char2mel_calculation_backup.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Character to Mel Spectrogram Sequence Generation Model with location-based Attention"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline\n",
    "import os\n",
    "import sys\n",
    "import pickle\n",
    "sys.path.append('../../')\n",
    "import IPython.display as ipd\n",
    "import librosa\n",
    "import librosa.display\n",
    "import numpy as np\n",
    "import seaborn as sns; sns.set()\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from tools.libaudio.utils import load\n",
    "from tools.libaudio.feature import melspectrogram\n",
    "from tools.libaudio.waves import sin_wave\n",
    "from tools.libaudio.display import show_spec, show_mel\n",
    "from mlutils.utils import plot, plots, to_onehot\n",
    "from models.phoneme import Phoneme43\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "**RNN Sequence Genetor with location-based Attention**\n",
    "```\n",
    "\n",
    "U: char length\n",
    "c: char sequence\n",
    "T: timestep length of input x\n",
    "wt: window into c at timestep t\n",
    "φ(t, u): window weight of cu at timestep t\n",
    "αt: parameters control the importance of the window within the mixture\n",
    "βt: parameters control the width of the window\n",
    "kt: parameters control the location of the window\n",
    "\n",
    "ht^1 = H(Wih1 xt + Wh1h1 h1t-1 + Wwh1 wt-1 + b1h)\n",
    "\n",
    "(αhat_t, βhat_t, khat_t) = Wh1p ht^1 + bp  # output of the first hidden layer\n",
    "\n",
    "αt = exp(αhat_t)  # importance of the window\n",
    "βt = exp(βhat_t)  # width of the window\n",
    "kt = kt-1 + exp(khat_t)  # location of the window (how far to slide each window)\n",
    "\n",
    "φ(t, u) = Σk=1->K αt^k exp(-βt^k(kt^k-u)^2)  # discrete convolution of a mixture of K Gaussian\n",
    "\n",
    "\n",
    "ht^n = H(Wihn xt + Whn-1hn ht^n-1 + Whnhn ht-1^n + Wwhn wt + bh^n)\n",
    "\n",
    "\n",
    "```\n",
    "\n",
    "ref. Generating Sequences With Recurrent Neural Networks  \n",
    "https://arxiv.org/pdf/1308.0850.pdf\n",
    "\n",
    "**Char to MelSpectrum model**\n",
    "```\n",
    "seq of char (n,)\n",
    "-> embedding (N_char, 512)\n",
    "-> conv (shape:5x1, filter:512)+batch norm+relu x 3\n",
    "-> bi-GRU (256 hidden size  x 2)\n",
    "-> (encoded features)\n",
    "-> attention network\n",
    "-> (attention context)\n",
    "-> residual uni-GRU (1024 hidden size) x 2\n",
    "-> fully connected (256) x2 + (attention context)\n",
    "-> residual uni-GRU (1024 hidden size) x 2\n",
    "-> Mel spectrogram\n",
    "```\n",
    "ref. NATURAL TTS SYNTHESIS BY CONDITIONING WAVENET ON MEL SPECTROGRAM PREDICTIONS  \n",
    "https://arxiv.org/pdf/1712.05884.pdf\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mel Spectrogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1befbda9b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "key = 'BASIC5000_0005'\n",
    "wav_005 = load(f'/diskB/6/Datasets/VoiceData/jsut_ver1.1/basic5000/wav/{key}.wav', sample_rate=24000)\n",
    "plot(wav_005)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "                <audio controls=\"controls\" >\n",
       "                    <source src=\"data:audio/wav;base64,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\" type=\"audio/wav\" />\n",
       "                    Your browser does not support the audio element.\n",
       "                </audio>\n",
       "              "
      ],
      "text/plain": [
       "<IPython.lib.display.Audio object>"
      ]
     },
     "execution_count": 265,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ipd.Audio(wav_005, rate=24000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [],
   "source": [
    "text_005 = 'ケツアツハケンコウノバロメータートシテジュウヨウデアル'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(88560,)"
      ]
     },
     "execution_count": 267,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wav_005.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {},
   "outputs": [],
   "source": [
    "mel = melspectrogram(wav_005, sample_rate=24000, n_fft=2048, hop_length=300, power=2.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1be9c65e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_mel(mel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([128, 296])"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mel = torch.Tensor(mel)\n",
    "mel.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Phoneme (Character) indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "segments[0] ('/diskB/6/Datasets/VoiceData/jsut_ver1.1/segmentation_results/all/BASIC5000_1701.log', [(26, 30, 'b'), (31, 36, 'o'), (37, 42, 'k'), (43, 46, 'u'), (47, 54, 'n'), (55, 62, 'a'), (63, 67, 'r'), (68, 80, 'a'), (81, 94, 'n'), (95, 99, 'a'), (100, 108, 'n'), (109, 114, 'i'), (115, 124, 'o'), (125, 133, 'o'), (134, 138, 'i'), (139, 145, 't'), (146, 148, 'e'), (149, 158, 'm'), (159, 172, 'o'), (173, 183, 'd'), (184, 192, 'o:'), (193, 204, 's'), (205, 216, 'o:'), (217, 226, 'k'), (227, 233, 'a'), (234, 236, 'i'), (237, 245, 'w'), (246, 250, 'a'), (251, 260, 'k'), (261, 263, 'a'), (264, 269, 'n'), (270, 274, 'a'), (275, 279, 'r'), (280, 282, 'a'), (283, 291, 'z'), (292, 294, 'u'), (295, 306, 'sh'), (307, 309, 'u'), (310, 312, 'q'), (313, 321, 's'), (322, 327, 'e'), (328, 332, 'k'), (333, 335, 'i'), (336, 345, 's'), (346, 349, 'u'), (350, 353, 'r'), (354, 358, 'u'), (359, 361, 'd'), (362, 367, 'a'), (368, 375, 'r'), (376, 392, 'o'), (393, 395, 'u')])\n"
     ]
    }
   ],
   "source": [
    "with open('/diskB/6/Datasets/VoiceData/jsut_ver1.1/preprocessed/segments.pickle', 'rb') as f:\n",
    "    segments = pickle.load(f)\n",
    "    print(f'segments[0] {segments[0]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [],
   "source": [
    "def search(key):\n",
    "    for d in segments:\n",
    "        if d[0].endswith(f'{key}.log'):\n",
    "            return d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [],
   "source": [
    "phonemes = [Phoneme43.to_index(phone[2]) for phone in search(key)[1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[25, 3, 26, 2, 0, 26, 2, 11, 0, 25, 3, 10, 25, 9, 33, 4, 23, 0, 38, 4, 32, 8, 24, 5, 24, 4, 35, 1, 24, 3, 13, 7, 12, 9, 29, 3, 0, 38, 2]\n"
     ]
    }
   ],
   "source": [
    "print(phonemes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['k', 'e', 'ts', 'u', 'a', 'ts', 'u', 'w', 'a', 'k']"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "phoneme_names = [Phoneme43(phoneme).name for phoneme in phonemes]\n",
    "phoneme_names[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Character encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {},
   "outputs": [],
   "source": [
    "vocab_size = len(Phoneme43)\n",
    "\n",
    "x = torch.LongTensor(to_onehot(phonemes, n_class=vocab_size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        ...,\n",
       "        [1, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 1,  ..., 0, 0, 0]])"
      ]
     },
     "execution_count": 277,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([39, 43])"
      ]
     },
     "execution_count": 278,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_size = 512\n",
    "embedding = nn.Embedding(vocab_size, embed_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_emb = embedding(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        ...,\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685]],\n",
       "       grad_fn=<SelectBackward>)"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([39, 43, 512])"
      ]
     },
     "execution_count": 282,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb.shape  # (T, H, W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [],
   "source": [
    "class NGramConvolution(torch.nn.Module):\n",
    "    def __init__(self, embed_size=512, kernel_size=(5, 1)):\n",
    "        super(NGramConvolution, self).__init__()\n",
    "        self.conv1 = torch.nn.Conv2d(in_channels=embed_size, out_channels=embed_size, kernel_size=kernel_size)\n",
    "        self.conv2 = torch.nn.Conv2d(in_channels=embed_size, out_channels=embed_size, kernel_size=kernel_size)\n",
    "        self.conv3 = torch.nn.Conv2d(in_channels=embed_size, out_channels=embed_size, kernel_size=kernel_size)\n",
    "        self.bn1 = torch.nn.BatchNorm2d(num_features=embed_size)\n",
    "        self.bn2 = torch.nn.BatchNorm2d(num_features=embed_size)\n",
    "        self.bn3 = torch.nn.BatchNorm2d(num_features=embed_size)\n",
    "        self.relu1 = torch.nn.ReLU()\n",
    "        self.relu2 = torch.nn.ReLU()\n",
    "        self.relu3 = torch.nn.ReLU()\n",
    "    \n",
    "    def THW2NWTH(self, x):\n",
    "        x = x.permute(2, 0, 1)\n",
    "        return x.view(1, x.shape[0], x.shape[1], x.shape[2])\n",
    "    \n",
    "    def forward(self, x):\n",
    "        if len(x.shape) == 3:\n",
    "            x = self.THW2NWTH(x)\n",
    "        x = self.conv1(x)\n",
    "        x = self.bn1(x)\n",
    "        x = self.relu1(x)\n",
    "        x = self.conv2(x)\n",
    "        x = self.bn2(x)\n",
    "        x = self.relu2(x)\n",
    "        x = self.conv3(x)\n",
    "        x = self.bn3(x)\n",
    "        x = self.relu2(x)\n",
    "        x = x.permute(2, 0, 3, 1)  # (T, N, H, W)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {},
   "outputs": [],
   "source": [
    "ngc = NGramConvolution(embed_size=embed_size, kernel_size=(5, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3.58 s, sys: 220 ms, total: 3.8 s\n",
      "Wall time: 196 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "x_emb_c = ngc(x_emb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.0000, 0.0000, 1.5784,  ..., 3.8419, 1.1681, 0.0000],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 1.5156,  ..., 1.5452, 1.3186, 0.0000],\n",
       "        ...,\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004]],\n",
       "       grad_fn=<SelectBackward>)"
      ]
     },
     "execution_count": 286,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb_c[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([27, 1, 43, 512])"
      ]
     },
     "execution_count": 287,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb_c.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Sequence Generator with location-based Attention\n",
    "\n",
    "```\n",
    "x: input sequence\n",
    "y: mel spectrum sequence\n",
    "U: char length\n",
    "c: char sequence\n",
    "T: timestep length of input x\n",
    "wt: window vector into c at timestep t\n",
    "φ(t, u): window weight of cu at timestep t\n",
    "αt: parameters control the importance of the window within the mixture\n",
    "βt: parameters control the width of the window\n",
    "κt: parameters control the location of the window\n",
    "(eq.46): discrete convolution with a mixture of K Gaussian function\n",
    "\n",
    "xt ∈ R×R×{0,1}\n",
    "\n",
    "h1t = H(Wih1 xt + Wh1h1 h1t-1 + Wwh1 wt-1 + b1h)\n",
    "\n",
    "(αhat_t, βhat_t, khat_t) = Wh1p ht^1 + bp  # output of the first hidden layer\n",
    "\n",
    "αt = exp(αhat_t)  # importance of the window\n",
    "βt = exp(βhat_t)  # width of the window\n",
    "κt = κt-1 + exp(κhat_t)  # location of the window (how far to slide each window)\n",
    "\n",
    "φ(t, u) = Σk=1->K αkt*exp(-βkt(κkt-u)^2)  # mixture of K Gaussian\n",
    "\n",
    "\n",
    "cf. \n",
    "normal distribution\n",
    "N(X|μ,σ2) = 1/(2πσ2)^1/2 exp{-1/2σ2(x-μ)2}\n",
    "\n",
    "mixture gaussians\n",
    "p(x) = Σk=1->K πk N(X|μk,Σk)  # where πk:mixing coefficient, μk:mean, Σk:covariance, N(X|μk,Σk):mixture component\n",
    "p(x) = Σk=1->K πk (1/(2π)^D/2 1/|Σ|^1/2) exp{-1/2(x-μ)^T Σ^-1(x-μ)}  # where Σ:DxD dim covariance matrix, |Σ|:det Σ\n",
    "\n",
    "wt = Σu=1->U φ(t, u)*cu  # the soft window into c at timestep t\n",
    "\n",
    "hnt = H(Wihn xt + Whn-1hn hnt-1 + Whnhn hnt-1 + Wwhn wt + bnh)\n",
    "\n",
    "yˆtˆ = (eˆt,{wˆj_t,μˆJ_t,σˆj_t,ρˆj_t}^M_j=1) = by + Σn=1->N Whny hnt\n",
    "\n",
    "et = 1 / (1 + exp(eˆt)) ⇒ et∈(0,1) : stroke probability\n",
    "πtj = exp(πtj) / (Σj't=1->M exp(πtj)) ⇒πtj∈(0,1), Σπtj=1 : mixture weights\n",
    "μjt = μˆjt ⇒ μjt∈R : means\n",
    "σtj = exp(σˆtj) ⇒ σˆtj > 0 : std\n",
    "pjt = tanh(pˆtj) ⇒ pˆtj∈(0,1) : correlations\n",
    "\n",
    "Pr(x|c) = ∏t=1->T Pr(xt+1|yt)\n",
    "L(x) = -log Pr(x|c)\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Input & Output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        ...,\n",
       "        [1, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 0,  ..., 0, 0, 0],\n",
       "        [0, 0, 1,  ..., 0, 0, 0]])"
      ]
     },
     "execution_count": 288,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# c1. onehot\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([39, 43])"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        ...,\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685],\n",
       "        [ 0.1277, -0.8859,  1.1535,  ...,  1.9341, -0.2470, -0.3685]],\n",
       "       grad_fn=<SelectBackward>)"
      ]
     },
     "execution_count": 290,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# c2. onehot embeded\n",
    "x_emb[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([39, 43, 512])"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[0.0000, 0.0000, 1.5784,  ..., 3.8419, 1.1681, 0.0000],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 1.5156,  ..., 1.5452, 1.3186, 0.0000],\n",
       "        ...,\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004],\n",
       "        [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0004]],\n",
       "       grad_fn=<SelectBackward>)"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# c3. onehot embedded 5-gram\n",
    "x_emb_c[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([27, 1, 43, 512])"
      ]
     },
     "execution_count": 293,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_emb_c.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[4.1781e-04, 4.2978e-04, 3.8574e-04,  ..., 1.8835e-02, 3.2648e-02,\n",
       "         4.3341e-02],\n",
       "        [1.1756e-04, 1.4426e-04, 1.5435e-04,  ..., 2.5821e-04, 3.2323e-04,\n",
       "         3.2827e-04],\n",
       "        [2.8792e-05, 4.0903e-05, 6.2976e-05,  ..., 4.9818e-05, 5.0324e-05,\n",
       "         7.5028e-05],\n",
       "        ...,\n",
       "        [3.7895e-07, 4.1616e-07, 3.6224e-07,  ..., 1.5572e-08, 3.1202e-08,\n",
       "         4.8096e-08],\n",
       "        [2.7661e-07, 2.4075e-07, 1.4813e-07,  ..., 6.5611e-09, 1.0549e-08,\n",
       "         1.5595e-08],\n",
       "        [1.1432e-07, 7.7864e-08, 2.2786e-08,  ..., 8.3096e-10, 2.5113e-09,\n",
       "         4.2880e-09]])"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# y. melspectrogram\n",
    "mel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([128, 296])"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mel.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [],
   "source": [
    "K = vocab_len  # mixture of K Gaussian\n",
    "padded_text_len = 999   # text length with padding (=max sequence len) from data set\n",
    "U = vocab_len = n_chars = 43  # num of characters\n",
    "hidden_size = 256\n",
    "mel_feature_size = mel.shape[1]  # 296\n",
    "batch_size = 1\n",
    "T = max_target_size = 800  # max time step"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Input data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {},
   "outputs": [],
   "source": [
    "# y: mel spectrogram (N, T, H)\n",
    "y = mel.view(batch_size, mel.shape[0], mel.shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 800, 296])"
      ]
     },
     "execution_count": 298,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# add padding to x according to max_target_size (T) \n",
    "y = torch.cat([y, torch.zeros((y.shape[0], T-y.shape[1], y.shape[2]))], dim=1)\n",
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 296])"
      ]
     },
     "execution_count": 299,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# time step 0 for x\n",
    "y[:, 0, :].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 39, 43])"
      ]
     },
     "execution_count": 300,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reshape x:text sequence from (T, H) to (N, T, H)\n",
    "x = x.view(batch_size, x.shape[0], x.shape[1])\n",
    "x.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hidden Layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 321,
   "metadata": {},
   "outputs": [],
   "source": [
    "# GRUCell only returns the next hidden state (batch, hidden_size)\n",
    "\n",
    "# h1t = H(Wih1 xt + Wh1h1 h1t-1 + Wwh1 wt-1 + b1h) = H(Wiwh1 (xt + wt-1) + Wh1h1 h1t-1 + b1h)\n",
    "H1 = torch.nn.GRUCell(input_size=vocab_size+3, hidden_size=hidden_size)  # NOTE: 3? window size?\n",
    "H2 = torch.nn.GRUCell(input_size=vocab_size+3+hidden_size, hidden_size=hidden_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 43])"
      ]
     },
     "execution_count": 302,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# begin with the first timestep\n",
    "t = 0\n",
    "xt_1 = x[:, t, :]\n",
    "xt_1.shape  # TODO: must be size of vocab len (43)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 3])"
      ]
     },
     "execution_count": 303,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the first window is from the first character (phoneme 0: \"a\")\n",
    "wt_1 = torch.zeros((batch_size, 3)).long()\n",
    "wt_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 46])"
      ]
     },
     "execution_count": 304,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# initial input (N, vocab_size+3(alpha,beta,kappa))\n",
    "xt_1_wt_1 = torch.cat([xt_1, wt_1], dim=-1).float()\n",
    "xt_1_wt_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 256])"
      ]
     },
     "execution_count": 305,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# initial hidden state\n",
    "ht_1 = torch.zeros(batch_size, hidden_size)\n",
    "ht_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 256])"
      ]
     },
     "execution_count": 306,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# h1t = H(Wih1 xt + Wh1h1 h1t-1 + Wwh1 wt-1 + b1h)\n",
    "h1t_1 = H1(xt_1_wt_1, ht_1)\n",
    "h1t_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 43])"
      ]
     },
     "execution_count": 307,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# initial kappa\n",
    "kappa_t_1 = torch.zeros((batch_size, K))\n",
    "kappa_t_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([1, 43]), torch.Size([1, 43]), torch.Size([1, 43]))"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (αhat_t, βhat_t, khat_t) = Wh1p ht^1 + bp  # output of the first hidden layer\n",
    "Wh1p = torch.nn.Linear(hidden_size, 3*K)\n",
    "bp = torch.nn.Parameter(torch.zeros(3*K))\n",
    "alpha_beta_kappa = Wh1p(h1t_1) + bp\n",
    "alpha_t, beta_t, kappa_t = alpha_beta_kappa[:, :K], alpha_beta_kappa[:, K:K*2], alpha_beta_kappa[:, K*2:K*3]\n",
    "# αt = exp(αhat_t)  # importance of the window\n",
    "# βt = exp(βhat_t)  # width of the window\n",
    "# κt = κt-1 + exp(κhat_t)  # location of the window (how far to slide each window)\n",
    "alpha_t = alpha_t.exp()\n",
    "beta_t = beta_t.exp()\n",
    "kappa_t = kappa_t_1 + kappa_t.exp()\n",
    "alpha_t.shape, beta_t.shape, kappa_t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[[0.9620, 0.9620, 0.9620,  ..., 0.9620, 0.9620, 0.9620],\n",
       "         [0.9624, 0.9624, 0.9624,  ..., 0.9624, 0.9624, 0.9624],\n",
       "         [1.0692, 1.0692, 1.0692,  ..., 1.0692, 1.0692, 1.0692],\n",
       "         ...,\n",
       "         [1.0156, 1.0156, 1.0156,  ..., 1.0156, 1.0156, 1.0156],\n",
       "         [0.9990, 0.9990, 0.9990,  ..., 0.9990, 0.9990, 0.9990],\n",
       "         [0.9674, 0.9674, 0.9674,  ..., 0.9674, 0.9674, 0.9674]]],\n",
       "       grad_fn=<RepeatBackward>)"
      ]
     },
     "execution_count": 309,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kappa_t.unsqueeze(2).repeat(1, 1, x.shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 0.,  1.,  2.,  3.,  4.,  5.,  6.,  7.,  8.,  9., 10., 11., 12., 13.,\n",
       "         14., 15., 16., 17., 18., 19., 20., 21., 22., 23., 24., 25., 26., 27.,\n",
       "         28., 29., 30., 31., 32., 33., 34., 35., 36., 37., 38., 39., 40., 41.,\n",
       "         42.]])"
      ]
     },
     "execution_count": 310,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.Tensor(range(U)).view(1, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 322,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[[ 9.6203e-01, -3.7973e-02, -1.0380e+00,  ..., -3.5038e+01,\n",
       "          -3.6038e+01, -3.7038e+01],\n",
       "         [ 9.6236e-01, -3.7638e-02, -1.0376e+00,  ..., -3.5038e+01,\n",
       "          -3.6038e+01, -3.7038e+01],\n",
       "         [ 1.0692e+00,  6.9233e-02, -9.3077e-01,  ..., -3.4931e+01,\n",
       "          -3.5931e+01, -3.6931e+01],\n",
       "         ...,\n",
       "         [ 1.0156e+00,  1.5598e-02, -9.8440e-01,  ..., -3.4984e+01,\n",
       "          -3.5984e+01, -3.6984e+01],\n",
       "         [ 9.9903e-01, -9.7132e-04, -1.0010e+00,  ..., -3.5001e+01,\n",
       "          -3.6001e+01, -3.7001e+01],\n",
       "         [ 9.6739e-01, -3.2612e-02, -1.0326e+00,  ..., -3.5033e+01,\n",
       "          -3.6033e+01, -3.7033e+01]]], grad_fn=<ThSubBackward>)"
      ]
     },
     "execution_count": 322,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "u = torch.Tensor(range(x.shape[1])).view(1, -1)\n",
    "kappa_t.unsqueeze(2).repeat(1, 1, x.shape[1]) - u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 323,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1c1c435f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "u = torch.Tensor(range(x.shape[1])).view(1, -1)\n",
    "a = alpha_t.unsqueeze(2) * (-beta_t.unsqueeze(2)*(kappa_t.unsqueeze(2).repeat(1, 1, x.shape[1]) - u)**2).exp()\n",
    "a = a.sum(dim=1) / K\n",
    "a = a.squeeze(0).detach().numpy()\n",
    "plot(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1bee3dfcf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# φ(t, u) = Σk=1->K αkt*exp(-βkt(κkt-u)^2)  # mixture of K Gaussian\n",
    "\n",
    "# unefficient way\n",
    "#phi_t = torch.stack([alpha_t*(-beta_t*(kappa_t-u)**2).exp() for u in range(x.shape[1])]).permute(1, 2, 0)  # (N, K, U)\n",
    "#phi_t = torch.sum(phi_t, dim=1)  # linear combination of K columns to a mixture distribution length U (N, U)\n",
    "#phi_t, phi_t.shape\n",
    "\n",
    "u = torch.Tensor(range(x.shape[1]))\n",
    "phi_t = (alpha_t.unsqueeze(2) * (-beta_t.unsqueeze(2) * (kappa_t.unsqueeze(2).repeat(1, 1, x.shape[1]) - u)**2).exp()).sum(dim=1)\n",
    "phi_t, phi_t.shape\n",
    "\n",
    "#u = torch.Tensor(range(U))\n",
    "#phi_t = (alpha_t.unsqueeze(2) * (-beta_t.unsqueeze(2) * (kappa_t.unsqueeze(2).repeat(1, 1, U) - u)**2).exp()).sum(dim=1)\n",
    "#phi_t, phi_t.shape\n",
    "#alpha_t*(-beta_t*(kappa_t-0)**2).exp()\n",
    "plot(phi_t.squeeze(0).detach().numpy())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 348,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[[0, 0, 0,  ..., 0, 0, 0],\n",
       "          [0, 0, 0,  ..., 0, 0, 0],\n",
       "          [0, 0, 0,  ..., 0, 0, 0],\n",
       "          ...,\n",
       "          [1, 0, 0,  ..., 0, 0, 0],\n",
       "          [0, 0, 0,  ..., 0, 0, 0],\n",
       "          [0, 0, 1,  ..., 0, 0, 0]]]),\n",
       " torch.Size([1, 39, 43]),\n",
       " tensor([[15.8055, 43.3287, 16.2702,  0.8441,  0.0062,  0.0000,  0.0000,  0.0000,\n",
       "           0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,\n",
       "           0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,\n",
       "           0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,\n",
       "           0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000]],\n",
       "        grad_fn=<SumBackward1>),\n",
       " torch.Size([1, 39]))"
      ]
     },
     "execution_count": 348,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x, x.shape, phi_t, phi_t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[[ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000],\n",
       "          [ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000],\n",
       "          [ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000],\n",
       "          ...,\n",
       "          [ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000],\n",
       "          [ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000],\n",
       "          [ 0.0000,  0.0000,  0.0000,  ...,  0.0000,  0.0000,  0.0000]]],\n",
       "        grad_fn=<ThMulBackward>), torch.Size([1, 39, 43]))"
      ]
     },
     "execution_count": 350,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# wt = Σu=1->U φ(t, u)*cu  # the soft window into c at timestep t\n",
    "# The size of the soft window vectors is the same as the size of the character vectors\n",
    "# cu (assuming a one-hot encoding, this will be the number of characters in the alphabet).\n",
    "wt = (x.float() * phi_t.unsqueeze(2))\n",
    "#wt = phi_t.transpose(0, 1)*x.float()\n",
    "wt, wt.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 343,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f1be9e88b00>"
      ]
     },
     "execution_count": 343,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWIAAAEZCAYAAACtuS94AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XtYVNXeB/DvnkGExPEW6ikU7xCvlhaGN8wDhLcskURQiCOWmSLpUwaUmietFLO8dZ5Mj4aAmJa3SvNYioqX1My8QMkhEaYLSiATAgrMvH/4Om8TKHv27GEzm++nZz+Ps12z9m8wf7P87bXWFkwmkwlERKQYjdIBEBE1dUzEREQKYyImIlIYEzERkcKYiImIFMZETESkMCZiokbk5s2bWLt2rd2vk5WVhYMHD9r9OiQOEzHVEhAQgFWrVtnUx8WLF+Hl5QW9Xi9TVA0nISEBUVFRottHRUVhyZIld21TU1OD4uLievt699134ezsbH599epV0XH81dmzZzFp0iT069cPQUFB2LJli/n3OnfujKVLl6KwsFBy/yQfJmKqZd++fZgxY4bSYSjmrbfewkcffSRrn7Nnz0ZGRsZd2+j1euzduxeTJk0CABQVFSE4OFjS9XJzcxEdHY2goCAcPnwY8+fPx+LFi7Fr1y4AgJubG8LCwvD+++9L6p/kxURMtWi1Wmg0Tfd/DY1GA61WK2ufJSUl9bZJSUnBmDFj4OTkBACorKxEeXm5pOutWbMGffv2xeTJk+Hm5oahQ4fiueeew4cffmhu89RTT2H37t2iRupkX033b1sTEhkZiWXLlplfHz58uFbZYPbs2Vi4cCEAYNy4cRalCS8vLxw6dAhxcXHo27cvgoKC8PXXX1tc47vvvkNoaCj69u2LyMhI/PTTT7XiyMvLw9SpU9G3b1/4+fnh9ddfx/Xr1wEAK1euxIQJE8xty8rK0Lt3b2zfvt18Lj09HWPGjLH58924cQMLFy6En58fBgwYgISEBFy7ds3cdsmSJRalicLCQsTGxsLX1xfDhw/H3LlzMWjQIBw7dszcxmg0Yvny5RgwYAAGDhyIFStWmH9vypQpOHHiBBITE+Hl5YWdO3fW+gwAkJGRgSFDhgC4VZIIDAw0//x9fHzqfM+dHD58GCNHjrQ4N2rUKOTk5Jj/bFq1agVvb28cP37cqr5JfkzETUBQUBAyMzPNrzMyMqDT6fDVV18BuJVEjh49av6LX5f58+dj+PDhOHToEIYPH47ExERUVFQAAPLz8zF58mT4+/sjIyMDs2bNskiMAFBcXIzIyEi0a9cO+/btQ2pqKs6fP4/p06ebYzx37hxKS0sBAEeOHIGrqyv27dtn7iMzMxMBAQE2f75XX30VZ86cQVpaGnbu3AmDwYDY2Ng7fvZXX30VlZWV2LNnD1auXIljx45h7ty5GDBggLnNtm3bcOPGDXz55ZdYtGgR/vWvf+G7774DAKxduxb9+/fHm2++iQsXLuDJJ5+sdY2SkhJcvnwZffr0AQC4u7ubP/uFCxdw7tw5AEBiYiIGDx58xyMnJwcGgwHFxcXo1KmTxTU8PDwgCAJ++eUX87k+ffrg9OnTd/zs1DCYiJuAoKAgZGdn4/fffwcAHDhwADExMeZR7fnz52E0GtG/f/879jFy5EiMHj0aOp0OUVFRKC0txeXLlwEAmzdvRqdOnfDiiy+idevW8PX1RXx8vMX7N23aBFdXV7zxxhtwd3dHz5498e677+LEiRM4ceIEfHx80LFjRxw9etQixqNHj6KiogLV1dU4fvw4goKCbPp8+fn5+Pzzz/HOO++gR48e6NChAxYvXozTp08jKyurzs/+7bffIjIyEu7u7vDy8kJwcDA+//xzCIJgbuPp6Yn4+Hi0bt0agYGB6Ny5M86ePQvgVqlDEARoNBo4OTlZvO+2q1evokWLFnB1dTWfu10ecnJyMpdKEhMTsWPHjjseXbt2NX9BtmrVyuIaWq0WLVq0sBj9u7u748qVK3V+bmo4TMRNgIeHB3r16oXMzEz8+OOPcHJyQnh4OM6cOYPi4mIcPnwYQ4cORbNmze7Yx6OPPmr+dcuWLQEABoMBwK0ZEg899JBFguncubPF+8+ePYtHHnnE4hqenp647777zKO9gIAAZGZmwmg04tChQwgJCUGXLl2QmZmJM2fOoEWLFujdu7dNn+/ChQsAbpUzbo8iR44cCZPJVGc5BQC8vb2xd+9eVFdXo6SkBJmZmbjvvvss2vz1S6xly5b4448/7vjz/Ks//vgDLVq0qLedTqeDu7v7HQ8nJye0adMGwP//+dxmMplQUVGB1q1bW8RZVlYmOk6yDyelA6CGcfuf74WFhQgMDESbNm3Qr18/HDx4EEeOHLFqutZtt3dQFQQBRqOx3rZ13QA0mUzm9wYFBSE+Ph7nzp3D3/72N3Ts2BFBQUHYv38/OnTogMDAwDpHk9Z8PpPJhGbNmmHbtm214rn9BfNXixYtwvjx47Fnzx5UV1djwIAB5pJKfZ9ZLDc3N1E35l555ZW7zv9NSUlBr1694O7ujoKCAovySUFBAWpqaiy+RMrKyuDm5iY6TrIPJuImIigoCM8//zyKi4vNSSQ4OBgHDhzADz/8AH9/f8l9e3l54auvvoLJZDInyu+//96iTe/evbF7925UV1ebZwUUFBTg119/NY9y+/fvj8rKSmzevNlcgggODsazzz6Lzp07Y9q0aTZ/vgceeABVVVXIz8+3GMX+9NNPuPfee+vse9myZUhMTMTYsWNhMpnQvHlzKT+mu3J3d0dZWRkqKyvh4uICAHV+6cydOxdz5sy5Yz+3R8PBwcHYs2cPxo8fb/69r7/+Gt27d0e3bt3M54qKiuDu7i7XxyCJWJpoInx8fNCsWTNcvHgR/fr1AwDz7IdHHnnEplHRhAkTUFhYiGXLluHatWs4cuQINm7caNFm4sSJMBgMeOONN1BUVITc3Fy89NJL6Nevn3nUptVqMWzYMOzcudOciHv16gVXV1fk5OTAz8/P5s/XtWtXjBw5EomJiThx4gTKysrw5ZdfYtKkScjPz6+z78LCQly8eBGVlZXmLxFr6XQ65OTkoKCgoM7pYm3btkWnTp3MpZPb7wGA7OxsnD9/3nyuvtIEADzzzDM4ffo0NmzYgLKyMmRmZuL999/H1KlTLa577tw59O3bV9JnIvkwETchgYGBeOyxx8z/JO/YsSN8fHzuOltCjE6dOiE5ORnHjx/HsGHDsGbNGixevNiijbu7O1JTU/Hzzz8jMDAQEydORK9evbBmzRqLkd/jjz8ODw8P9OzZ03wuKCgIgwcPvmsN25rPt2TJEgQFBWH27NkYNGgQPvroI6xatQqenp519rtgwQLs3r0b/fv3h4+PD3x9fZGQkICbN2+K/hlNmjQJ27dvR3R0NC5dulRnm6FDh1pMiWvZsiUiIiIQERGBZcuWoaqqSvT1unTpgnXr1uGLL77AwIEDsWDBAsyZMwdjx441tykrK0NWVhYGDhwoul+yD4GPSiK6M4PBgPHjx2P69OkYMWIETCYTvv/+ezz33HNYvnx5ndPppMrLy8Ozzz6L//znPw2yoGbz5s04c+ZMrS9NangcERPdRVFREfLy8lBdXY3q6mpUVFTg119/BYBa83Rt1aVLFwwaNAhbt26Vtd+6VFRUIC0trUkvZW9MOCImqsenn36KlJQU5OXlQavVomfPnnjhhRfw2GOPyX6tiooKrF27FnFxcbL3/Wdnz56FXq/HqFGj7HodEoeJmIhIYSxNEBEpjImYiEhhTMRERApjIiYiUhgTMRGRwpiIiYgUxkRMRKQwJmIiIoUxERMRKYyJmIhIYUzEREQKYyImIlIYEzERkcIa/Jl1OlduQk1E4hgqEmzuo/Rmt/ob/Z9WznU/ydveJI2I9Xo9Hn/8cfPrnTt3IiHB9h8YEZHsjFrxh0L4FGciUjXBWPtp2I0NEzERqZtJpYn4rw82rK6uliUYIiK5OcKIWFKNuG3btigtLUVZWRlu3rzZIA87JCKSQqgWfyhF0ojYxcUFL7zwAsLDw9G+fXsMGTIEer1e7tiIiGwmOMBTORv84aGcvkZEYskxfa28yFt023vu/cHm60nBm3VEpG7Gxj8kZiImIlVzhNIEEzERqZtR6QDqx0RMRKomVDf+IbHkTX9SUlLw1FNPYdy4cUhNTZUzJiIi2Qgm8YdSJCVio9EIFxcXbN26FVu2bEFqaioqKirkjo2IyHZGKw4Jzp8/jwcffBDffPMNAOD06dOYMGECxo0bh5iYGFy5cqXePiSvrKuursaUKVNgNBpRVFSE4uJi3H///VK6IyKyG8GONeJr165h0aJFGDRoEADg+vXrePHFF7Fu3Tp4eXkhPT0d8+fPxwcffHDXfiSNiE+dOoWNGzdixYoVSEtLQ48ePdDA05GJiMQxmcQfVjAajXjllVcQGxuL1q1bAwAOHz6MBx98EF5eXgCAsLAwnDt3DiUlJXftS9KI+Pr163B3d0fbtm2Rm5uL7Oxs1NTUSOmKiMiurFm6bDAYYDAYap3X6XTQ6XQW51atWoU+ffpgyJAh+PzzzwEA+fn56Nq1q7mNVquFh4cHfv75Z7Rp0+aO15WUiIcMGYIdO3YgNDQUnp6eCA0NRX5+Pjw9PaV0R0RkN9bchEtOTsbq1atrnY+NjcXMmTPNrw8ePIjz589jzZo1Fu2cnJwgCJabDAmCUOvcX0lKxFqtFu+9956UtxIRNSwrasTR0dEICQmpdf6vo+EvvvgCP/30E0aNGgUAuHr1Ko4fP46nn34aeXl55nYmkwl6vb7e+2ecR0xE6mZFIq6rBFGXpKQki9cJCQkICQlB3759ERwcjIsXL6JXr1747LPP0Lt3b3MN+U6YiIlI1YQG3Bi+efPmWLZsGeLj46HVaqHT6fD222/X+z6bd1/T6/WYPHky9u3bJ6o9d18jIrHk2H2t5vteottqH7po8/Wk4IiYiNTNASZ0MRETkao5wqOSJCXiLVu2ICUlBe3bt8fDDz8sd0xERPJxgLVmVq+sy8nJwYcffohNmzbh3//+tz1iIiKSj1EQfyjE6kT8zTffICgoCC1btgQAPPnkk7IHRUQkGztv+iMHq0sTgiBY7CthNDrArstE1HSpsTQxYMAA7N+/H9euXQMAbNy4UfagiIhkU6MRfyjE6it3794dM2bMQFRUFMLDw9Grl/g5ekREDU0wij8Ui9HWBR3W4oIOIhJLjgUdpgP/I7qt8PcLNl9PCs4jJiJ1c4AaMRMxEambWhd0EBE5jBomYiIiZTXg7mtSSZ6vERMTg71795pfjx49Grm5ubIERUQkF5NREH0oRXIijoyMxPbt2wEA2dnZaNWqFbp37y5bYEREsjAJ4g+FSE7Ew4YNw6VLl1BUVIRdu3YhLCxMzriIiOThAEucJSdijUaDsLAwfPbZZ8jMzMSIESPkjIuISB4OMCK26Wbd008/jaeeegoBAQFwcXGRKyYiIvkouHRZLJsScatWrdChQweWJYio8VL7POLvvvsOgiDA29tbrniIiOSl5pV1x44dQ2JiIlasWCFnPEREslJyWppYkhPxwIEDkZGRIWMoRER2oOYFHbddvXoVERERcsRCRCQ/B3hUks1LnN3d3ZGeni5HLERE8nOAWRM2R/jbb78hICBAjliIiGRnMok/lMJNf4hI3RygRsxETETqpuZZE0REjsDEETERkcIc4GadzYm4pqYGTk7M50TUODWJEfHZs2fRuXNnOWIhIpKfgttbimVTIj527BiWL1+O9957T654iIjkZacRcVpaGrZt2waNRoN77rkHr7/+Orp164Zjx46Zt35wcnLCvHnz4OXldde+bErEfn5+Fo9LIiJqbOyx10RxcTFycnKQnp4OZ2dnHDhwAP/85z+xcuVKzJs3D6mpqejYsSOOHz+OxMREbNu27a792VTF1mg0WL9+vS1dEBHZlx02hm/bti0WLFgAZ2dnmEwm5OXlQaPRoHnz5li5ciU6duwIAOjVqxcKCgrq7c/mGnFqaipiYmJs7YaIyC5MVsyaMBgMMBgMtc7rdDrodLpa59955x1s374dPj4+WLBgAVxcXODj4wMAKCsrQ0JCAiZOnFjvdW0aEX/wwQfmTX+2bNliS1dERPZhxYg4OTkZgYGBtY7k5OQ6u3755Zdx4MABPPTQQzhy5Ij5fHZ2NiIiIjBw4EDMnj273hAFk8m2FdYBAQHYv3+/6PY618W2XI6ImhBDRYLNfVQsf1x026qYT0WPiE0mEwThVjmjoqIC/v7+OHXqFE6ePIklS5YgKSkJ3bp1E3VdTgAmInWz4mbdnUoQf3Xw4EHs2LEDS5cuhZOTE7755ht07NgR165dw8KFC7Fhwwa0a9dO9HWZiIlI1eyxoMPf3x/nz59HWFgYnJ2dodFokJSUhIyMDPz++++IjY21aL98+XJ06NDhjv3ZnIi1Wi3Ky8vh6upqHqYTETUadkjEGo0GM2bMwIwZMyzO+/j4YOzYsdb3Z2tAQUFBmDBhAnbt2mVrV0REsjPVaEQfSrH5Zp21eLOOiMSS42Zd2eLRotu6JXxh8/WkYI2YiNTNAfYjljQWvz1hmYioseOjkoiIFNYktsEkImrMlLwJJ1a9ibisrAyzZ8/G9evX4ebmhoULFwIA1q5di4yMDJSWliIpKYnlCiJqnBxgRFzvV0VmZiY8PT2xadMmzJ49GzqdDjU1NbjnnnuQlpaG6Oho7sBGRI2WySSIPpRSbyL29fXFyZMnkZSUBDc3N7i6ukKj0SA8PBwA0LVrV1y5csXugRIRSWIUxB8KqTcR33vvvfjkk0/Qu3dvPPfcczh79iwEQYBWqzW3aeCpyEREoqli1kRBQQHKy8sxatQoXLlyBadPn26IuIiIZOEIsybqHRFXV1djyZIliIiIQEZGBkaPFr9KhYhIaVziXAcucSYiseRY4lwUHya67b1LlHnABecRE5G6OUBpgomYiFRNFTXi+nAOMRE1Ziaj+EMpNifi1NRUOeIgIrIPKx4eqhQ+xZmIVM1o1Ig+lGJTjXjatGnYsmUL0tPT5YqHiEheDlAj5s06IlI1kwNsDM9ETESq1iRmTdx+ijP3myCiRslkxaEQPsWZiFRN9TfrACA+Ph7x8fFyxEJEJDvWiImIlOYANWImYiJStSZxs45LnImoMVPFo5LqwyXORNSYOcITOrjEmYhUzRE2hucSZyJSNUeoEfNmHRGpmiMkYuXG4kREDcBeN+uys7MRFRWFiIgIhIWF4cSJExa/v3//fnh7e0Ov19fbl80j4ttLnF1dXSEIjf+bh4iaFnuMiKurqzFr1iysWLEC3t7euHTpEqKionDo0CFoNBrk5+dj/fr16Nu3r6j+uMSZiFTNVCOIPsQSBAFLly6Ft7c3AKBTp064ceMGrl+/jsrKSsyZMwdz585Fs2bNRPXHJc5EpGrWjIgNBgMMBkOt8zqdDjqdzvxaq9XiwQcfBABUVVVhwYIFGDp0KFq2bIlXXnkFEyZMMCdpMXizjohUzZpEnJycjNWrV9c6Hxsbi5kzZ9Y6r9fr8dJLL8HX1xezZs3Cpk2b0Lx5c4wbN86qGAVTA+9fqXNd3JCXIyIHZqhIsLmP7LA40W3vX7dI1IgYAHJzc/Hyyy/j9ddfN9eCo6KiUFhYCI3mVtX3119/hbu7O+Li4vDkk0/e8bocERORqlkzIq4r4dalqqoKc+bMwbJly9CtWzfz+ZSUFIt2UVFRePvtt+Hh4XHX/piIiUjd7DBr4vvvv0dBQQFee+01i/Ovvvoq+vTpY3V/TMREpGr22PDd19cXJ0+erLfdX0fIdyI6Efv5+WHs2LE4efIkOnbsiLFjx2LDhg2orKzEO++8g+7du4vtioiowZiMSkdQP9FfFdeuXcOjjz6Kbdu2obS0FJ999hnS09MRERGBdevW2TNGIiLJVLUNpkajQUBAAIBbk5cDAwMBAJ07d0ZhYaF9oiMispEjJGLRpQlBECyWMN/+tSAIMBodYOxPRE2SI2z6w5t1RKRqTMRERAqzx6wJuXFlHRE1WnKsrPt2RKLoto98+bbN15OCI2IiUjWWJoiIFMZETESkMCWfziwWEzERqZoj3KyzOUI/Pz988cUXmDx5Mhr4vh8RUb1UtaDjTm7evImrV69iw4YNcsRDRCSrJlEjLi8vx9ixY+WIhYhIdiZjE0jEWq0WrVu3liMWIiLZNYkRMRFRY8ZETESkMEeYNWFzIs7KypIjDiIiu+CImIhIYUzEREQKYyImIlJYk5i+RkTUmDnCiFjS7URfX1+sX78e4eHhCAkJwbfffit3XEREsqgxCqIPpUgaEZeVlUGj0WDz5s04f/484uLi8OWXX8LZ2Vnu+IiIbKLaEbEgCIiMjAQA9O7dG61atUJeXp6ccRERyULVm/5oNP+fw//8dGciosZEtSNio9GIXbt2Abi1oKO0tBRdunSRMy4iIlmodkSs1Wrx22+/4R//+AcMBgOWLl3K+jARNUrGGhUvcZ42bRqmTZsmZyxERLJzhNIE5xETkaqpNhFzox8ichRGB0jEjb94QkRkA3vcrMvLy8Pq1asxaNAgnDp1yuYYWZogIlWzR2ni008/Rbt27dC+fXuL8waDAQsWLEB+fj5MJhMGDx6M2bNn1zvFV1Ii9vPzQ0xMDA4dOoTS0lIkJSXBx8dHSldERHZlrJE/Eb/00ksAgK+//tri/FtvvYWQkBD4+/vDZDIhJydHVH+SShPXrl2Dm5sb0tLSEB0djfXr10vphojI7owmQfRhMBig1+trHQaDod7rFBcXIy8vD9nZ2Zg4cSKmT5+Oe+65R9SCN0mJWKPRIDw8HADQtWtXXLlyRUo3RER2Z02NODk5GYGBgbWO5OTkeq9z+fJl/Pjjj2jTpg02bdqEiRMn4vnnn0dNTU2975VUmhAEAVqt9k8f1CSlGyIiu7OmRhwdHY2QkJBa53U6Xb3vLS8vR7t27TB+/HgAgL+/P5KSknDp0iX06NHjru/lzToiUjVrErFOpxOVdOty//331xqUarVai0HrnXD6GhGpmjU1Ylt06dIF7du3x+7duwEAZ86cQU1NDTw9Pet9r80LOnx9fZGSkiKlGyIiu7PHrIk7SUpKwmuvvYZ169bB1dUVy5cvt9ip8k5YmiAiVbPnEue/DkI7deqEjRs3Wt2P5NIE5w0TkSNoqNKELTgiJiJVc4RJXUzERKRqjrD7ms2zJnJycjBixAjo9Xo54iEikpVqn+J8W0lJCeLj47FixQp4eHjIFRMRkWwcYURsUyKOi4tDWFgYvLy85IqHiEhWqt6P2Gg0YsyYMUhLS0N5ebmcMRERycZkFH8oRXIi1mg0CAsLwxNPPIHFixfLGRMRkWwc4SnONt+smzJlCs6dO4eMjAwZwiEikpeq5xHfXubs5OSE7du3yxYQEZGclJwNIRbnERORqnFBBxGRwlQ/fY2IqLEzckRMRKQsVZcmdu/ejY8++giCIMDb2xvz588XtRM9EVFDcoSbdZKmr+Xl5WHdunVITk7Gxx9/jObNm2Pz5s1yx0ZEZDOTSfyhFEkj4iNHjqCkpARTp04FAFRWVvIBokTUKDnCEmdJidhkMiEgIADz5s2TOx4iIlk5whBRUmli8ODB2Lt3r3nry5MnT+Ljjz+WNTAiIjkYTeIPpUgaEXft2hXz58/HzJkz0bx5c7Rp0waLFi2SOzYiIpuZoNLSBAAEBwcjODhYzliIiGRX4wC1Cc4jJiJV44IOIiKFOUAeZiImInXjiJiISGEOkIeZiIlI3RR8ApJokhNxSkoKPvnkE2i1WowbNw6RkZFyxkVEJIsapQMQQVIiNhqNcHFxwdatW6HRaPDEE08gNDQUrq6ucsdHRGQTR9h9QVIi1mg0qK6uxpQpU2A0GlFUVITi4mLcf//9csdHRGQT1ZYmTp06hY0bNyItLQ1t27ZFeHg4N/0hokbJETKTpER8/fp1uLu7o23btsjNzUV2djZqahyhEkNETY1qR8RDhgzBjh07EBoaCk9PT4SGhiI/Px+enp5yx0dEZBN7DRGPHz+O5cuX37pGTQ3i4uLg7+8vqS9JiVir1eK9996TdEEiooZkjxFxVVUVXnzxRaSnp6Nbt27Izc1FWFgYjhw5AhcXF6v74zxiIlI1kxVVYoPBAIPBUOu8TqeDTqczv75x4wacnJzQpUsXALd2pHRzc4PRKC3tC6YGvsumc13ckJcjIgdmqEiwuY/nBPGPcXtw5VWsXr261vnY2FjMnDnT4ty2bduQlZWFJ554AkePHsUDDzyAv//975Ji5IiYiFTNmpFmdHQ0QkJCap3/82j4tsLCQvTr1w89evTADz/8gG+//VZyIpb0hA4iIkdhtOLQ6XTw8PCodfw1ER87dgxZWVkYPXo03NzcEB4ejry8PGRmZkqKUVIi1uv1ePzxx82vd+7ciYQE2/8JQUQktxrBJPoQq3nz5rh48aK5nlxSUoKLFy/C2dlZUowsTRCRqtlj1sTDDz+MyMhIPPPMM3BxcUFlZSUmTpyIRx99VFJ/TMREpGrWzJqwRlRUFKKiomTpS/JeE39WXV0tSzBERHJzhJV1kmrEbdu2RWlpKcrKynDz5k1s3bpV7riIiGRhsuI/pUgaEbu4uOCFF15AeHg42rdvjyFDhkCv18sdGxGRzRxhRCy5Rjx58mRMnjxZzliIiGRnzWwIpfBmHRGpmqpHxEREjkDJ2q9YTMREpGocERMRKcyo5hExn+JMRI5AtTfr+BRnInIUqq0R8ynOROQoVFsj5lOcichRqLZGbMtTnN2MzaRckohIksafhiXuNTFkyBC0a9cOoaGheP/9981PcSYiamyMgkn0oRQ+xZmIVK3GAcbEnEdMRKqm2hoxEZGjUF0iXr9+PWJiYmy64IpOnGtMRA3HEaavWXWzLjU11V5xEBHZhao2hv/ggw9w9epVREREYMSIEdizZw8EQUDPnj0xb948NGvGaWlE1Pg4QmlC9Ih42rRpcHd3R3p6Oi5fvoyYmBikp6dj7NixTMJE1GhVCybRh1Ik3awbPnw4Fi5ciF9++QWhoaFyx0REJBtVjYj/zM/PD6mpqaipqUGvKUbrAAADC0lEQVRYWBjKysrkjouISBaqqhEDtxZylJeXIysrCz169MCUKVNw6tQp5Ofnw8fHR1QfwVnvSAqUiJqi6Tb3oLoRcVBQECZMmICCggI8//zzCA8PR+vWreHl5WWv+IiIbGKESfShFMHUwNumld7s1pCXIyIH1sr5J5v78HJeKbrtjzfjbL6eFFxZR0SqZhSUjqB+km7WARBdEyYiUpIjlCYafES82yu+oS9JRA4q4pLtfTjCzTqWJohI1RxhG0zJpYnbcnJyMGLECOj1ejniISKSlepLEyUlJYiPj8eKFSvg4eEhV0xERLKpEsQ9xs1aubm5WLBgAa5fvw5nZ2e8+eab6N69u6S+bBoRx8XFISwsjPOIiajRqoFJ9CGW0WjEzJkzMXXqVGzbtg3Tp0/HrFmzJMcoOREbjUaMGTMGaWlpKC8vlxwAEZE9WZOIDQYD9Hp9rcNgMFj0eeHCBTg7O8Pf3x8AMHToUGi1WmRnZ0uKUXJpQqPRICwsDCUlJVi8eDHeeOMNUe+LuPS81EsSEVmtpFL8TK1Vq1Zh9erVtc7HxsZi5syZ5tf5+fno2rWrRZtu3bohPz8fDzzwgNUx2jxrYsqUKRg/fjwyMjIwbNgwW7sjIlJMdHQ0QkJCap3X6XQWr52cnCAIlitFBEGodU4syYk4KyvLHND27duldkNE1GjodLpaSbcunp6eyMvLsziXl5eHzp07S7quzdPXiIiaGm9vb1RVVeHo0aMAgFOnTqGqqgre3t6S+mvwTX+IiNTgv//9L+bOnYvq6mo0a9YMixYtkjx9jYmYiEhhLE0QESmMiZiISGFMxERECmMiJiJSGBMxEZHCmIhJUbm5uYiKisK4ceMQHh6O3NxcpUMianBMxKQYuXewInJUTMSkGLl3sCJyVEzEpJi77WBF1JQwEZNi5N7BishRMRGTYuTewYrIUTERk2Lk3sGKyFFx0x9SlJw7WBE5KiZiIiKFsTRBRKQwJmIiIoUxERMRKYyJmIhIYUzEREQKYyImIlIYEzERkcKYiImIFPa/xlIbpIs2AVsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f1be9e88978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# window weights φ(t, u) during a time step sequence\n",
    "\n",
    "# the window weight φ(t, u) can be loosely interpreted as\n",
    "# the network’s belief that it is writing character cu at time t\n",
    "#frame = pd.DataFrame(w.squeeze(0).detach().numpy(), index=phoneme_names).iloc[::-1]  # reverse by row\n",
    "frame = pd.DataFrame(phi_t.squeeze(0).detach().numpy(), index=phoneme_names).iloc[::-1]  # reverse by row\n",
    "sns.heatmap(frame, cmap=\"plasma\")\n",
    "plt.suptitle('window weight (t=0)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'xt_1_wt_1_h1t_1' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-346-3f824b56ea34>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mh2t_1\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mH2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mxt_1_wt_1_h1t_1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mht_1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mh2t_1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'xt_1_wt_1_h1t_1' is not defined"
     ]
    }
   ],
   "source": [
    "h2t_1 = H2(xt_1_wt_1_h1t_1, ht_1)\n",
    "h2t_1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Encoder(torch.nn.Module):\n",
    "    def __init__(self, encode_type:('onehot', 'embed', 'ngc', 'rnn') ='onehot', hidden_size=256):\n",
    "        self.vocab_size = len(Phoneme43)\n",
    "        self.encode_type = encode_type\n",
    "        \n",
    "        # embedding\n",
    "        self.hidden_size\n",
    "        self.embed_size = embed_size\n",
    "        self.embedding = nn.Embedding(self.vocab_size, hidden_size)\n",
    "        # n-gram conv (ngc)\n",
    "        self.ngc = NGramConvolution(embed_size=embed_size, kernel_size=(5, 1))\n",
    "        # rnn\n",
    "        self.gru = nn.GRU(hidden_size, hidden_size)\n",
    "    \n",
    "    def forward(x):\n",
    "        \"\"\"\n",
    "        args:\n",
    "            - x (batch, sequence_len, vocab_size): onehot encoding\n",
    "        \"\"\"\n",
    "        if len(x.shape) == 1:\n",
    "            x = torch.LongTensor(to_onehot(phonemes, n_class=vocab_size))\n",
    "            \n",
    "        if self.encode_type in ('embed', 'ngc', 'rnn'):\n",
    "            x = self.embedding(x)\n",
    "        if self.encode_type in ('ngc', 'rnn'):\n",
    "            x = ngc(x_emb)\n",
    "        if self.encode_type == 'rnn':\n",
    "            out, h = self.gru()\n",
    "            \n",
    "        if self.encode_type == 'rnn':\n",
    "            return out, h\n",
    "        else:\n",
    "            return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 356,
   "metadata": {},
   "outputs": [],
   "source": [
    "class AttentionKGaussianWindow(torch.nn.Module):\n",
    "    def __init__(self, K=20, hidden_size=256):\n",
    "        super(AttentionKGaussianWindow, self).__init__()\n",
    "        self.Wh1p = torch.nn.Linear(hidden_size, 3 * K)\n",
    "        self.K = K\n",
    "        self.bp = torch.nn.Parameter(torch.zeros(3*K))\n",
    "        \n",
    "    def forward(h1t, kt_1, cx):\n",
    "        \"\"\"Compute Attention Window.\n",
    "        \n",
    "        h1t: (batch, hidden_size)  Wh1p: (hidden_size, 3K) -> out: (batch, 3K)\n",
    "        -> alpha, beta, k : (batch, K) x 3\n",
    " \n",
    "        args:\n",
    "            - h1t of shape (batch, hidden_size): hidden state from the first hidden layer\n",
    "            - kt_1: previous kappa\n",
    "            - cx (input_sequence_size, embed_size):\n",
    "            \n",
    "        returns:\n",
    "            - wt: window of time t\n",
    "            - kt: kappa of time t\n",
    "            - phi: phi of time t\n",
    "        \"\"\"\n",
    "        \n",
    "        # (αhat_t, βhat_t, khat_t) = Wh1p ht^1 + bp  # output of the first hidden layer\n",
    "        alpha_beta_kappa = Wh1p(h1t_1) + bp\n",
    "        alpha_t, beta_t, kappa_t = alpha_beta_kappa[:, :K], alpha_beta_kappa[:, K:K*2], alpha_beta_kappa[:, K*2:K*3]\n",
    "        # αt = exp(αhat_t)  # importance of the window\n",
    "        # βt = exp(βhat_t)  # width of the window\n",
    "        # κt = κt-1 + exp(κhat_t)  # location of the window (how far to slide each window)\n",
    "        alpha_t = alpha_t.exp()\n",
    "        beta_t = beta_t.exp()\n",
    "        kappa_t = kappa_t_1 + kappa_t.exp()\n",
    "        alpha_t.shape, beta_t.shape, kappa_t.shape\n",
    "        \n",
    "        # φ(t, u) = Σk=1->K αkt*exp(-βkt(κkt-u)^2)  # mixture of K Gaussian\n",
    "        u = torch.Tensor(range(cx.shape[1]))\n",
    "        phi_t = (alpha_t.unsqueeze(2) * \\\n",
    "            (-beta_t.unsqueeze(2) * (kappa_t.unsqueeze(2).repeat(1, 1, cx.shape[1]) - u)**2).exp()).sum(dim=1)\n",
    "        phi_t, phi_t.shape\n",
    "        \n",
    "        # wt = Σu=1->U φ(t, u)*cu  # the soft window into c at timestep t\n",
    "        # The size of the soft window vectors is the same as the size of the character vectors\n",
    "        # cu (assuming a one-hot encoding, this will be the number of characters in the alphabet).\n",
    "        wt = (x.float() * phi_t.unsqueeze(2))\n",
    "\n",
    "        return wt, kappa_t, phi_t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Decoder(torch.nn.Module):\n",
    "    def __init__(self, hidden_size=256, out_size=298):\n",
    "        super(Decoder, self).__init__()\n",
    "        self.hidden_size = hidden_size\n",
    "        self.embedding = torch.nn.Embedding(out_size, hidden_size)\n",
    "        self.relu = torch.nn.ReLU()\n",
    "        self.gru = torch.nn.GRU(hidden_size, hidden_size)\n",
    "        self.fc = torch.nn.Linear(hidden_size, out_size)\n",
    "    \n",
    "    def forward(input, h):\n",
    "        out = self.embedding(input).view(1, 1, -1)\n",
    "        out = self.relu(out)\n",
    "        out, h = self.gru(out, h)\n",
    "        return out, h"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 361,
   "metadata": {},
   "outputs": [],
   "source": [
    "class CharToMel(torch.nn.Module):\n",
    "    def __init__(\n",
    "            self, encode_type:('onehot', 'embed', 'ngc', 'rnn') ='onehot',\n",
    "            K=20, hidden_size=256, out_size=298):\n",
    "        super(CharToMel, self).__init__()\n",
    "        self.encode_type = encode_type\n",
    "        self.hidden_size = hidden_size\n",
    "        self.K = K\n",
    "        self.encoder = Encoder(encode_type=encode_type, hidden_size=hidden_size)\n",
    "        self.vocab_size = self.encoder.vocab_size\n",
    "        self.H1 = torch.nn.GRUCell(input_size=self.vocab_size+3, hidden_size=hidden_size)  # NOTE: 3? window size?\n",
    "        self.H2 = H2 = torch.nn.GRUCell(input_size=self.vocab_size+3+hidden_size, hidden_size=hidden_size)\n",
    "        self.window = AttentionKGaussianWindow(K=K, hidden_size=hidden_size)\n",
    "        self.decoder = Decoder(hidden_size=hidden_size, out_size=298)\n",
    "        \n",
    "    def initial_states(batch_size):\n",
    "        h1_0, h2_0, h3_0 = \\\n",
    "            torch.zeros((batch_size, self.hidden_size)),\n",
    "            torch.zeros((batch_size, self.hidden_size)),\n",
    "            torch.zeros((batch_size, self.hidden_size)) \n",
    "        kappa_0 = torch.zeros((batch_size, self.K))\n",
    "        w_0 = torch.zeros((batch_size, 999))  # TODO: window size\n",
    "        return h1_0, h2_0, h3_0, kappa_0, w_0\n",
    "        \n",
    "    def forward(features, labels):\n",
    "        if self.encode_type in ('onehot'):\n",
    "            encode_out = self.encoder(labels)\n",
    "            for t in range(T):\n",
    "                wt, kappa_t, phi_t = self.window(h1t, kappa_t, labels)\n",
    "                self.decoder()\n",
    "        return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "window = Window(K=20, hidden_size=256)\n",
    "wt_1 = \n",
    "for t in range(T):\n",
    "    # generate next hidden state from mel of time t and previous window of time t-1\n",
    "    h1t = H1(torch.cat([x[:, t, :], wt_1], dim=-1), yt_1)\n",
    "    # create attention window with hidden of time t and texts\n",
    "    wt, kt, phi = window(h1t, kt_1, cx)\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# h1t = H(Wih1*xt + Wh1h1*h1t-1 + Wwh1*wt-1 + b1h)\n",
    "h1t = H1(torch.cat([xt.narrow(1, _, 1).squeeze(1), wt_1], dim=-1), ht_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# hnt = H(Wihn*xt + Whn-1hn*hnt-1 + Whnhn*hnt-1 + Wwhn*wt + bnh)\n",
    "h2t = H2(torch.cat([xt, w1t, h1t_1]), h1t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.Tensor(100).chunk(3, dim=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = torch.Tensor(10, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ln = torch.nn.Linear(3, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ln(a).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "kt = torch.Tensor((3)).view(1, -1)\n",
    "at = torch.Tensor((3)).view(1, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "kt, at"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "u = torch.Tensor(range(44))\n",
    "u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "K_U = torch.ones((3,44))\n",
    "K_U"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "K_U - u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(at.unsqueeze(2)*((kt.unsqueeze(2).repeat(1, 1, 44) - u)**2).exp()).sum(dim=1)"
   ]
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}