered_train_wavfiles/1/DrSun0822_4711_1_rec.wav\"\n",
    "home_sun_test_file_path2 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_1_rec.wav\"\n",
    "#sun_test_file_path3 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_2_rec.wav\"\n",
    "#sun_test_file_path4 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_3_rec.wav\"\n",
    "#sun_test_file_path5 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_4_rec.wav\"\n",
    "#sun_test_file_path6 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_5_rec.wav\"\n",
    "#sun_test_file_path7 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_6_rec.wav\"\n",
    "#sun_test_file_path8 = \"../DysSpeech_Corpora/DrSun/filtered_train_wavfiles/1/DrSun0822_4711_7_rec.wav\"\n",
    "#calculateMFCC(sun_test_file_path)\n",
    "#wav2mfcc(sun_test_file_path)\n",
    "Draw_MelSpectrogram_LogS_MFCC(sun_test_file_path1)\n",
    "Draw_MelSpectrogram_LogS_MFCC(sun_test_file_path2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/scipy/signal/_arraytools.py:45: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  b = a[a_slice]\n"
     ]
    }
   ],
   "source": [
    "_melspectrum, _log_s, _mfcc =  calculate_Mel_LogSpectrum_Mfcc(sun_test_file_path1)\n",
    "delta2_mfcc_ = librosa.feature.delta(_mfcc, order=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "def output_spectrum_info(specData):\n",
    "    print(\"Spectrum is type of {}\".format(type(specData)))\n",
    "    print(\"Spectrum shape is {}\".format(specData.shape))\n",
    "    \n",
    "def save_mfcc_to_array(mfccData,targetPath):\n",
    "    np.save(targetPath+\"mels.npy\",mfccData)\n",
    "    \n",
    "def save_mels_to_array(melsData,targetPath):\n",
    "    np.save(targetPath+\"mels.npy\",melsData)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "folderList = ['1','2','3','4','5','6','7','8','9','10', '11', '12','13','14', '15','16','17','18','19']\n",
    "#save_mfcc_to_array(delta2_mfcc_,\"./test_mfcc/\")\n",
    "def save_train_data_to_npy(rootPath,desRootPath):\n",
    "    for folder in folderList:#DYSCmdCategString:\n",
    "        # Init mfcc vectors\n",
    "        mfcc_vectors = []\n",
    "        wavfiles = [rootPath + folder +\"/\"+ wav for wav in os.listdir(rootPath + folder+'/') if wav.endswith('.wav')]#if wavfile.endswith('.wav')\n",
    "        #wavfiles = [wavfile for wavfile in ]\n",
    "        print(\"Current folder is : \",folder)\n",
    "        #print(\"the whole files are : \",wavfiles)\n",
    "        #print(\"+++++++++++++++++++++++++++++++++++++\")\n",
    "        for _wav in wavfiles:#tqdm(wavfiles, \"Saving vectors of label - '{}'\".format(folder)):\n",
    "            #mfcc_ = wav2mfcc(wavfile)\n",
    "            mfcc_ = calculate_MFCC(_wav)\n",
    "            mfcc_vectors.append(mfcc_)\n",
    "        print(\"length of mfcc_vectors is : \",len(mfcc_))\n",
    "        np.save(\"./\"+desRootPath+folder + '.npy', mfcc_vectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current folder is :  1\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/scipy/signal/_arraytools.py:45: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  b = a[a_slice]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  3\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  4\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  5\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  6\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  7\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  8\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  9\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  10\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  11\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  12\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  13\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  14\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  15\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  16\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  17\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  18\n",
      "length of mfcc_vectors is :  21\n",
      "Current folder is :  19\n",
      "length of mfcc_vectors is :  21\n"
     ]
    }
   ],
   "source": [
    "save_train_data_to_npy(\"../../DyAudio_Corpus/DysSpeech_Corpora/DrSun/filtered_train_wavfiles/\",\"./delta_mfcc\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spectrum is type of <class 'numpy.ndarray'>\n",
      "Spectrum shape is (21, 44)\n"
     ]
    }
   ],
   "source": [
    "output_spectrum_info(delta2_mfcc_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "drawNpy(\"./test_mfcc/mels.npy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Spectrum calculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def log_specgram(audio, sample_rate, window_size=20,\n",
    "                 step_size=10, eps=1e-10):\n",
    "    nperseg = int(round(window_size * sample_rate / 1e3))\n",
    "    noverlap = int(round(step_size * sample_rate / 1e3))\n",
    "    freqs, times, spec = signal.spectrogram(audio,\n",
    "                                    fs=sample_rate,\n",
    "                                    window='hann',\n",
    "                                    nperseg=nperseg,\n",
    "                                    noverlap=noverlap,\n",
    "                                    detrend=False)\n",
    "    return freqs, times, np.log(spec.T.astype(np.float32) + eps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Calculating Spectrum manually"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.io.wavfile as wav\n",
    "from numpy.lib import stride_tricks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def stft(sig, frameSize, overlapFac=0.5, window=np.hanning):\n",
    "    win = window(frameSize)\n",
    "    hopSize = int(frameSize - np.floor(overlapFac * frameSize))\n",
    "    \n",
    "    # zeros at beginning (thus center of 1st window should be for sample nr. 0)\n",
    "    samples = np.append(np.zeros(np.floor(frameSize/2.0)), sig)    \n",
    "    # cols for windowing\n",
    "    cols = np.ceil( (len(samples) - frameSize) / float(hopSize)) + 1\n",
    "    # zeros at end (thus samples can be fully covered by frames)\n",
    "    samples = np.append(samples, np.zeros(frameSize))\n",
    "    \n",
    "    frames = stride_tricks.as_strided(samples, shape=(cols, frameSize), strides=(samples.strides[0]*hopSize, samples.strides[0])).copy()\n",
    "    frames *= win\n",
    "    \n",
    "    return np.fft.rfft(frames)\n",
    "\n",
    "\"\"\" scale frequency axis logarithmically \"\"\"    \n",
    "def logscale_spec(spec, sr=44100, factor=20.):\n",
    "    timebins, freqbins = np.shape(spec)\n",
    "\n",
    "    scale = np.linspace(0, 1, freqbins) ** factor\n",
    "    scale *= (freqbins-1)/max(scale)\n",
    "    scale = np.unique(np.round(scale))\n",
    "    \n",
    "    # create spectrogram with new freq bins\n",
    "    newspec = np.complex128(np.zeros([timebins, len(scale)]))\n",
    "    for i in range(0, len(scale)):\n",
    "        if i == len(scale)-1:\n",
    "            newspec[:,i] = np.sum(spec[:,scale[i]:], axis=1)\n",
    "        else:        \n",
    "            newspec[:,i] = np.sum(spec[:,scale[i]:scale[i+1]], axis=1)\n",
    "    \n",
    "    # list center freq of bins\n",
    "    allfreqs = np.abs(np.fft.fftfreq(freqbins*2, 1./sr)[:freqbins+1])\n",
    "    freqs = []\n",
    "    for i in range(0, len(scale)):\n",
    "        if i == len(scale)-1:\n",
    "            freqs += [np.mean(allfreqs[scale[i]:])]\n",
    "        else:\n",
    "            freqs += [np.mean(allfreqs[scale[i]:scale[i+1]])]\n",
    "    \n",
    "    return newspec, freqs\n",
    "\n",
    "\"\"\" plot spectrogram\"\"\"\n",
    "def plotstft(audiopath, binsize=2**10, plotpath=None, colormap=\"jet\"):\n",
    "    samplerate, samples = wav.read(audiopath)\n",
    "    s = stft(samples, binsize)\n",
    "    \n",
    "    sshow, freq = logscale_spec(s, factor=1.0, sr=samplerate)\n",
    "    ims = 20.*np.log10(np.abs(sshow)/10e-6) # amplitude to decibel\n",
    "    \n",
    "    timebins, freqbins = np.shape(ims)\n",
    "    \n",
    "    plt.figure(figsize=(15, 7.5))\n",
    "    plt.imshow(np.transpose(ims), origin=\"lower\", aspect=\"auto\", cmap=colormap, interpolation=\"none\")\n",
    "    plt.colorbar()\n",
    "\n",
    "    plt.xlabel(\"time (s)\")\n",
    "    plt.ylabel(\"frequency (hz)\")\n",
    "    plt.xlim([0, timebins-1])\n",
    "    plt.ylim([0, freqbins])\n",
    "\n",
    "    xlocs = np.float32(np.linspace(0, timebins-1, 5))\n",
    "    plt.xticks(xlocs, [\"%.02f\" % l for l in ((xlocs*len(samples)/timebins)+(0.5*binsize))/samplerate])\n",
    "    ylocs = np.int16(np.round(np.linspace(0, freqbins-1, 10)))\n",
    "    plt.yticks(ylocs, [\"%.02f\" % freq[i] for i in ylocs])\n",
    "    \n",
    "    if plotpath:\n",
    "        plt.savefig(plotpath, bbox_inches=\"tight\")\n",
    "    else:\n",
    "        plt.show()\n",
    "        \n",
    "    plt.clf()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}