Transcrição_de_Videos_ou_Audios_com_IA.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/fmbento/Jupyter-Notebooks_ad-hoc-solutions/blob/main/Transcri%C3%A7%C3%A3o_de_Videos_ou_Audios_com_IA.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Transcrição de Vídeos ou Áudios com IA\n",
        "### Optimizado para o Google Colab\n",
        "<br>\n",
        "Última revisão: <br>\n",
        "28 Janeiro 2024 -- uso do **Faster-Whisper** em vez do Whisper (MUITO mais rápido)\n",
        "<br><br>\n",
        "\n",
        "\n",
        "### Na primeira execução vai aparecer este alerta:\n",
        "\n",
        "<center><img width=\"500\" src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA74AAAFgCAYAAACc1aV3AAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAAFiUAABYlAUlSJPAAAJ4uSURBVHhe7Z0FgBxF1sdf3ElIiEECIbg7geBwuEPw4w6HQy/IHXq4B4K7BviA4O4ePHgIEggkQNxdd77+1U5tajs9u9O7s8vu8P9BZ2d6qrtLX71X9aq6wYMPPpixLJlM6Uf/11Pu+5yxNnHMB3bVty/YzMbL2FItl7AlmzXO/hijQQdr2ry37bVSN1uqWRPLEWoxyj+/JDpm2I9f3WmvjW9uczpub0evs7K1js42dL9HLJxuUycPsXc+ucPeaHag7b76BrZF16WsydxJNmPcIBsw9BX7ZeES1qJle2vfopk1Lrtwjk2fNtomzO1kLdvvZmdsspa7b3I8/7Bvv33W3v59jo3t3McuWrd79nxAJorr/DH2+TcD7c2xI+yXRkvYMm3aW6vGDaxBg9IgC+ZOtKkzxtvI+atan/UPso07drSOTUt/s5IFZtO+sReHPmvfL1jBluu2m+25bBNbOPYde/DHL23BEr1tw+U2sw2WapK9oJQpo16xz4Y9bwNnrenuuUH7Jax9+SAxyN+MTRvxlD03YrB9OXO+NWre2Tq3bGYNo3i6uC6cbTPnTLTRM+fZ9LZH23/X6mmrtm3hri7H3Gk2JcrDPyb/ar9N/cN+WTDWxkybaA2armWrLbO17dJzOWvVKCqrbPqFEPWDhtPetEZTHs9+E0IIIYTIH2/75Eva8HHil/v7hfdttM8++1zAh5ISDMzFiRvBlplvCxZOtrFjhljDNrvZVj13sT1X3MJ6L7OpbdptE9tkmdKj19K9omNt27DTkta6caNFRmoCiz2jHPw2zyaO/dyGz2xk81sub+t16mDNorMNopSURMbigtnjbNyEb+3Ncb/YzNa9bJ2OS9uyrZpFtts4Gz/iFXtpxmRr3mJb27LHbrbXKtvaFt02tl5dNrD1Oy5jbWeNt6nT59usFivbZt06GzZoclyn2/jxP9iv0xZEz1jdtunSNns+ILMwMlx/sEGjv7bv57S1Tu23s/3X2Mt26BHlzdIb24Zd1rNVmjS05tOG2tfzlrLVu65ly7Rq5QxDBwXTvKVlJv9gI6YvtAmNozBLNrcZI16292Z0ty5d17B1uyxlrWIRnDP9Zxs96Uf7dn4nd8+lI+O+hb9nIuTpfPtj+ED7dOoUm9t8Q9uk+7524GrbWO/uvWzjrhvYBu2WsY4210ZNGm2Tm29gm3Ve0pZqXmpNl5QstIXzZtjcKLmRxWwtWi9lnZda0VbsurZttOTS1nLmr/b71Ik23lpbt449rX1jGb5C1DcazP3FGs4Zmv0mhBBCCFF10hi2VTWCky7DzPT3a4jBm2T0YowmGqSNW1iT1kvbKk0b2vQp79knI7+yL8fMsJnzS3/OeV2AD5NP2HJg5M6fa7PmLsieMFswa6yN+OMze+fnz+wn62LrdVjGurVoaQ2i+y4smWdz5k2MHrjQmjVtY62atbRmjaJnLpxv86f9bj//8Kl9Pn6U/UIUMqX5kMnmR9KRKcnGNxt28WOhlcydblMXRsZxgxbWtPES1i7KJxd+3jSbNGqoDR35g30xnXtE/y92n0x0tLall1jJOjTN2IyZ39tP0ybY8AkjrEWLLlG62lv7BvFrSq9jsjm6oYt/RWkoPaJ4lsyK8qbEFi5oZs0at7E2zVtZ04ZcO8/mRnnz68iv7evffrLRUTxLYvGcN2uS/TFkoH38w/f288SpNnPegtLfoueXNO1snaO4tm/c1BYunO2eUbKw/PU6dOioDwfySIcOHTp06NBRDEfc/srnKCTcLjwqIp8wSXBJ0mWRWuOOBgMGDFjs94oTGhlL86balJEv22O/fmnD55dYg2ZtrHWTxtYoZma3a7GcbdD9YFtnSYuMq8ruuzil4bHocHW+y14b/YsNbdDOOrdqZc2yU4glC+fanLlTbdb8xta05R520Frr2AptW1nzyN5cOHusTfntJbtj+Kc23tpGxmMbW6JJ6XRpZuEcmz1nmk2dPzMyUpe3Lu13sdN6rWVtot+YLJ03+Tv7YfRge3HKKBeeWedZsybalLklNr/pUrZcK+acI5qsZ+t0Wce26t7ZWkQGpc0ZZu9886S9N2GsjW3azrq0alnqOh0Z3wvmzbaZc6fZtAULbLatZvuuf4BttFTg6pylZNaP9tnP79o348Zb83Zr2/RxX1uH5fvY+t2Wt56tF3fEnjLqVft82Av22Ow1bN/1DnSuzktW6OpMnpbYhKH32yOjv7Oh8xtau+ZLWluM9IiMy9PpNn3eDJtuLWxh2yPtjDUXuTrPnfm7/fFNP3t8Vhub1aiFNWsSzuhjGI+36Qt7WvdOW9n+a61lHeXqLES9o+G0t+TqLIQQQvwFqeqMK+CRm4Z8glclPkmXNNp7772dq7OncuOU9apNrFlkKLVp3MSaLJhjs2aOs7GzImNo9kSbGBxzSxpb17abW/fWkX24yDIqR+XP4/esq/OMP2zkguk2ZU50/8gInTh7hs2Z18RaN+1hK3bqZZt3X9tWWjIyehs3dIlt0LCRNWrS1prMn2vz5k6ycbPG2qjZk2zynIXWKLO0dWu/snVrvMCaLmxk85utZL27dXYu1ER1/rRf7Jdxn9sbE3/OpmdKZLDOi2Iy3xYunFqWxonzOlqH1svYah3aOTdpa9QsMoBLrOGC2TZj9jgbNXO0jY3iOXt+M2vTbDnr0XppW6ZpYxsxp6mt3nVNW6Zl4OqcpUF0j8zsMTZl6s/25axpNq7BqrbRcqvY8ku0jgz+bKCAUlfnYTZ0Aa7Oa+bh6kzuNLTGjZtag4ULbMGcyTYpet7vs8dHaVoYGeqdonJbPiq3JazlrMk2sdn61rvTIlfnBpFxy+2nz51j02dNsLEzyNcJ0bWzIoO5mbVtsYqt2mU926BrdI82za1x9Lgq1FchxJ9Ig7m/ytVZCCGEEItRVcO4MqO4svumfW48eNmMby4DtGfPnrbssstax44drUWLhM2NhBBCCCGEEEKIWmb2jFE24fdP7PcfnrNfv3k4e7Y83gB2hm+S0duqVSvr3bu3de3aNXtGCCGEEEIIIYSoe4wZ/rp98sKJNnPqyOyZRWD8Nsw107vFFlvI6BVCCCGEEEIIUefp0vNvtule92S/lQeLN3Hl7YorrmidOnXKfhNCCCGEEEIIIeo2Hbv3tp7rHJr9Vp5Ew7dHjx7ZT0IIIYQQQgghRP1g2dX7ZD+VJ9Hwbd++ffaTEEIIIYQQQghRP1iyy9rZT+VJNHybN2+e/SSEEEIIIYQQQtQPmrfqnP1UnsUM38rfqyuEEEIIIYQQQtRNkkzacoZvSUmJDF8hhBBCCCGEEPUWTNrItC2HM3wxdjF6hRBCCCGEEEKIYgAT18/ruvf4apZXCCGEEEIIIUSxganL4QxfIYQQQgghhBCiGHGGb/ZzGXJ5FkIIIYQQQghRXykpWXxyt8zw1TpfIYQQQgghhBDFAMZv6N1ctrmVEEIIIYQQQghRTHhbt2yNr4xfIYQQQgghhBDFgjdxsXU14yuEEEIIIYQQoijxpq52dRZCCCGEEEIIUTTELVxM3sV2dZYhLIQQQgghhBCiPhO3ahczfIUQQgghhBBCiGKinOHLbK9mfIUQQgghhBBC1Fe8SRtatmWGr4xeIYQQQgghhBDFQNz4LdvV2Ru9Mn6FEEIIIYQQQtRXyozewPh1uzrL6BVCCCGEEEIIUSyExi9H2euMZPwKIYQQQgghhKjvlGRN2tC0XWxzKyGEEEIIIYQQoj4TN221xlcIIYQQQgghRFGBVYtp683bcmt8QYavEEIIIYQQQoj6SmjS8pmjbI0vlJSUZD8JIYQQQgghhBD1D2/sekvXGb7Zz2VoxlcIIYQQQgghRH0mbtaWGb4yeIUQQgghhBBCFAuhiSvDVwghhBBCCCFE0bCw7V5lx4K2e9vCdvss2tXZ/5UBLIQQQgghhBCivrKw9RZW0mbL6G/pUdJmq/JrfGX0CiGEEEIIIYSo78Rt27LXGcnoFUIIIYQQQghR3/G2rf/L24vKuTqDDGAhhBBCCCGEEPWVJPu27D2+MniFEEIIIYQQQhQT2LkcWuMrRCXMnj3b/v3vf1vPnj3dceWVV2Z/ScfAgQPL7rHFFlvYTz/9lP1FJDFp0iTbe++9y/KM/BNCpEPtSBQK+j5fj+gT6RtF3UJ6hhDlidu2i73OqL4ZvzNnzrR//etftSqMFyxYYM8995wdfPDBdu2119rkyZOzvwghhKgt6K/++OMPGzRokA0bNszJZs/YsWPtlltucYrgvHnzsmfrBvRbp512Wlm/td9++9kvv/yS/VX81aAOn3322XbEEUfY+++/r0mIPJkzZ469+eabTgfceOONy9rTlltuaf/973+dXKhrbV8IUbsgT0OZWra5VX3lt99+sy+++CL7zeyjjz6yESNGZL/VDEOHDrXzzjvPPeumm26y//u//8v+IoQQojbAyL3xxhttm222sX/84x+244472lZbbWXXX3+9vf766/a///3P+vXrZ88//7xTkOsKxPuOO+6wp556yn3fYIMN3ADq8ssv776LvxYYZrfffrs98sgj9vbbb9sFF1xgv//+e/ZXkQQ66yeffGI777yzHXXUUfbKK6/YhAkTsr+ay7/HHnvMyYVDDz3Uvvnmm+wvQoi/Kt7WLVvj66lvRvAbb7xh48aNy34z9/nTTz/NfqsZ5s6da7Nmzcp+K/0uaga57QhRitpCeZD9DDyGs7yjR492hu8xxxxjr732mjVu3Nj22msvW2KJJbIh/lzoXzF4b731VvedWSri2717d/dd1Cwff/xxWRvi4PufDXUCDwDP/Pnzy9VpUR7yi4Et2ng+kxzog8cff7x9/vnn2TNCiL8KSTZuuTW+9Y0pU6bYBx98kP22CBSeadOmZb8VntVXX90OP/xwa9q0qW200UZOsRJCCFF7fPvtt85AOO644+z77793njjMnG2yySbO4O3atatdccUVtscee2Sv+PPBMB8+fLitscYa1rt3bzfTu/TSS2d/FX9FmjVrZgceeKCtssoqTqdghlIDIblhSQPt2ut4DGpddNFFzrClbXHwmaUE5CfgSn7dddc5nVEI8dem3OuM4pZxXefnn3+2r7/+2n1u3ry5U3YA12eEY03RqlUrO+uss5yy9eijj7qRYyGEELXHqaee6pTc//znP07BpQ/Yfvvt3dKTH3/80a2V3Geffcr6hboARi5rD5n1ffDBB2X0CgcD6C+99JLTKY488sg6VWfrEriF33///WXr4TF6b775Zvv73/9u7dq1c+eAz8zyXnzxxWV5yXrf9957z30WQvx1CG3bxBnf+mL8MtL/8ssvl7kI7bvvvrbnnnu6z5xjw4P6ZsgLIYQQQojFYaCLJQ4eNoXr1atX9lt5GjRo4AbCGFTw4CGoza6E+OuQZAfW29cZTZw40W1uAIzo/e1vf3ObnHgY3QvX/npwdTnkkEPK1vhceumlietp4q+wYZTeC0zW1rHGzv/G2rskCP/WW285V7x1113XhV111VXtgAMOsPvuu8+9ZqIyWEvMJg1cw7Xcg3VhJ510klufVNFaIDZ7wN2HTV/uvffeaq0bGjlypF1yySVut0TisPLKK7vBhnzSkTYfwtdvnHnmmdmzpe5KO+ywgzvP70nPLSkpcTM94bP4y3fiUKhOj3K5++67bZdddilLDxtp0CkX4hnVKXdPRfeg7ZBXFYE8+OGHH9xuo2G54zrK5jzjx4/Phqw6Po7UJe7NwW7p5GOa+kqes4kRZRCmlXabT1qTCNu5X1Obq9yZSaiJ/KxOW2CDF9aPcn+eQ9g0ecI9aZ++bPz17KBa1bYUl6vITuJB/oVlR/6Sz+R3RfD7k08+af/85z/LyRbuxfnKrgfSQX0L78FfvrNsplAbY1VHhlZGodpRdetMZVS1vPJdm5v0up2wzh100EHZkKXwnfO51sxXtW7E6zmzkrz9Add28pO8feaZZ7Khy8c7V3uGQtT3iqB8KecTTzyx7P7bbbedk09plo/VVD3Cw8/rdXje7bTTThXOjrdt29Y222wzW2edddxBueQqs0L1d74tVrfvzqVf8BsboHGeI5f+mQ810W8KUdeI27YNIoEcncuUVXA+n3zyye5zXebFF190whnWWmstJ5hoxLwOABdohOGdd97pdvkMIX10QnRAsN5667lrO3To4L576ARZx4uCCQjx3Xff3X2O/8Z6k/3339999owaNcopqhjguVhqqaVcPBDMjE7GYVMG3Pgq2sABgx93ns6dO2fPLIJ7s/kLkB8PPfRQudHPfCBPcS265ppr3OcklllmGfc7AjNOVfKBTh93r6+++iobYnHoxOgY2rdvnz1T+voS8oK6kQvy67LLLnPPzBc6S1zbn332WfedgQSUGDqGJPr06eN2/W7Tpk32TCl0UN54Ic8YjFhxxRXd95Dqljvkc49c8QS8Jvr3728DBgzI2VHjZkYHjGLQsGH67QJQYE4//fScdQPDjnWcFbUzwK2VdFS0qR2KIs9CWcqXsJ1TXigD7CKcaxkFig1HkiJW1fysSlugnVK3aJMVKVm58gQZyeARrsThTqlxNt98c1cmadx1423p3HPPdeWXS3lDptxwww3WqVOn7JlSiCPtj/j7+pEEeUM+oNAlwQwS8amo7nAPDBOU4KpQXRlaGYVoR9WtM5VR3fLCYAiN1ocffjhxpo9yYp030I4uv/xy9zmsc0kkyePq1I18+oywHMJ4x9szFKq+VwTlzuu/4hvGebg365F9Gnz+tmjRwn2HmqxH5AF5hl4H6H3kU5q+PBeF6u8K0XdT75iwyCXzMfZnzJhR1t7j7TlfPaOm+k0h6hLYbtgVHL7d1ssZX4TrO++8k/1WqhxhuCJINtxwQ3cO4YUrdFzRIPEobL4h0/h5JVIczvsOZoUVVnCjn/kyffp0u/DCC8sEE4owBvbRRx9dtvEKoFQiVFBK4qB4Ej4UoAh4NsAIFWt2NzzllFMSZ7erC3nIIAGdm89Hnk0cws6GfKKTJ89CqpoPrNdjkAEjgtFmD2XGWh7O87vfuAJIP/kQGr10VKuttlq5cOQX6fEu8lWBVyeECkycxx9/3O66664KO/5cFKLcKQc6vsruQTzpNONthLzhVTD33HNPWRqSyp0ZADr5J554IrXs4Bkoe7mUdXj11VcrVPKAtDIDGXbeSWnF8GCAqyplAsSDAcGK9g4gv5KUlerkZ9q2wP1vu+02l7f+WTwDpZUjfB55gqIXLzvaIe3RG720I4wnBgHCV+5QdrRv2nlVYQa0ohkL2hm/x+PI+WOPPbasfpB+Nh0kjaHST3mgrCZtakPdoa2FdSdJZnAP6lhcvuUDZVAdGVoZhWhHhagzlVGI8qoqjRo1cn0+bYUZ8RC+cx6jOhwALHTdqKzPqIyazj/Kk3XnuYxe4N4VpaGm6xEztQxue5Zddllr2bJl9lvVKVR/V4i+m3PMOJPXuUCvrai95wP1tTb6TSHqCmGbdYZv2k7sz4adMQcPHuw+00jp1PhLZ4DLkgchTdg4dBQYYIDQC98DDDTyUMBjTOcanUuCHQVxBQQ6S4wgjAwUGzZeYRTYK5AIOlyeQsFCp8Xor3ctIiybaPHeYDbAwN3njDPOKBNQxJWNUuLCid2mmeElHK4sdJRpYJSdzgC4B8/k2cSB3xDgvIMS2GyC0dfQiKpqPrRu3drN3Pft29et0fGwYQXp4Dy/Ew64hvT7MmOUk+88/4UXXrAhQ4Y45ZM4AB08a8CrA26EH374oRudxcMgzAugw0j77sBClDtGCKPtfvOP+D2+/PJLp0D5e+CSRTpCGFR67rnnst/MdeafffZZWbkzuIAiAzybGTnyIA3xZ+y2225l+clMK8qFf0YumFVhhsKnlfDEjTj6/Dr//PPL0prLMM0X6g8GDLsHE0+UB7wH/P2RJeGAnKc6+Zm2LaAYhy6UPOvdd991dZ6D9ogrpQclKlTAeD7X+3MMKhI/FFpmB3DvJA98mrkf7aw6xNsSeeXlApB/U6dOzX4rHfjErdO3E9od8cJljzTyGeXbx5G6Hy937kEbjW+SQ7/iZQbtyMsMwlHXqHNpqK4MrYxCtKPq1pnKKER5VQf0ArxbaCv8DfHn2QjJ9/E1VTeIBzNo9IO4tzNjmQ+1kX+44YeDtb7foA5Rl6hTtNOKqOl6xGueQhdwBjSSPOXSUh357ClE340eTthQ9wxlY77tuTJqu98Uoq7gbd3F3uNbH0CY+kbLCCxrEzx0JszQAmFozHHosEI3KQQLriMelKzQaMGYDkd5K4NRybADWXPNNcsJaEbVcHfxMEIYrjthAwY/osfMDkIIA9ZP07N7KcKZw0OH7EeDPRj4CF8UdFwK07ir0NlimKLMA7NLPI9nA+nh/sws+vsSbzpQT3XzIV9IN+kHhDWzULwqxOcX53AP4kX3HtbKpFViPRgguL16RcnnBbNXGN1AB5j0qq2KKES5Y5T5wQbSfc4555S7ByPkKHl+PTzlQ0fr5QDxfuSRR8rKDUWamU4/G0JaaW/MFPu08nyUmHxlCfnOLJR/Booc8fT5SVwZmAqfkQRppRwBV1jcx4ibr2PkFzM5XtmiLldn0zsGbHAp822A/CV/tt56a/cdqP9hvaqN/AxZuHChewazWCj0tFsfX6Be8boU32a/++67cm0W5TJcy0Y5hK7MlA0zz8hdIF0oZVWFuJKvYVviVT/E2zNmzJhyLtfIJmQHaeRgyUv4+hfugWspXiVAHFFkQ6g7oesrcQjXC/KXgUPOe6hrXJcvhZChFVGodlTdOlMZhSiv2qQm6gb5j/GKTN92223dGlLfhiqjNvIPPcl7s1DO8b6HOkV6qSe5qOl6RLqq412SRKHkcyH6bvRO2rMnrmfk254ro7b7TSHqEtRj1yrDCl3XKzeNMDQoEATh+lyEhHd3BowAr3iEoKz66xDAKFceRta84YsRne/IrMfPvgDGNyPEYb4iYOhEUBg5cCUJZy/DEb9NN93U1l9//ey3RdD54uLo177xnCSX6aqC0epncngWHavv+EOWW245p7wBQjx0RapOPqSBdPuBEMoqaUSUZ1FXPIRPqhf5QHqT1sVi3ON94OHVFPluNlKocmdWwnfiufKCuO+6667uN465c+c65Qpw+6c9eFDwvKISQrtAGfQQ93yVEoyYMM5s4OE795COHTuWczWLw2i6L0OUPj/gFcKAVVjuccM0DT169ChTDjzU16TnemojP0OomyidHCigSe8DJc7+1R/UlXADE2ZRwjV7eMOwTj+EMmFmx7fbww8/PPtLelDgkvKjIu8U4u9nwTni+zhAkyZNyq2RxKAPIV2+7mCAcI942fKdfsLLf8JT5/KlEDK0IgrVjqpbZyqjEOVVm9RE3aBs8jV049R0/iH7Q6+3XH0P7bQig6um61E+MCOMwUZckg5+C2eNCyGfC9V30/ZDl3mM8CQ9o7L2XBm13W8K8WeSZN+WDkfVIxiVDNc3MHIaKhM02NDdGUPg119/zX5bBMqGb9gInFD4hcYDGy517drVfc4X1gN7xY0RxcMOO8zNbLLImjUVnMsFhhLKpAehRGeRBDMx4aYaGFqFAmXXu/OQF373y/iBK2Q4Ox6OYFYnH9IQPp8OnJHWpLgy2uyh8yvkOjKg7oXlwc6W+c5gF6LcMWDDe6B45FoDxQYd3gUNbwA2LQFm3dkxHVByWEeVBG0u3jH66yqDcOFAU1oXfKATDtfbMkPDTGFSubNmypOmTApBbeRnEiiUuLXjhbDzzju7+uLzg/W6YTsNoQ4z0+BlKgodCj4ud957xA+S1AWQU7gHH3jggWW70HJQF8JZuxDiHyqYK620ktv9NYkll1yyXP2kfdHO8qEQMrQiCtGOQqpaZ9JQlfKqTWqqbjBbFzeeq0JN5B/yMCzbivqefKipeoRMSjIEq0Mh5HOhdDb0EX9PdE5m+QtNfek3hahJFjN86/KML3HD6A1HY5M6e0Zh/SgWguTtt992n0MQTKGBzCwynR4uz+F6hrhhnQ8It379+rlO08PoHgYfW9zTYfEXl21vYHv4Hs70MIuYi5ocJfcCuDpUJx/yhTIrtAFbHZJGi/OhEOWOwoG7mYfOM5y9y4fQ5Z9R5fA5cUJPC54bPrsiUA59GyaOKJBpIc3hyH1dpTbyMw4zhrjr7bPPPm59Fq/oSNO+MHR5zRtGMHAtswTe5XPttdd27rm05T+rv6Dds7srbqPEFQM934E08jWcvaionZAHfoYKuDbfWapCyNCKKEQ78lS3zlRGdcqrNqmtupGWmsy/eN/DpEBVqcl6FO/3MDjjMpIBhm7dupV5M3FUNDtaCPkcz7+q6mzhgADp8G7ShaS+9JtCFIokHWVRy1pYYgsmzbLxr1bvPXA1CWsgWI/rYZaWtZzxkSpcgcLNBzBqk4wjZga92wkL+hEIjGzxGXIZ1vnAmgncAXlND2tb4zDjyQwoazh4zUFdI+wQMOZYr4P7UmVHXOjXdD7EFRWMbHYrTIpbeMR38RR/DlXt4OPKBq5lSeUcP8IdkIsR8oQdStmsBEjrMccc49YX4snCwQZzKPS5oDz2228/t+kL6wnjiiNK+MCBA91sDq+VKaSBlA90ZDw33CmZNetszoOBThoZIA1dE/8MCiVD86E6inIh6kxF1JfyqqvUdv7V1XrE2tPQlR9DMWxjwEAEr5zz3kwc8Z28/8qo3xR/ReLGb9nmViXzMjb7t7k27Yu648YWh5nYcPv1fOEaRh7j4Arq3Vdw/yAMRq/faRCjuqLRwspAULNmBHcS1nLQAbCNfSj4EcxsruDLIe7OgyGeC0b8w1FCRhMLRXzUlxFcv36nooPBhDhVyYd8YTQ+vA8dI7sWJ8UtPMJdPAuJn4FJSyHKPb4+k3qcy/UuF6GLFgMR4W66cXC78/BsjnzgGX70nHRUZWaMdIZ1lAGPE044IbGswyPcAbk2qI38DInvpI5bJLOzeFcw08DB+XyUW9oH7/Jl4xtkKIo2csArQCjhV111VTk33dqAeo0M8aCYsaMqs2EMZJJGZj9zKWpp2gnnw51nuTZfw6CQMjSJQrQjKGSdSaK65VWb1FbdSENN5x/XsW7UgydHVajpeoSrODO4Hq+zVYdCyOdC6WzVmWnPl/rSbwpRSEK7gs9lEqhkYcbmz4hO1lG7l5EqRjWrMrvANbw2In4to/Dh7ra43PpNHhBm/jVJacDVidllZo85/AgtAocOAIOPnfvCbf3ZMc8LXIzEcAMJ1oDERzU9rO8K30HM7GqhCJUqjDl2iUxjlFY3H9LgZ+0BL4CqdtzVhTSG68NwuaI886EQ5Y4CwzM9rG3PdQ+upz1xMNjj2wbu6d5dm3VMufKS8OEupqyNCl3BKoJwYZmFeZYvKAyhmyFrysKdf+sKtZGfIbRTX5asHQ0VxXygDvs2S/ulHaNwEhcUbZYuMIDlZyVxt/TKbm1BPvo9GYgX6wbTyOl4O6Gt5fI2IQ/Y7NCDR5FfD18Z1ZWhlVGIdgTVrTOVUd3yAgy7qs4UpqG26kYaCpF/FUHfg5zy0G9UZQC3pusRsB7Vyx6exet9qjrYDIWQz4XS2eIDWcSn0NSXflOIQhP2vWWGb8OmUcNbrqE1XjTwV6fgfbzhWl3cWXFnZlOBpAMl4Mgjj8yGLnV3TmrguHp4ocW73Hg3HuDmXBVDks7w8MMPdztLc4QjtR4Ebfg6JYSqXxtEx8tMswfXbr8zaAjX4FLkR5zpDOgUCgU72Ia7Y/Pe3XBThBA6HmaFfKcH1c2HNLCLolcAmfUYMGBAzs6QzoSOqTowa0364tBpYkh6qD/5vmC/UOVOXnuliJm4pPfvcQ/Wi/EeWI6wbCh31nB6nn766cS8RDnilQ4eFJ18XcfZLCZc9817KJPyk9fq5OqUSSODVj6tjPyz+VJYB0NQWsKR9tqiNvIzJFwzRltImq2iDeTaOZiBP99m2fk7VOw9eMr43WwhfGZtgIzw5cxAWZKSibGSFHcPu5l6hRejArfuJKOU835Gm/D+lTH5UF0ZWhmFaEdQ3TpTGYUoL2YJw5lCNiWKQx4Woo3XRt1IQyHyryKQoeGSLvqepEGUyvK3pusRoKshlzzITPqvNO0mpBDyuVB9N3I13Pk7V1wqa88VUV/6TSEKRZLsdj0JPzRo3Mgad25tyxyUXtmqDRiZ8ut2abRsOlXRLomECWds6cCShBGClPcfAsLECxoEmx99SwOuPKGyg4ERN7RYY8HaHE+XLl3KjRRjyHEA8eG9tLgaeqOQWRmUqNtuu819BzqDcNQRSM9pp53mXqNx7733puoc6MhZ1+HzD6OOdbgYUr4iER/ylc0s2DE5fEYh8gHCNNFhEo94RWaknXeLenCbZj1U2DmwIyEvZz/kkEPcu/TC1w+kBeOWd9/5Dpz4MNjC7si+k8AQD+OUD4UodxSY0IuBjVCo9z7P/D1wKwfKN+wIGQ1mTZT/zkAQL+wP1wVRBmx05NPK81lflu+upcz6h7MW8fwkvRhguMlV1OkyoxAqm7feeqs7wldIEe8HH3zQ7UrMelXKqTYpZH7m0xb8bAiglPOuSMoc+MvzyddcsgBF0Bu1DDZS18K4AkZH6N4czhbVBshlnxek46GHHipr6+QHMumMM84oN0MTh3YSrolEXqDY+nyhDr7yyit22WWXue9Ae06z50N1ZWhlFKodVbfOVEYhygt5Gsbzsccec/npIc2UoR+4zkV85phBTC9fPbVRN9JQiPyrDAag/SAKXhyUN2tyfd7kk781XY+Aus5aefQz4F7U+QsuuMAZpz6+/EV+IWtpU7kolHwuRN+NzsSO+h7iEuox3Cuf9lwZ9aHfFKIm8P1ugxtvvDFDg+KEb6T49NclEBrnnXee6+wAAXPzzTeXc9lIAmHN5gpeSWPDFoQkI3QhzPTieutBCD7wwAPlZiNDGFk9/PDDy4QPLxTff//93WdgtJQZad8xcz9m/1i7xToRXIK8MOO36667zr3nL+T99993ay/CnRtZb4wCwPN9hwJ0Agjr0O0N2FDqpptucp95Dh1mvuvHgE6FjSI4QogHygMKcBi/eDwKkQ/xvPYgvFlDxLohYBQV5TFu0GIU497DoEnY4fpNG3huZbB5Fh1fmtdFJN1/4MCBrtMCOjw65BVXXNF99xSi3OP5DrnuwSZfvGw/bBN03GxSwgCCh3SwUzq/hWuYOI9x3adPn7wNX0h6RmXE2xkkpdXHlRmIsMNOG9d43WODmSSZcOWVV7o1d8BrolBWMEo8hcrPfNoCbRKZF84solBR32gDYdl7kHW8tg3oB9iA5pxzzilrL1zPM2i7DEBiuPn7oCzz/u1816fF2xLr8TEG4zD7Sd2EeFvh2SiWlIfH5yczTV6mhDDYRdsL8zSp7pBW3BiZmQrbIEo9ymH4KpJ8qK4MrYxCtKPq1pnKKER5US8xGq6++mr3PR+S2iLPI61+SZMnXseqWzfyreeeUIbE+7ZC1feKSGr3lRHP35quRyFJ5VMZ5BkDBMhQPnsKJZ8L0XdTluii4SaulRFvz/noGTXZbwpRl8AOot5y4DXk/rKh1eT359j4l6PjldKjrkEjxK3Ig+Gb6916IQgVZn09uDvTacXB1SUcrWTGIy4o0kAHSGb7EVQ6kiFDhjgBxEix76gwNhipZNQ+Dp0Bs6ShQsl1jOqGAtS/FzdfRSkNCDyEIxvc8NlDPBhMCAU8a3Xp/MN4FCIfmIFiU5jK4LnXXHNN2Wynh7qDO4/vzHkW7xfM1+hNgmvpuHJBJ4QCUJX7F6LcyXc6w4ruQdzYsIJ39cUHgpipuuiii9zvPg3kH/kYKgEoNWxuxIh52g6RZ6AIhu0zDqPqFf0OpJUBE5RFj49r2HmjgFA//4zOu1D5mU9boMwZ3EPh8dBOmRWi7Hk+bTWUd+GaM57L77RHXy+4/r333nP1klkMX4do18Q3X6O3UBAvBmdpAx6fn16m0EfwWiYPA23h7u9A3SFN4WAgaUVGhbKNulUVoxfI7+rI0MooRDuqbp2pjEKUF/WSV98l9REeyil8V3sSGJO8AzcsiyRqo27kS6Hqe0X4ds/sXq68ob0zA5iLmq5HIeT3gAEDys2QVgRxY4CO5W/x9BVKPhei76afIg/D/iwObSDM46pA/tX1flOImoBBvkZbLLv1BZPfn2dTPp9vs35e4I7dLyhvPPzZ4KKK2wwgpDA88nGvo6EiyHidDrPZuG9g0MaFCuswafDeVQjhjOtSrobOxi/c07vDIMji62tx22WGGaOadTmsvWLNC/HBJYrZDIwThGWuXQ4RbhhRCFLugRDl9T0IIww8BCQGVq61gOzKSqfDbChrORmhjRs5lUF8MfLIEwS/jwewCQhu1IwG4n6UtJ61uvnAOXbexkWNTpLrSQMKCfkeukYTP1wLWW9D2ZFu3Hf8s4gjMwYoCbnyPAk6hDfeeMPVESAvGZUlTn69DXHivngm8HqmcJbBg1sdm6wBcWWnaz+qH1LdcoeK7kE9YIaBMs1VHzjPe64pX64j31FifF6iJHAPyqaqHSJtGXcv2iTeGcSR+/t8xGBgpBzPAEhqZ0A9xyAkLri0UybUMdLAGn7yipkJZizTxDXezun8w41vPIz0+ziussoqLp7hbp1QiPzMty1Q9pQx31GAeQ71DZlG/UdxQ4H39Zly4BVwPs48h8E/DA3qC88hL5Ch3Ad5gHcFrv2sS0tDvC2xHCLJKGNWm41rIKmtIGvIy7DuEG9kO7Mdp59+unuWb2/IAdIfb2/k5Z577umuI50c1B2fTuohy0XSGKNxKOPqyNDKKEQ7qm6dqYxClBcylQ3WQrnr2w8znHiuUH7+HkltkbbFeTyBMGroI7gH9Z0BAtLsqU7dyLeee0IZQp/Jc8M+pFD1vSK4H+kij6mjPIf0IjePOuooZxxyP98uk/K3putRCPel3pMvbDBFnP1GZMhFXp9Im2PmnfyhzHPJ1kL1d4Xou6l35GFYz4kf7ZnrkcvULcoH4u05Xz2jpvpNIeoSeE/4+ouMgwaXHnVFZsrnC2zOmIWWif5bMLXEbp1xoftRCCGEEEII8eeDmzLu2qxhBpazxZeICSFK8a7O4F9B1rDN2i2t2z/bWM8zouP0FtZug/xnwoQQQgghhBDVB1djlpTgkhmHc8xgeaOXjc9C12ohROXIyhVCCCGEEOJPBLdmXOj9Du/hTsu4srO7eP/+/bNnStcV48IthKgcP5gkw1cIIYQQQog/EWZz/QaCbMK57rrr2s477+zWfLNOPNw1mk3Cjj/++MT9RIQQuXGGb5JLhRBCCCGEEKLmYbMuNgFjMyxglpeNwOI7wLPDNmsXa3I3cSGKjbIZXxm9QgghhBBC/Hmw6yyzu+za3K9fP7cTOLssAzs0s/M0r0LibQPs6C2ESE+DqBFFtm8msoRLomO+jXt6tl3+zvnZn4UQQgghhBBCiPpDuKszfxlcWmyNL680EkIIIYQQQggh6iNJXs0Np7w5y367e2Z0RH/vmWPTv5XhK4QQQgghhBCieGg4Z+R8m/5NdAyZbzOGLLR5E2X4CiGEEEIIIYQoHho2X7aJtVkrOtZsYq2jg79CCCGEEEIIIUSx0LDdti2t+5Gt3NHt8JbR0Sr7kxBCCCGEEEIIUf+Ir/NdbHMrIYQQQgghhBCimJDhK4QQQgghhBCiqCkzfJO2fBZCCCGEEEIIIeojoY3bUAavEEIIIYQQQohiBZtXrs5CCCGEEEIIIYqaxQxfzQALIYQQQgghhKivJNm0mvEVQgghhBBCCFHUyPAVQgghhBBCCFHUyPAVQgghhBBCCFHUyPAVQgghhBBCCFHUOMNXG1oJIYQQQgghhCg2vK2rGV8hhBBCCCGEEEWNDF8hhBBCCCGEEEVNOcNXLs9CCCGEEEIIIeo7cdtWM75CCCGEEEIIIYoaGb5CCCGEEEIIIYoaGb5CCCGEEEIIIYoaGb5CCCGEEEIIIYoaGb5CCCGEEEIIIYqahtrJWQghhBBCCCFEsYLNqxlfIYQQQgghhBBFjQxfIYQQQgghhBBFjQxfIYQQQgghhBBFjQxfIYQQQgghhBBFjQxfIYQQQgghhBBFjQxfIYQQQgghhBBFjXudUfwQhWPgwIHWs2dP+/jjj7NnKiZt+LpOsaXnr8aCBQusf//+tvLKK9udd97pvgtREWrz6Zk3b57Nnj07+63wXHnllbbFFlvYTz/9lD0jknj//QXWoMFkd1x88Zzs2eKAsqcOUBfqA5MmTbK9997b/v3vf5drG3ymvVSV6l4vRD7UpMxVH5s/STZuw2nTphnH1KlTbfr06e6oi1DAFHRFhzp2IQrLs88+a7feeqv961//ssMPP9waN26c/UUIUQjoc4855hjbd9997ffff8+erT/MnWv21FPzbf/9Z1iHDlOc0bjqqlPtlFNm2YcfLrCSkmzALD/8UGKbbz7devSYai+/PD97tmIwQr1BinGaRBjmoYfqnmFTkQ6z5ZZb2rXXXmsTJkzIhhZJ0D5oJ7SXquiq1b1eCFG/8Hatt3U5GrZp08b80bp1a3fUZdZff3076aSTEo+DDjrIpSMtuUYWxV+TYq8PpIm0kUbSmgsGw1DW+vTp4xQFGb1CVB1v+DBaH9KwYUNbYoklbMkll7TmzZtnz9YPxo7N2KGHzrB99plhjz02P5InpR5jGLc33DDXdtxxul1zzRwLHUUGD17gjNcRI0rsxRfnO8MZMFbrsuFaKOI6zLHHHmvt2rWzm266ycnkTz75JBtSxGnatKm1aNHCOnbsWKX+KNf1udqmEKJ+4+1aDm/rNmzWrJnFj7rMRhttZH379k08jj/+eOvcuXM2pBCiOrRt29a561x22WXWqlWr7FkhRCGhbd1www2RsfeQLbXUUtmzdZ+ZM80uumi2M3hhrbUaRd9b2COPtLJTT21u7ds3MCbVLr54tr3wwqKZ3Q03bGzbbdfYVlmlofXp0zTSObI//EWI6zD//e9/7ZlnnrHbb7/dzUxgAE+ZMiUbWoR06tTJnnjiCbv66qudAZuW6l4vhKhfJNm42txKCCGEEKn46KMF9sADpdO1u+7axF58sbWdd15zO+CApnbNNS0ig7i1de1aavw+/fQ8ZygDBu/rr7ex779va1tuKS8SaNCggW2zzTa2xx57RPn6kQ0bNiz7ixBCiEJS1IYv62VYN7Pxxhs7Nxb+hutovEvrhhtuaF999ZVbz7jGGmss5uIav892221nd999t82aNSsbonK43z333OOu9fd48MEH895kgXDEj46R61dddVU3wz1kyBC3WDsOcSOOu+yyiwu/7rrr2llnnWV//PFHNsQi0t47iTlz5rj0bbbZZu4e3Ou5557LuRlS/Jk+fiNHjsyGyE24ScfQoUNdXIkzx8knn1zlNOZTHwg7ePBgO/HEE12cuRdlSdrJgziV1cHKKCkpsffee88OPvhgt8FURWVTWRq9izNpIhxpJK1xl+fRo0e7EXHWnYVlk5SvSaTNoyTS5Fu+eeQ3hHjjjTfs8ssvd3E77bTTbK73tUyA67lPmJZcdZvv3Psf//iHi0NFYf3GFx9++KGb9SE86avsN0jTdpADrNH2+ejj89prr7nvuPiFpCn7tG0+Xk4cfOYcv+UD63PuuOOOMrnmyxo5kA/kLfX9559/Lle/iPvbb7+9WDzS1OVc9Yv0cZ6lOHDmmWe6796tMmnpgZdFnH/33Xdd/MivTz/91P2ei8mTJ0dG5zXu2T5/33///XLtIKSqcnjhQovSOd8ZtXDEEc2sW7fy6kSvXo2j5zezHXZoHNUJsylTSuMQ30Bq4sSM7bTTdPv737OWcQSfC+HyjMv1uutOc/c67rhZUV5nf4iozLV69uxMVM7zbO21S6//29+mRwb74muWCwWut2uuuaZrP7/88kv2bCmUB+US1sFcOghlTZkfcMABLqyXnaGMD0nTxySRT98akiYtccJ2EepoPCcfOR2/3rs452qbUCi5nou0Mo38O/vss8tkVxr9I1dep5HNacJCdcobKNs0+kRt1kdIK3PrUh9bXf20vtIoamwX8CEspF69emU/1R2oFLiooKhvvvnm2bO5+e233+yoo46KOqrXbdddd3WjqYyqPvnkk/bFF1+4Sse6KioGv3333XeugVxwwQWuknXp0sWtvWLklft88MEHtvPOO9vWW2/tGtb//d//uTgRF9aN5OLbb791cRg0aJB9/fXXUef5N1tvvfXs+++/dxWYOOH6xLPC8Kyr7Natmzs3c+ZMu/DCC13jWm655ZzgXm211Vzjuv/++22ZZZZxcedegHD/z3/+4wxrjJzdd9/dhXnqqaecAN90002tQ4cOLmzaeyfBPc477zwnvAlP3FmvhusWStv8+fMT03P99de7ZxG/7t2724svvmivvvpqufglQfpwDfvxxx9dncB9aYcddoiUsYVOeSVvt9pqK2vZsqULn28aUTwqqg/kweOPP24nnHCCuy/x7t27d1l9GDVqlOv0mjRp4n7Ppw5W5EKMcLv55pvt3HPPdS78e+65Z1m8H3300UghW9uWXXZZF5Y44JJMx0saCLv88su78uZ5PixlQB0cN26ci+ell15qO+64o/utUaNGLk+PPPJIpxBsv/32rr77snnllVdcfW/fvr17ZhLIkTR5lESafEuTR75toVBQTzi/wgoruPjlig/ppsOdOHGi63SoC2yQct9997nffdslHjfeeGNZPAhLJ0I+04mStxtssEFZOyd+rOVDLjAAseKKK7r6xjUV/Zam7RD2f//7n3s+YcJ2+fLLLy/WLtOUfdo2H5YTLob77LOPW+tIW/V5GeZPEuQlCgxyjD6KNk++kF+PPPKIk6v+ebkgLGmhbqBwUvdJ45dffunqV9euXW311Vd39S1tXc5Vv7bddltXh7kvCsjRRx/tFG+MHJYTkDe0U9bUU3/JH5RylLcffvjBlRX5TXvmPsQxCfKHuvr000+7es+9kInUS9KHPNxrr73KlWFV5fD06Zmo3OZGaS2xlVZqGBmVzaJ6X77s6BZ32KGJHXpos6i8m0b9bWkf8ttvJc6ghG22aRK1ocZR3s+zn39eXDHjurXXbhTVqQX21lulyh5G9rLLLl5PwjD+Oozqxx+f59Yi42K9005NovJyQeybbxZGzy11wfbhw7gNGrQgKoP5Ub6W6ka//FLiZq579GgUlV2jqI6403mTjw7zzTffuLpAOa200kruHHWWdkld2G233crpIJQrMsnLQ19nGQCmfrORE4r1m2++6fo6XKlpd/75aeRnEvn2O540afFtgPXP9Fm0s6RzkK+cjl/Pej/acEVts1ByPYm0Mu3zzz93cUTWeH2UdoyuhzwK9dF8dVfqzF133eWMfmQ4spn8+Oyzz+zhhx9eTCbmGxbSlHcSaWWwl2n56rPE4dBDD3UDTZQteYIhS95joIb5mURamVuX+tjq6qf1BfLalzd/3eebbropw3HDDTdkog7QHXWRjz76KBMJ1JxHVGEyUUPPhs5k7rnnHnc+6sCzZzKZkpKSTFSh3fkBAwZkz2YykbDMRJUzc8opp2RmzZqVPZtxnznHb7/++mv2bPn7REI0ezaZqPNw4fr27ZuJOqLs2Yz7HDWYTCRgM4MHD86eXRSe9Hr8uWuvvTYTVfbs2UwmEl6ZAw44IBMJzExkqLlzxO2WW27JRJ2mi2PUCN15iCqACxsZxZm5c+e6c2nunYsXXnjBPe/UU0/NzJgxI3t20T2S0kO6I4XOxdcTNVBXjmH8kqCcCbfOOutkog4mezbj4h8JPfe88HzaNOaqD4SPBEUmEvrlytI/lzRFAiN7Nl0dTII0cM9IaJWL95gxYzKRwMscfPDBmUhIu3NDhgzJRApVJhKU5cL6PD3uuOPKyias16TVE3UmmX79+mUOOuigTKS8Z8+WEikyroyRFRWRNo+SSJNvafLI1wPKnHhWBm0+UkQWCx8pT5mzzjrLPde3XfKLcFGHWy4e5MExxxyTiZSszMiRI7NnM5krrrjCxSUuF6Ci39K0nTTtMm3Zp23zvpzi4Ukf6eRekbGYPZtMpNy5OvvOO+9kz5QSGQuuDVPnKpIb4PM2Xl9o/5GB6g4v69PW5crql+/DCBeS1B69DCJf4nI8CS/3uX9czvm0xfvINHUpzoQJJZkdd5wWWYSTIjk8Nbqm4viFDBo0313HcdFFs6O0ZTKTJpVk7rhjTtl5PvMML34J53/L53jwwdJ4Ey/ix7ljj51Zdj8gTDx8GLf27SdH5Tw7kkOU92z3nfO9ek3NDB+ef3o9vvypg0nQLk466SQnx5HnECnWrj5RfpHx4M4B9YF6Qf245JJLysrby6x4eMoR2RR/fhr5mUSafidtWpL64aRzaeR00vWQq20WUq4nkUamUQ6RkebyL9RXiNdtt93mnuv10VCmVKa7RgZeZpdddslEBllm6tSp7hz4OhAZhO4zpAmbtryTqKoMzkfX4/fzzz/f1d/ISHXngPNXXXVVpX1SWplb1/rY6uqn9YXrrruuzLa98cYbXR7nHl6vo+Ta1RlXlXBH56jhZz8tAkufkS82Eck1AhfCKBUjl4wEMXrk4T7MKDKSxEgXIy2VwchrGD8+//3vf3dulpHQy55dnKjS2vPPP29rrbWWc1kIdyJceuml3WtmmAElnsBnRnAY4WQ2I5xBIe922mkn5/JBuLT3ToKRN0ZKGaViJDIcIeIejDqF+Gcym8Bonx+JAWZHcDPx8auMqPG70T8P8Y8EjfscCQf3txBp9BCed9nixhiWJfdkFJEROsrTU506SL4Sb0bLKccw3ow84xbF6KFPZyR0E91+yNOoodsRRxxRNiqaC37HNZORVOp2CPHguZFAzZ5JJm0eJZFvvqXNIw/ygnhWBiOFjJofcsgh5cKz8y7n+IsXB5BfUYdhUadTLh7kAS5ElE083YyOR4phuXzyJP2Wpu34dsmsBWVfWbtMU/Zp27wvJ8LFw5M+dgznXsxsEjYX5DkzI4xEh/BM8ovZnEipyp7NDaPn5F9YTtRN5DEj/8ymAPetSl3Ot37lA/WaWY6KZsKBmTzcK5PkHHGNDIPst1IKKYfjhK8V8gfuxrgdJ0HSllyygbVsuSgOfO7QoYH9mfsPnXRSM7dmefvtG7u/55xTuuP2xx8vdLPFhWT8+PHOxZAyYRYImQG4tjODeNhhh7kZVQ/1gRk0ygnvAmangLqLzIqHZ9YKPYa676mq/AxJ0++kTUu+pJHTaSm0XI+TRqahj0aGjpvFXGWVVdw5IF5+pt7ro2l0V8owSe7iSXfxxRc7N1zvPZcmbCHKO40MTqvrRUZeovzmun/+85/Oxdi3wyTSyty61scWwkaqr9Q7wze+I6I/cBeh4nhwI1h++eWdEoLrCq4EY8eOdZWAzj4UHLlA6NOwaNA0gPDA4Pruu++M3RcRBpURNgoPFRihxBoDBFwSuFKgjOE+gptDHNKx0korOcHPPXDT4BrOxV9NRRxw3X3hhRec627aeydBQ/3++++dkEb4VYZ/JsY5riFhniIcBwwY4ARYZQYWIEBDhQ3CBg+FSGMIwhJhjnuLXxPCceaZZ2ZDLKI6ddDnK3WMuhbmEwd1cuLEiWX5RBoQtLg80Znee++9Lk2wzjrruHZTkctOCC5i4RoUDu5dkQIUkiaPksg339LmkSepLSZBvcHVk3vFYQkBHfvhhx+ePVPa+dCphmvBOHBNSoL6y5FE0m9p2o7PG9ygaJv5kk/Z+3vn2+Z9eOKIzItDHEkPYeJlFQfFJVwPx4HrKG6F1QUZAcTDU5W6nG/9ygdcKeMyLglfN2g3LOGpjELK4drmllta2ltvtVnsOProwm4P3agRbnGln/mLq7Tn55+rbvgiD8L85ujVq5eT2dRrdnn2Ow7j+kyfhnIdh/Mop6xXx+UTCA+hoZEL3y7Tys+QNP1O2rTkS1o5nZZCyvUk8pVpI0aMcIZq6ErsQZ9Dr0O/Q/6k0V2R4zwbd3gmE2655RbnpstvpHeTTTYpMzrThC1Ueecrg9PqetRLDHDijqFL3pEu8oV+jUkU3JBzkVbmeupKH1sIG6m+Urgeuo5BBcCnn3UHjz32mFvXARQmHQuVLl/hxGxtUmUCGgbKSVWg8ef73mTWpCQZLpxLOk/Hma8ClvbeSWDA+846H5h9ZqYhCbYb98KzUBQijXQ6t912m3v1CAL+kksucfcFRucYmQwpRB1EULMWKVdZ8j5CoC7RyRGvBx54wH0GBDIjgHT8fiS2Ihg9RxByPzoDL/zonBDWlZE2j5JIm2/55lFVYLaA+lgZdKQXXXSRW8PHehnWTZKHMHDgQLdmplCkaTt0eKQhH9KWfdo2z4hzktcB8jOfuomSwUwKg3t4+KD4UA8Yee/Xr182VNVhdhwlww9kFqIu1zYoSrUhh2lqbdqUKt+zZmVsLl7DWfr2bW7HH98sKqdMVF4zI6Ohco+oNLAWd7PNFm/r771X2OfECbv5CmzBSknKcww3b3zG+wTqXC49wcuYEGZ1c63NTqI68jNtv5M2LfmSr5xOS03L9arItDT5lI/uSrmzhhbDiHXL1113nXs2ehEzmUwq+despQkL1S3vqshgfs9X16PNPfroo24NLOt0GfADDD/2ySGdlQ08ppG5damPLaSNVN8ozlRlQek744wznHsII0a43lCQTPsz5Z8vLPpOmmXm6NOnT2Ijywc2M0Hg5QOjUN5FIYRRvaQRWUbJMplFykhFpL13CPmJMGS0KsltJBd0tgjQpDyNz94Xguqk0fPrr786IYmrERs8MFrIqCAHrl1JVLcOkg/HHntsYj5xoCh5EGz77befE2C4diHMtt12W7fRA5tYJaU/hE4eIYiyS0fAJgw+fShr+RhQVcmjJNLkW5o8SgvtiKMyyG+UIzaVwWUR10Gf7tC9sBDk03Z8u2R0OckVMU6asq9qm2dmwxuVIZyrzKUWWcYmJygI7FJNWumYiR+zZWlG3HOBPGYU3ysOharLtQGKDQfxrw053Lp1g+jaUktw2LAS+/TTRUYnei5uyrgvR9WkaAi96PMcr04kyWuN2UQ2QqJtxaHvytVH4SIdQt1l1iZfvQKqKz/T9Dtp0pKGfOV0WmpSrldVpmEM5ku+uit/cT3GDZeZZmb/2PSJmXzcvCk3T5qw1S3vqshgnpmvrodRi6GHKzU7Rw8aNMht1oYXJoYps9m5SCtz62IfWygbqb5RlIYvlRv/fEbkaAAUJKNQuGhcddVVbmQf15XKKhSjLVQ+1uBWZjRURpKwonEh9PDvzzWqwwwE7ghDhw5N7MyGDx/uXGBQYLgHLh5cw/osFLkQBC2uKay74dlp750EnR4dNjvEjRkzJns2Nz5+NLK0a3mqQiHS6MHtC0HEznjxwY54p1vdOkhcqH+MKHvXsYpgR0B29PN5yigr8WS0mtF8djUMXWmSQDCTF7gmkW8h+SoWafIoiTT5ljaP0kK9IT9ZLxWHkXqUVUa9wbuhodSGI8S0uUIpZGnajm+XhKNjrow0ZZ+2zVMXcYPD/Q55F4dz/EaYXLMDGO/UK55LuYTE41dVkBFAPKC6dbk2YYaPfoRXvYWKZy6qK4cjfS9SfptESlzp9zvvnGvffVe38gTQJVu1Km2PEyeWRDJl0WBwZcvBFy7MRO239DN/Bw9e1IevsELVvLzSQntEYfcuzCGcp5/HAPCzetRd5FDorp+LQsjPNP1O2rTkSxo5nZaalOtpZZrXRykvnh+CUYMrMAYcfWMa3RVX8P79+zsZDNQL6hGzfsziYhT5+pEmbCHKO40MTqvrMUCEPvzSSy+5/CS/iMuBBx7odoZmqcdbb72VvXpx0srcutTHFspGqq8UpeFLA0EIMnJDQw3xBiijLozWhLDYO9wchZEg3CsY0eIIhQ2uKFQQLwAqg1FDrvHQMNk0AfccNhvIBaN++OIjPBhhCw1oFFpGCnFvYAQO+EyHQ7p5V2cYZ/KEeOC6xJqQtPdOgrxmO3kEL24nCDQPjY2t0UN8/OiQGMELw9PwCc+riuJCrapUJ43x+kBZ4aLCqGAYb5SMuMtNVeugB8HMug8ED4pF+F41yhQ3H9ZT+U6Nv4ywU6fCNHqBSry5ZwjXhJ0igp84sekFAyMe6i3uRvkoyGnyKIk0+ZY2j9LCqDudJBs9hMYj92NDEhQQ1hyBN9h4zYCPp48DI9aFIE3bSdsu05R92nsTnk31ksLzmXP8RhjCJuHLGyUDhclDWeCelqRc5QKFh3SGMgY5zowF5c26QKhuXc5FeK9CQdnhnpck54gvrycKKYQc7t2bTWBK3UvZ8GnvvadHBs+cqN9ZYLffPtcOO2xmZBgtikdldO++SB0ZMGCu8SqhXBti5Uu7dg0iZbbUYHnllfmRjJzjXK/7958TyZiK93S48ca5xkZdhL/yyjl26aWlnhO9ejWKlNdFchuFFiU57GsLBQo/ijIzRKGuwbPQSVBO2QwLd0ug7lKH4+FpJ8zIeiMUCiE/+S3ffidtWvIljZyujHjbrEm5nlameX2UNIUDFeQzuh5tfM0113RyK43uitGDEcSa5Xj6uTdGEIYYpAlbiPJOI4PT6nrcm4EZXLXjAz9e70taK+xJK3PrUh/LUR39tL5T7xyRKCRGnJKgIrOmAdcd3AhwXeAvhc0IGKNH7HxGoe+1116uIgLCh99pjOxKR+fBRg2sFeCdWOwGh7//c8895wQLoyW4AeA+wFodfOUrg8ZBPDiA66no+PuzCURFsIMbuzXybjBG1NgUAKOMe5AmKi+CDhiVxO0IYX/OOee4MMTZp5284d1d5AGkuXcuMNzZeRSBzGgbwozGzPPiu6wSP0ZgEfZ0VJQR4SkLZh+INyOXCFGEbCFIm8Zc9YEwlB+DB9wPwYRAJw3UPWBEj46Y/E1TB5NAKaHuEW92beQedMR0knS8pIs6yL1wLcMdKEwjAoxdE9lA4vTTTy9zW8SlBjctFCHqNWG5lnPEiXVGbFzBPYENH+iMiSt1FkXPKzNx0uRREmnzLU0epYXRXOoi79Ijf7g3nXpYT33bJQ/pBOlceS4zBMgqOkDiisJJGhjFrSpp206adonSkKbs09wbkJPEDVcqHx58/eS3XGtNwZc7o/O4YeESSH1mxgl5TPyoVxXVzRBG9NkMhvdkYrTwmVF7XDMpd6huXY7DDAFtjDqC4oMyTp0tBNQN+j5meEIZwIwBCiZlEq77LIQcJpsvvLBFFIYNp+Y6I/X88xc3Jtu3b2Cs+11hhYrlOe/mZe3u++8viJRGjhn24IOtonJI33Y9bduSL02d0RtVT2fIckDXrhWv3Zs0KbNYeiJd0E4+ubn16FGaFuo8fTgukRgrrAksJNRxZDSuqfvvv79z9aQcvXzDuGEtLeUG1F36d/p+Hx4FlnZGXfXhPNWVn2n6nbRpyZc0cjoXudpmTcr1tDINfZS0sDNwUv5RVl5up9FdyRu8AClDL5t5NvlHWpETDCxAmrCFKO+0MjiNrkcesZyDpR5h/Lgf4dl5mcHBXKSVuXWpj6XuVVc/rddEmVQU7/Hl2Dz2nsJRo0a593FFwsr9zl/e6xZV7GyIRUQVw72zjnC8K21y8O668ePHu/e45XOfOI8++mimd+/e7r1sjz32mHu3F/fgL9/nxt6TSHh+J70hhHvmmWcyUUNwv0eNNxMJNXffSDnJhlrEzJkzM3fddVdm5513LoszaSAtcdLeO4mosWbuvvtul1buwb0iw8rdN9/0kO/vvvtupe+spIwp66T3Ifp6Qj6GpE1jrvpAvvLuNl8XfDo/+OAD95406lt4vzR1MAnygjzh3W/cnyMS7i4t8bpDGVCn8slT6kGksLn7RUIv88svv7jzhHvrrbfK7uHrDfnBe/p49oQJE1zYXKTNoyTS5Fu+eZSrbVUE8aSORJ2je2809+Y5ufL0oosuKhdu0KBBmajTds8N6yR1Ny6vPBX9BmnaDnUiUtrKlQXXPvDAA4vlRdqyT9vmk8opV14mQVnwvkXSyrXkc6REZCLjLcO7CyvKM4/PW+oi75AM84W0x+ORpi5XVr8IFykVLr+4NlKW3PlIMcv5Ht/4+0YrY1JkrfXr12+xOkgfn5Q/1ZHDHl5f+dZb8zP//OeMzHLLTclEJqN75+1uu03P3H77nMzYseXbevw9viGffLLAXcdv3OvJJ+e58+F7fLk+iTCMfy8vEL+HH57r3r/Lb6usMiXSc+ZkXnttUTyS3uMbGcuZ666b48LzfbvtprlrwmyhbKiDkQGZGT58ePZsMr5vSuq3KgO5x3MoV+6B/kD/Tv2MQz2jzCNl3oX17Zi2Eynyiz0/qV3m6mOSSNPvQL5pSWoDudpFvnK6ouuT2iYUSq4nwXPTyjTyj74wLtOTyipf3TUyhDOvv/56XnUgTVhIU3eTSKtPJMm0inS9yJAuF7/IuM6pKyeRRuZSF+tSH1td/bQ+kPQe3wb8gwEcVYgydwgW8gshhCg+kPO4dzFDFRnAebsBFgNXXnmlG9HGhbM6s+9CCCGEqNuwEzuz84D3Ep8L40sqhBCiTjFz5kz3GgjWBIUMGzbMuXLhltmjR4/sWSGEEEKI4kaGrxBCFCGsNWLtKjs1st6LvRH4y3qmiRMnuvVNrHMSIg3sfdKgweS//HHffVXbNE8IIcSfhwxfIYQoQnh1Ae8NxMD9/PPP3QYcuPnuvPPO7l2bFW0mJYQQQghRbGiNrxBCCCGEEEKIokFrfIUQQgghhBBC/OWQ4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuEEEIIIYQQoqiR4SuqxJVXXmlbbLGF/fTTT9kzor4wcOBA69mzp3388cfZM/UT4k86SM+fzaRJk2zvvfe2f//73zZ79uzs2cKh9lZ46nI7mDdvXrXqUSaTsRkzZlhJSUn2TM1CvaR+Uk/FXxvq7hdffOEOPsPo0aPtwQcfdHVSVJ+ZM2faggULst+KA2TWY489ZiuvvLJde+21ZXUnH5JkeW3LwMqgvJ588knbbrvtXFzPPPNMmzt3bvZXUZvI8BUiJbNmzXICet9993VCGiG2yy672B133OEMICGEqCrTp0+3Y445xsmX33//PXs2Ha+++qqtv/76dtddd2XPCFHzYKxceOGFru5yHHDAAa6vPPfcc+311193xkhdwg9Y0ofHj7/97W929dVXV7kN1hRff/21bbbZZnbRRRelMg7rOh988IFdeumltvvuu9uxxx5rTZs2zf5SNeqSDKTeP/XUU87YnTNnjh1++OEubo0aNcqGELWJDF8hUjBu3Dg77rjj7L///a9NmTLFDj30UCekGzdubFdccYUddNBBNmzYsGxoIYRIR8OGDW2JJZawJZdc0po3b549m46WLVtas2bNbKmllsqeEXGYUcdDA8NHA5aF4Y8//rCvvvrKnnjiCRsyZIgzHjFmxo8fb3379rU2bdpkQ9YtOnXqZEcffbSddNJJ7qBPpw3deuutts8++9j777+fDfnn06pVK+vQoYNr28iKYmDq1KluYGTnnXd2Bj1prC51SQYymPncc8+5iRIGgs477zzbf//9nd5Y09Qlz7i6ggxfIfIERYkR4I8++siNar/yyitOgGEEP/PMM3b77bfb2LFj7eKLL3ZGsRBCpAWl74YbbrCHHnqoykobbsfffPONU9qFqC2WX355e/7552299dZzhscJJ5xgX375pT377LO2zjrrZEPVPbp27eqMXYxzDt+nY/jS7+OmjXtxXWCFFVawN954w04++eRaMZxqg7Zt29r5559vl19+eUGMXqhLMnDixIk2cuRIW3HFFd2ApvhzkeErRJ588sknrjPcb7/9nAtX2Ok0aNDAttlmG/cbhvEPP/yQ/UUIIYQQ9Qn6dIynrbfe2s1es05ZiKqwcOFCd4i6Qb0yfKdNm+bWUbKekqn7VVdd1Y4//ngbOnRoNkQp4UYz7777ru2xxx7OxeDTTz8t++2ss85yI5Fcz334/R//+IcbIWIx/Ntvv+2u4znrrruuC48bT5x848SmH/ye64hvCjJhwgS3wH/jjTd2v7Mg/u6773brS0P8pjfE+5577ilbOM9fFtLHN0DgO6OFpJW4EpZ04oaRtFnC5MmT7ZprrnF5QB4dfPDBzu0n11qdeLz5y3fOxyGf33vvPXdP7u3vz7nKNiQIy5iNVSgfH0dfjnHSpj3OZ5995sKRt0nrTzCE99xzT/vXv/5l8+fPz54tJW1a04anXjA67fPdp+u1115z3/PZwCf+TF+X6fTzXZvFqKYvC56bq95695ukzXD8RhVVdc3JNx1+Qx5c1F944QUXV5ScX375JRsiGdo2swPc28uGXOvAarq9scaL2RQvqyoqs7R1Kon483z6Kfd88Nf79fE+vnF56esAeccsAM857bTTyjYDSRMP1lQhG1kXV1n+Q7739nGkfLifD08bvO2229xz43CPs88+u5xcJ25h2IpccFG+8TrZcssty8Ut3jflcm9LU1+SIAzpZeDPpxX5nstVuC7KFJ+/a6yxhssLXHM33HDDcvnt+9UPP/zQzQASb9IJxHvw4MF24oknlotTvBw9afrEJNJcn6Zd5JuvkCZs2vTGw+e6dyHkV1ro55Nm6PKp18gE0sPfOOgSa621lt10003l2kE8nylHypNy9eTaSC5tX5NEIfLY18HKZDykfV4aWZ5LBlKv0uhK1ZGZXtbssMMOTkZzH+QO5RduVJnmGfnq/D79LL8D1hfzPZ4f+dS5YqNRZLBdwIcwc3v16pX9VHdgbSUVgQXixI+KROWgE37kkUeca023bt1cWCobBces28svv+xcVHDBYUauXbt27jeue/rpp22llVZy5xndY0YP45gKetVVV9kGG2zg1qigbLFQnrWbhPXrrtLEiThQqQjnDxoe98Y9dvvtt3fPA55z1FFHucX+rHlAGacS/t///Z+L2+abb15mePEs4v3mm2/at99+68KuvvrqrjN/6aWXrHv37rbaaqu5sDSIG2+80W000blzZ1fBiQPpoKNhoT1x8OtGOE/nTj6tvfbazqhj1Ip7MGiAK9Nee+1l7du3d+F9vIkL+UR+sFbt0UcfdQJ5k002cWtTgPrGpgM0RjoW3FGINx3Cww8/7FyP+E65JBGW8eOPP+7ivtNOO7k4kW8oLOSTj1vatMfheQgVdgn85z//WXbfOKwV2nTTTW3ZZZfNnil99s033+ye3aJFC5dWNjb4/vvv7b777nNhwmenDU/d+t///ufSQXn36dPH5Smu19R/jHDO+bpIPWE9TXgufCb5Q1lTb6hflB/lH6YpCcIeeeSRrkx22223cvWW+kJn5d2YqMesA0PhpJxCfPxoe3QSufD3CMOlSQdKLjP41FvaSpMmTWy55ZZzdTeXO5JP488//+zaJh0QndMtt9zi2jHyhPhwr7R1Lm17o9xxucdQJt4o7j6t999/vy2zzDJOHtGGqtvewD/v+uuvd89hExLq24svvujkI/Xet+8kaEPIVQYauI70rbLKKvbOO++4zhhF0JeNrwMoSwxCch4Xv969e7s6lW88iDPLERicJD/CtsGgaLxtpEmjjyObzfAb9Zv6zMAJyyCoA2H5fv75524dIdd5uc7zcKMcNWpUmVyn3iAvWfdGHiED4Mcff3R1D4WG/oLrfdx4XijvktoG+XbZZZc5xQkFk3vTL/IsZFtlbZw6hKzFxZIBXxRblCXkPfWNdWzIKd+e66pMWXPNNV1585k2RzmxDnXHHXd08aFd8lz61UGDBrm+FBdF2tJGG23k8gAXXqB++DpJnChHZAL3hN9++831iTx71113LdM1yG92PWYAoyLXzjTXp6m75B37U1BXkUvkF4Nu6C0Mrvi6mDZsGh0A8tV1CiG/kvB6BIRtzYOLKu0TOUF9xyU333qNvMaA49qtttrK1SsgLZQfcccoIu7g6zqDqOQzdZv6iXFDuqh7yBLfb9GGyR9I29ckEaYrH70jiTQyPu3z0sryJBnIPdLoSr5N5dPHJsF57oWcpL3i7n/BBRe4tozsZQ1y2md42VSZzk+eIov4jXpI30N9Q/5Rj/298qlz9Rn6S593/HWfb7rppgzHDTfckIkEpjvqIpHwyURKXyZqQNkzpXzzzTeZyIjMRAIxExmR7lwkrFzYSAnNRMI5EymP7jz437gmypDs2YwLQz5ElTETCZNMVKGyv2QyUWPIRBXT3e+jjz7Knk0XpySiTjUTNYTMqaeemokMKndu1qxZmVNOOcXd99dff3XnoKSkxKWF+EWCOns2k4kEjDsX3gN8HCJFOhM1LHcuUiIzBxxwQCZqYC5NnkiJyRxzzDGZSEBkRo4c6c7xvEiZd/e+9tpry4X/7rvvMttuu63Lp6jjcud8vEkP6fJwH/KH88cdd1xZHKMOM7PLLrtkokaXiRQ8dw7GjBmTiQRPJmqE7nMuwjKOBKp7DvA3Epou3lGH6M5BmrQn4dNHvMaPH589mx/kRyRsFisjnt23b1+Xhki5z55NH/6FF15w5+LhI8Hv0kxehPU26pwXO+efSd6F+ePL4+CDD85Eyk727OJEwtI9i3oRKanZs6XtinpL/C655JKye/Ns4kD9jePjx9+K8PcIw6VJB3WXOkzdpI76OpQLrosUwMXqeJhG6gh1BWqyvYHPp3h4X+60f66F6rY34HnkbaQglMurSFFxcfvPf/5Tobzz9ZT8CMMNHz48s9NOO5UrG5820kF6QtLEoyptI997h3EcO3asOwfcO1JsyskKX3coR18mQLnddttt7j5erod9AHIOImMg069fv8xBBx3k6lVIZFC4NNJ/eZLaxpAhQzKRUZiJlMdy9cWnLZTPSdAfRYbMYm2c/KBMeV7Ynuu6TEnKZ4/vV5G3tFcPZRsZaq5vD88TB3QE0hspuNmzmUykQLr7RIZn9kz5vnzAgAHZs8mkuT7fuktczz//fFcXIqM2G6o0DZHR4vLV9y9pwqbVAcL8r0zXKYT8SsLrEfE6QB2LDP3MySef7OKBbPZ5mm+95jfqJ/U3TB+/ESbMC1/XORcZtu4c+HzmeYMHD3bnfL8V1vPq6jfg05Wv3pFEGhmf9nlpZXmSDKyqrpRPH1sRvsxC/cCT9hleNsXTkKTzQ1I+QJo6V5+57rrrymzbG2+80fWT9caUP+SQQ9zMKiOcIUsvvbQbMWOkKRJW2bOlMILLqGfSiAUjj4w+eQjjR49xGWA0zYMLK6PIjFCNGDEie7ZqcfIwY8TIE6NzjKD7UVtGthkZjQS5G/3xMErBqCEjOYwUEhcPz8L1Mhw57tGjhxt5ZCTHu19xbdShWNRgyq1PZadFZqMJFwkrd47Re2ZNyaNIWJULz+gTI7khPt6scY0aX/ZsabwZEed8JOjcCC9EjcuN7MZhxpTNoXC9YMS0MihjRs/CER3KilHlsKzSpD0JypL7UU/C+sR5RtG4R3h4NyTSyGYflA0jbmEZ8WxeW8IsDaONhK1KeEbzGME74ogjyoWnHjKKWhn+mYzEkp9h/lA/ceNntikSxNmzi4OnBKOQhx12WLm2Q14xU0ObijqxGl0nVdV0RMLfjYz6OpSLSHlxo7bxOk4amYGLz07XZHuLlAOX1qTwlDvu9swM0Cahuu3NP49ZI+RkmFfMxFK+zHzzzCRoJ8w6MRrNzImfIYKoU3Y7XCaVDTuskh5Pmnj4tkF7ibelpLZR1TQSR/LRw2dkEOH8JnfIx0gBcaP5zIB4KDc/YxSX6yHMIOLqzUwYYUOo79TvSAnKnkmGOuD7ghDSFikDTn74mcokmF1AfsfbOGVJf8XMhKcYZAr9aqQUuvbqod7ceeedzv0+PE/6aKfM3oT9SKTkZj8tgnrF7CablzErVxH5Xp+m7kYGXGJfRxrwZmK2B/kEacKm1QHS6DqF0hdy4d3dff/NDH9ksDpPCtodr58hXmnqNb9Fxo6btccrxEM/wnfyxMskX9fJI2YgPdyDe+JlGN4jTnX1G5+ufPWOJNLI+LTP40gjy5Pw98hXV0rbx1aFqj4jX52/Iqpb5+oz9WoOmwoQrqflQFghtJLAvSPsACqjdevWzhUaF4GwAkLonhOSNk5A54hr5MSJE53xGwp9BAO/I8z9/fyBoPzuu++cMkVH4IkbY0DakzoCGj+NKFwHwoGrRwjuDrg2IbBwVaoMH2/cu+N5x3c6ZH4nHNBwyTPcNTA8yA9cqkgX8WJgIlQscpFUxriPJL0GJN+0J8H9cFnxQtjD8+nc/GsQEPghKKPff/+9U9wQZHG4J8KKMIStanjOk6dVwd+DukUd8/niD+oidZVwuWC9CUIYAR6H8yhoDPbgClhTVDUdtJ185ISv4yuttNJidTwXNdXefHjcmMJOy4NxRTxRMlFIqtve/PNwuaT+hflKXR0wYIDrxHPVET9whBKeJEtRLOmI44MH8XxOEw9fH0h7aJjmoqppTJJ3cZdJ0o4Cn+SO2aVLF7fGHBe4yuoVSk24xpeDQZGKDEgP9YGwuG4yaHvvvfeWGSHrrLOOc2sLldU4tHEgLyrD5319lilJ/SpgDKIwhmt8OXDDjcOAGG2avoEBUtyOWRJBW6NPDAdBksj3+jR1lzJm4IC2j/FKvUMuoFfQTujPMF4gTdi0OoAPn4+uUyh9IRekBYPK9+MsmwKMXtzavT6Rtl6vueaarv5i8Hu9AfdS9Ezyw+PbFm7f8XsyecM9KeOKqI5+49NFXclH70gijYxP+zwfPl9ZnoS/R766km9T+fSx2AHxCRAG08N1vEmkeQb560mj8+eiEHWuvpKf9lYHoHIw+jp58mS3WJtOhoJnpqRfv37ZULVLVeKE8oOhzEwxo8YIzyQwoJIEAtDRUMnTQsPhHWmse2CWlLVAfsRo4MCBbjYrDgIirsRVRK4BAv8cDx0hHQrCjfUW1113ncszOlpGvlg7Xcj3r1Ul7SEoq+Q5HTEKBMoqEN8+ffq4z8B6Ap4Rh5HKpNmUXMIqbXiEeZKxnwYELOthcinfHTt2zH5Khs6cwaMk4uVfk1Q3HZURDlRVRG20N/I8yVjhXHi+UO2NNVjMKCXBgFNlyid1tyqyK06aeFBeafK0ummsiOq0A2QLSjn3wAjxRg+GH8ZwZXAdijlGwgMPPOA+AwMtKP0oppUpTszq5pLxSRSLTPHQf7NxGa+bIh8vueQSF0dgNonZ4BAUe9bpsQ6V93f69aSUHa/MYQAjybj2pL0+37qL3vHoo4+6tZTsKYBhDBhj//nPf5yc8AM0acJCvjqAJx9dp6b1BeQuM2j0u8B7hzFmWEuL7I7HL996zf1Y64y+h2cCxg3tmDxl1jiE/MEjJMkAAoyjXFRXv/Gk1TuSSCPj0z4vrSxPIq2ulE8fS3wZBAp1gzR9Rb79eKGpTp2rz9QLwzeT3VSDDh7hy4J3D8IpnxmSQlPVOLGJA+5BVDZGUnPNNCEsGeEsJLgtIBhxraYTCZ/NKGQoHGnIHIz44CJDI84HRomSSDpPg8YFiwPBzWgh8aOjZ6QM5cIrFdUlTdqTQNjS2aF44KKFQpIG6gSj03GhzTl+i5NveJQeOl/v3pJv55QErlp0/lVtTwwK5BoNRpGoLaqbjlz4DjrfGabaaG/kOSP98c6RmZR4WRSivdEGSEtVO2OWfiA7q0s+8SBvaBvMhpKn+SpM1U1jReRyZa4MygvjB0WK/iNUsCi7fBU55AOubRzUD2QZLrPXXHONq9e8SzNXuqn/zDYy0JsvxSJTPLipYgTSfzNwEOZVLrmAon3GGWe42UNkN7PFbOqDOycGXGX9fJrr8627yCKMZwbfSQdyhw2CuC+DKngDMHObNiyk0QEgX12HNNWWvoDhiusrkxkYv0xwhPI733rNNRhEzLrSH2DII/dpf0llxDPxIkhLdfUbT756R0WkkfH5Ps/rOWlleUhVdaV8+lh+CydA0pKmHy80Va1z9ZncQ411CCopFZ51F7j9hOB2xFHbVCVOuPawexuuHozWJo2AotDQON95550y15hC4d3tcGkLBSNCKh5fRm0ZkaRDoVFWBoYg6UHAxpU7vnOe373BiMtL//79y9yeEGS455AvjOCyHs674RWCNGnPBaO0dHSsyWBWPx+YrSBdpDNJMeIcvxGGsGnDI7ypg4wmV9Uthbyn3lFGVc1z3LkQ0t59JoTzrC/DBcqPnPtOqJAUIh0VQTunDg8dOjQvA6Ym2xvuXsSHuCQZIsOHD3fPRxEmX6rb3hgR5pko3VVZU+nLhvgmGSwMKDHzU9ESEUgTD982WGM3ZsyY7NncVDeNFeHlOnUzrhSi4OE2i2GRax2eV/RpZ3FXv1z9TRzWK2Ks+LQhPxiwZaaI2SsU5Ipcpqkv1GfcBSujWGRKHIw38oh8iyup8TJAYWWQYuDAga4vJ37MSmLksfMtaw1xTc1V5mmuT1N3GbzAVZgdYKmL5Bl5eOCBBzr9BI+mt956K3XYtDqAbxP56Dq1rS8ABjnu9Mxy88oXqEq9RgYxO467M/s4IOfpE0KQ5dRpwsTlQz5UV79Jq3ckkUbGV1XPyVeWJ5FWV6I9peljq0JtPCMX1a1z9Zl6YfgyE0KhoxTS2XkQlrjdJHWMNU3aOGEoMapOx4m7Wlx58TCyigsVa3k4wgrJPejwvPBPixdYjNb6jon746LFKHYIMwsIa9LBhiphR4biwyYGITRMRjZxx8KVJ4TvnOd3wgGdOJ06I6E0vhAENZ16viNy+ZAm7blAEWOEm/e9sRlMfOMRyh7lEfzsIMoR64VQbnGDC9PKZ87xG2EIW5XwrCVJCk/nwWsTKoN6zNo/FCgU4/B9iz6PGNWvSDlBESR/8H4I6yfXU49R0Nh236+twVWcTpKR6tDQQ/h7RSothUhHRbB5BDMb8TpOfaVOsPlPSE22NwZgWP+XFJ5OnfcU4kbG7AhUt71xL4wjlnfwWpHwHlxPPWNmJJeS5csGxYWwYXypp6SBWUvqREWkiUfatlHdNFaEl+u4PIYKM/ei7lC+rAfMNdOPgUAdYbMfr4QDfQKut/kY6tR7+iA2wQnznzhwYJRQTrlgwBb5HW/j3BelNjSa65NMIQ75ygTKh3xCWQzrB22UOhZC/WNGnVdIYbiF+PynTNElkkhzfZq6Sxrop3AVjhtvfjNO7/qYJmxaHSCNrpNWfjFQhFEe3jMtftaXdD/33HPuXlWp1/QDyCHSyG8MoMfdstkMjHxh/X28rJFVV1xxhRuEyEV19RvqWhq9I4k0Mj7t8ziqq+ekvUfaPrYq1MYzPGF6obp1rj5TL1ydqbB77bWXG3XEvYd1LbiaMILNqChKAaMiCLuKOu5CkiZOGEEIPBQcNhGhE2BkNgTBza53uOrwXjN2c2P9DAIXhYh7ssMdLiCs4UnraguMXqJc05gQiIwOUuHptIkviguGOaNijBoyOsdoLO+HY0SV6xFqKBx0euFaHpQBdhRk5oL3giGwEHDkAfHmM78TDsgHdtbj3ox2ocAQB/KFOLFBg+8gC0GatOeCcJQ1wgDBiRLLzo4IJtxnuC8dIQoI8fdQXuQLxrJPK7AjKZuz8Fu4LitteHbAZDMC4uPDozgwM+2Vk8qgw6LOUR7sMkv50ZkyqEO6UACoe9T7JKj71Ne+ffu6eowbP/XDX4+CwzpC8hBQVumEUGpZH0+cqeO0H9pMValuOiqCtol7G65kvo7jukadpeOKU5PtDUgLxnYYnsEY2hv3RWFGuYTqtjfiRxgG+nCNHTx4cLl7MANCvUQJRUlNgnrKpkpJ8aVMmNGpzFUxbTzStI1CpDEXpItrcVdNah+UJbvI5gIFif4G10s29/Fh2ewHJZs4YvxW1AdyDS66Yf6jaHm5cvrpp7t2nAtmqtgI5ZxzzilLA4YX1yPniUNIXZcpKOGsWcZo5znkB/kT7k4dh/ZEOnArpe3xPIwz6ooftCAvUCpJF3KC38I+kbZJ/fM6RDzfPGmvz7fuUhdZE4scC/PV99XsEEwfBmnCptUBuHe+uk4a+UX7ZnKBza+YhGBdcFWhDjNggPFIepDVVanX9AOkF90B4z9e5uQNG52Sz6QlXta0BfqHXO2Tultd/Sat3pFEGhn/Z+g5ae+Rpo+tKjX9DNorMo26waApbs3U2+rWuXpNVOnqxXt8I4Ht3iPHuxAjAefeCRcpAZlI0GciYerekTUs+47LqODce9mS3plV0W9cz33C96N5ogq52Luw8o0Tz+F5XJ/rIE7EzTN+/Hj3LrRIgLnf+XvWWWdlok4kG6IU4hqm3eOfmXTfiy66yMWVOPNeyEGDBmWijtw9J0wfTJo0yb0/Mh6eOpP03KR4853zcSJBmHn99dfdPbk3RyQEMlFHU+4dcElUpRzTpj0XkYB010VGkLuXTyfvQ4sU0cS4c827775bLq185hy/xUkbPlLsMpFgK8v3SLi7fHzggQfcd+qvh3TGz0HSM/MtDw/1kzbg8yVStjN33XVXZmbwXjkP52655ZayOBP27rvvzrz11lvue2XlkdQmId90VNTeKyLqpNy7Ebkv6SS9nEMOxOtjTbc30kO6KG/uF3WQmUgpc+/0Qz6FVKe9eZKeR7pz1cs4/nqeG+ZfXK7lqqOeNPGgbVCvevfu7cJyTWTsuOuTnpHvvSuKYy65TDqR4/F2GuZ/LtnNs2kbPl5etkZKont/KHk6YcIEFzZX2yAvHnvssbzyLQnqFPUxMoLKxYF+MDKAFmtL3DOftlgRNSlTaJ+RkeTiFSl+7t2tkKv8IP4MX58++OADdx/egRm2vVGjRrlzPjx/k/ryXKS5Pk27iAz2cvkaGfE5++o0YTnHb2F8c4WFpPBJ6ctXftF+iGtkoLr3x1aE1yPibS2EsiVe4Tuj09ZrzkXGfbk2mkS8riflc03qN0np4nO+8gF8HSStXE98SFNSfU37vDSyvCIZmK+uBEltKlcfmwtfZkn6KqR5Ri7ZlKvf4PrIkHV5Rv5GBnb2l1LyqXP1mesS3uPbgH8wgKPMKXMLYYG8EKJ+Q3vGRY1R70io/+U2MBBCCCGKCWb1Dz/8cDf7zhpnUX2kKxUv7HCONxfgqcXnerHGVwiRm5kzZ7odLVmXETJs2DDnLoOrF+tThRBCCFF/mTt3rttcVaRHupIAzfgKUc9hHc8pp5zi1g/5NVgTs+s0gE04KlubI4QQQoi6CRudsQHTu+++6z4zk7X77rtnfxX5IF3pr0fSjK8MXyGKAHYq5L3SvHoB4c5mODvvvLPbEGSlAm4SJoQQQojahR3d2TiKTdnYZBF353zf3y0WIV3pr4UMXyGEEEIIIYQQRY3W+AohhBBCCCGE+Mshw1cIIYQQQgghRFEjw1cIIYQQQgghRFEjw1cIIYQQQgghRFEjw1cIIYQQQgghRFHTcP78+RY/hBBCCCGEEEKI+kiSjasZXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRY0M33rCwIEDrWfPnvbxxx9nzwhRONLWr2Krj2pfQgghhBDFjQxfIYQQQgghhBBFjQzfiNmzZ9u///1v23vvvW3SpEnZs6K+U2yzeNRN6ih1lTor/tqoPgghhBBC5I8MXyGEEEIIIYQQRY0MXyGEEEIIIYQQRU29MnynTZtmd9xxh+2yyy7OhXXVVVe1448/3oYOHZoNUUroAvjTTz/ZWWedZeuuu66tvPLK9o9//MO++eYbF867OK+xxhr27LPP2ldffWUbbrjhYi7P8+bNc7/vscce7rnci3uOHDkyG2IREyZMsGuvvdY23nhjF5a/fOd8PsyZM8fuuece22yzzdz1PPO5556zBQsWZEOUJ03cckFYn0f+mdyTe8PMmTPttNNOs7XWWsveffddd87Dd87zO+E88XzYbrvt7O6777ZZs2ZlQyyCc/zmy9Wn4Y8//siGWFRWW2yxhSvTEF/evty8i/OZZ57pfj/ooIPc99DluSp1ibSSN9SjTz/9NBsimULWGR8H6iZ1lPtSZ+MurmnyPBfcj/rHtf4eDz74YFldqIx4un2+DhkyxDKZTDbUIvIpe0/aeydRm+2Lekp9vfLKK129Iq7EmePkk0+uchrzqQ+EHTx4sJ144oll7ZqyJO3kQZzK6qAQQgghRH2n0U477XQBH0LFsXfv3tlPdYdx48Y5BfCpp56yXr162Q477OCUwvfff98eeeQRW2+99axbt24uLMofyuAPP/xgjz/+uDVq1MiidFrLli3tgw8+sA8//NA233xza9++vbvmb3/7m7t/kyZN7NJLL7Udd9zRll12WXcdxtyFF15o119/va222mq2++67W/fu3e3FF1+0V1991TbddFPr0KGDe+5vv/1mRx11lL3++uu266672jbbbGMNGjSwJ5980r744gvbcsstrVWrVi5sEjzrvPPOcwbZMsssY3369LEll1zSbr/9dmd0zZ8/353z6UwTt1yQf0ceeaT9/vvvTtnGICAvUJCJ+0YbbWTNmze3FVdc0d58802Xp1tttZXLS5TiSy65xNWd888/37p27eruOWzYMJcP5PXOO+9sW2+9tVPo/+///s8p++R906ZNXViU+P/85z/OuEJ5Jw2knXJ+4403ytKAYcL30aNH21577eXKzuPLG/bcc0/r3LmzbbvttrbUUkvZ559/buecc44dffTRtsoqq7h4V7Uuvfzyyy7Pl19+eVe2Pr1xCl1nllhiCWds89t3333n4nrBBRe4surSpYs1bNgwVZ4n8e2337o4DBo0yL7++mvXJsiH77//3hmGvi7wrDB8Un285pprbLnllnPGGeknX++//35XrsSde0G+ZQ9p750E96jN9kX6nnnmGfvxxx/tiSeesE6dOrm6tnDhQnv77bdd3vq2BPmmsXHjxhXWB/IAuXfCCSe4+xJvZLqvD6NGjXIGOfIOqiu3hBBCCCHqGujEXi/kr/vcv3//DEekbGX69evnjrpIpBxnIoMn884772TPlPLNN99kIuMlc+aZZ2bmzp3rzk2cONGFXWmllTIDBw7MlJSUuPP8ve+++zKR4ZJ5+OGH3TmYNWtW5pRTTnHXcG3Io48+mllzzTUzkdFTdh+IlNZMZExkIsW97LmRsejuHSnF7jtwTWRMufMDBgzInk3mhRdecHE+9dRTMzNmzMiezWQiwyVzwAEHuHt89NFH2bPp4pZEZAC6+x533HGZSEnPns1kIgMgc9VVV7l7Dx48OHs2k4mUdxc/8jBS3t3feB6Hefnrr7+6cxDmQ2RIlp275ZZb3D34jXt6Pv74Y1euPg3+vqQrMvKyoUrx5R0vP/InnmdQ1boUj2MuaqrO+LiQD+SHJ02e58LnVd++fTPTpk3Lns24zyeeeOJidSEpb/25a6+91tUhj6+/5G1kqLlzacoe0tw7F7XdvqinhFtnnXUygwYNyp4tbV+RgeueF55Pm8Zc9YHwkSHr6nFYlv65pCkyaLNnqy+3hBBCCCHqGldffXWZbYtuhb1bb1ydDznkEDcTxOxDyNJLL+1m3piZYyYlhJlbZjBCax/3QGZpRowY4c5VBO6wzz//vJvZYTbF3wdWWGEF5zqICyIzOxApn+5vCNcwA/fQQw8598FcMBvz3nvvuZlMZifDGRbSuM8++2S/lZI2bkngrvvJJ5/Yfvvt52a+PMwo4XrKTC+zfx7SwYzqXXfd5Way+MtsEmH985kZZbaO2WNmrTz8zuwWM1ifffaZm8ElbswyMZNIWfnZRFh//fXdLH1laagKVa1LpDWMYxK1WWc8afK8Mvbdd19r06ZN9pu5z3//+98tMvDsnXfeyZ5dHJ9u3N4PPvhgV4c85Ou//vUvl2biCWnKPu29k/gz2pcnMlrLedEQ/8ggdp8jI9X9LUQaPYS/88477fLLLy9XltyT2WFmlilPTyHqoBBCCCFEXafeGL6A4heuy+Tw69ySwFU5VFihWbNmzqDLB1xif/nlF6egr7322mXP5MDVcMCAAU5hnTFjhgu//fbbOzfYk046ya23wyVy7NixTvlEgcbVNhfcA9dHDC9cIisjbdyS8GudcXMMr+fAyJs4caJ7jqdFixZ23HHHOZdZ1vTy99hjjy1nRODWiWJ98cUXL3ZPlH9cM6dMmeLcSidPnuzuv9JKK1nr1q2zdygFJR33zRdeeMG5bxaaQtSlJGqzznjS5HllhAaXB0MKg5p1rQwKJOHTvfrqq5cbRPGQDsoZl2zukabs0947iT+jfXkw6uN1J+46XIg0hpSUlLhBrXCNL4df9x5SiDoohBBCCFHXqTeGL0ohmxQxy4iidt999znl8+abb3ZKW03CDBRKYdJBnPysCgoxa/FYM4tfOUoniiPGFWv6UEYrAwMDAzNf8o1bLlDAmdFLup4DpTkE5ZmZRWA2nbW/STBzmHQ/Dma7MCQ9pDfJ4KopaqMu1Wad8aTJ8zRQR+LGaS7atWuXuJaYc0nn05R92nsnUdvtKy2FSCMz+7fccotr1xjnrMOnfnMw2x2nkHVQCCGEEKKuUi8M30wm4zZrYVOWW2+91fr27evcVFHme/Xq5Tb/qUmYaWGjGJ4bP9gkic2UPGyMc8YZZ9hHH33kZlxuuukmN+NzzDHHuM2RckEYDABcH0M3xMpIE7dc4OaZdD0HbqghbGr08MMPu88o0rlmSNkgJ+l+HGwgFCrxKNaUcW1QW3WpNupMnDR5noapU6e6Gdp8YGYZt+I4zEgnzY6mKfu09w75M9tXGqqTRs+vv/5qjz76qNsci8Gd3XbbzdVvDty0kyhUHRRCCCGEqKvUC8OX12+gsDK7GJ+RQ3GuqRkJXA5xi0QRZDfhikApZSdkXqWD4orSyK7CzJpcddVV1rZtW7c+L5fSzc6upI8dVseMGZM9m5s0ccsFeYlCzU6++Rgf06dPtxtvvNHNALKGEKW/f//+Tln3MKOGgcF60CQFPsSngbWSGFchxIdZqyOOOMK52JKfVZ2xDKnpulSbdcaTJs8rI2kdMPnPQAE7neeaLSXN5Cev7UkykocPH+7W1WNIco80ZZ/23kn8Ge0rDYVIo4clCtTxDTbYYLHBjnj9LlQdFEIIIYSo69QLwxeDB2WP92KiKHtQ1Jh19GtVqwv3Cw0HNsLhtS64xjJzgpHoQYHkdR9s8sRnFEw2GbrssssWe8erNyZwi8xlvHE9rxhh7SlGZfgsjA6eFZImbrlghhNFmtfJxOPMM6+44gq31g8wRrjna6+95maBeF0Q63uZIWIG2BvOrAfcZJNN3DpBjtCgxnBGmWZNKvg08GzuG4YlL3kFTMeOHd06T9Zm42aNQs+aVQ/XcH383b4hYd7UdF2qjTrDZkTh5ltp8rwyyHOu8WCEPfLIIy7/2SgrF8yU4zZO/vHKnNCAZv0qs+vkDZtCQZqyT3vvJP6M9pWG6qQxXh8oKwanGNAK480aZ9IekrYOco6BrjB+QgghhBD1gXrxHl+ULpQvNrthN152GmYmhvezYngBChyzFBg1bP7Cu1dZL4fy6t9XCSi+KKysa8P1D5gtwwh66aWXnBKIccU7M1FG2WyGGSJmRHiPLb99/PHHdsMNNziDEaODTZF4BjNvhOE8hhUGGnHG8OAZuBKyUVAu+A0lnDT6Z+FmyDtyMUDZnMi/Z5TNcvKNG7M4SZA/GJPk1WOPPVYuzjxz/PjxTtFmF2xcnP/3v/+5uoHrJ8o1M4DMQmEEYHixcRAbh/FOVtYGYrzgCs1sFZsEsWER+YwBwrWkgVmun3/+2c06+bA8H0Wc/D/33HPdbBgwa8fsE++VJa7ECZdM1idSf5mpYtdpPyPGbBZlzbtIvQHP8wpVl5JIUy5VqTODBw92hiLurMyGci9m5fLN81z49/JSx+69915XD3m37UUXXWRffvml29SMtd2+LiW9x5cZVWYbuZ68JH6UF+/GxoCk/tDmyKO0ZZ/m3rmo7faVJGs83Iuyok5RdlCVNCbVB/KMmW2eTfq4lo3ceD8wM7ekk920eS6yNU0dZBDg0EMPdYY1M8pCCCGEEHWRpPf41gvDFzBEiRfKHTMZKHM77LCD9evXz23gwm6nrEdlViSt4UtGoDDi0ojxMGHCBPcqD7/RDGsnUUpRzjGO+YuyeOGFFzpF2c+GYCASJ5TFV155xd2LexIvlEieURE8i1k13CxRfLmecjn99NPdDCtKbGhopIlbLjCGMPIwEjFkeOULs3y8ggWjp3v37m4GkA1yMHL5y/2B5/M7s4wYQqyVRSEmH9j4inxHoSYdv//+u0sD+cArWzwYsxjXGG8YoNyLPOOVQ/75HmYAebUKdeCtt95yBi2v6iGtzHrhMhsavswWYuTiAoyBgPFHGRSqLuWipuoMz6beMjhDvlJvee0Pgw1p8jwJys8PJOAFwDpo8pg8P/vss52xE6Y9yfAl3dQB0k1ZYjzhustGSVdffbX7zQsgSFP2ae+dBPeozfaV1vBNm8Zc9YE6z/NoixjqH374oau/Z511lmvX1GfW+tIGuCZNHaTNcD/qGm1PCCGEEKIukmT4NuBlvpzAZc8bv7yqRgghhBBCCCGEqG8woeUNX7zz+Fwv1vgKIYQQQgghhBBVRYavEEIIIYQQQoiiRoavEEIIIYQQQoiiRoavEEIIIYQQQoiiRoavEEIIIYQQQoiiRoavEEIIIYQQQoiiRoavEEIIIYQQQoiipmHfvn2N4/HHH7f58+dnTwshhBBCCCGEEMWBZnyFEEIIIYQQQhQ1Dfv3728cffr0sSZNmmRPCyGEEEIIIYQQxYFmfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUyfIUQQgghhBBCFDUN+vfvn+FDSUmJZTLuo5122mnurxBCCCGEEPWdhSWlR0mk6np9VwhR92jQoIE1bGDWqGHpUVX69evn7gUNGzYsva/7JoQQQgghRJGBjTt3fsbmLchEhm9GRq8QdRzaKG2VNkvbLWSTleErhBBCCCGKknkLM26WVwhR/6Dt0oYLhQxfIYQQQghRdDjX5ugQQtRfaMO05UIgw1cIIYQQQhQdCwo4UySE+PMoVFuW4SuEEEIIIYoOuTgLURwUqi3L8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEEIIUdTI8BVCCCGEEIn89NNPtsUWW1jPnj3t448/zp4VQvwVmD0vY/e+M9EdfK7vNGzQoEH2oxBCCCGEyJdMJmM//PCDnX322bbxxhs743Dddde1E0880YYMGeJ+F/WL7/6YYxue9b2tcPK3FR73vTMpe4UQxcvgn2fZJU+McceLX0zNnq2fYPNqxlcIIYQQIiULFiywu+66y3bffXd75JFHbMKECe78tGnT7MUXX7R99tnHbrzxRhdOiLrEuQNHOeP9iNtG2Ky5JdmzQixOl3aNbel2Tdyx9rItsmfrLzJ8hRBCCCFS8uqrr9rVV1/tDNsDDjjA3njjDRs8eLC99957dtRRR7kwN910kwsn6ifn7dvVfr5hjcTjsK3aZ0MJUbys1LWZvXfRyu7gc31Hhq8QQgghRAqmTJliDz30kDN6DzroILvwwgtt+eWXt/bt29syyyxjp59+uu29997u9+eee85mzpyZvVIIIcSfRYPrrrsuwxqUkpKSsrUop512mvsrhBBCCCHKwyZPhx56qDVr1szuu+8+22CDDbK/LOKpp55y+tSyyy5r9957rzOMAZfoAQMG2NNPP22///67LbHEErb55pvbcccdZ2ussYZbhwYDBw60M88809ZZZx278sor7dFHH7UnnnjCZs2aZRtuuKH17dvX/Q33apk3b569/PLLzgWbNcZNmza1bbfd1o4//vhy9wb0vvfff99uvfVWN1PdsGFDt0756KOPts0228x9Bza3Ovzww+2PP/6whx9+2Hr16uXO++vvvPNO++STT9yzV111Vdt3333dYEDLli1duNmzZ9tZZ51lzz77rB1zzDG21lpr2WWXXWadOnWyu+++2w0W1BRV2YyHNb6H3vSrTZ650M34VjSzO3XWQjvurpH2yU+z7G9rtbH+/+hmLZs1jPRps1tfG2/XPD/OOi3R2O46bjlbo1tzd83YqQvs9ui3Jz+ZYtPnlFj71o1s9/Xb2rHbd7TObRu7MJ6Z0e93vzXRHn5/ko2btsBaNGlgW6/Rxk7drZP17FQ6+zZh+gI7+Ppf7edxcxeLr08LPHBiD/s4iufFT4x230O2Wr213XR4dxf3kijubwyZbre8Os6G/jbHFpSYdWvfxPbttaQduU0Ha9W8tF6Ezz1rry5RXpe4dG28Uquye8UJ49P/n93trSHTyvJhuaWa2ll7d7Ht1mxjb387w/q/OMa+/2OuNWtcmuYz9ujswoTMmZ+xZz6dYre/PsFGTJjn8meDni3tlF062Xo9Wkb1PRswYngUz/4vjLP3vpvhntcmSscWq7W2vrsuykvcvk+89zd7Z+gMl6YlWjSy26I0cW/CH7pFezt+x05RWcy3q58dG8Vzus1dkLFVl2lmfXfpYtus0brcM6kH3/4+29WDT3+aabOj+FIfDtqsfbm8rAifxnvenmjDx85153p2bmZHbN3B9tyonTWP0uzheV/8Osuuf3GcfTZ8lntervqV733D9nDfv3pEedaq4PmULy2apruoX79+ZTIPecbRaKeddrqAE+EGDL17985+EkIIIYQQIYMGDbLXXnvNVlxxRfv73/9urVu3zv6yiBVWWMEOO+wwZwR27NjRKV0//vijO4f7M2uBu3btalOnTnUbZL3wwgu29tprO0MZvv32W3v99ded4Ygx++6779rcuXOdwYnBzHd2W15qqaVceGaVzz33XLvuuuts3Lhx7tzChQud4frYY4+5Z62++upOEWQm+oYbbnAGKffinoQdOXKkPf/88+XCTpo0yZ555hmbPn269enTx7p161bueq5p1aqVtWjRwt0LV+/Ro0c7Yx7Dm7C4gZPGMWPG2EsvvWQTJ060Ll262J577umuqykWLMx+SAEG3VORMYZhsNXqbWzdHrnj17xJQ+veoam9+PlU+2nMXFttmRbOHfT7UXPsfwNHu3v0jYzU7ddawin63/4+xw6+4Rf74MeZNi8yBADj/KsRs6NzM2ybyOhr07yROz96ynz7x82/2vOfT7OZ2XW4GKE85/EPJ0dGRAtbvlNTmxUZnE98PMUZJvH4+rTAPhu3c4b68LHzonzJuOdjKBJ/4rz1Gku4cJc/NcYueXKMM9AxOptGaZw4Y2FkNM+0H0bPsW2jcE2j8+Fzh0Xn34qMm/lRfvfo2NR2Wa+tNYnCxPHxIR7PR3n2+a+zy/KBc699Pc2+HDErMrrHR89faPzi04xBt+M6bSPjp9RYnBEZr6fc/5vd9eZEdy0QduTE+fb8Z1NdWZA/QL4fdvOI6N6zrVEUra5LNo3ivSBKz1x75ctptunKrZ1BOj/Klxej7yPGz3OG6itRfPy9ieenkTH56fCZdv9bk6L4zHbPI47jpy2016OwG67Q0rq1L30mZtVD70+yE+7+zX6J7kdYoCzJy7eHTi9X3kmEaZwUlUGLZpHhFsWfQRAGJ2bNXWi9V2ltjRqW5vVb386wY+4YERn581w+LdWmsctz0h3WL9J51TNj7JoXxrn7dooMYsr0j0nzF7tv2B72igzi5aLyLWQ+paEJiU/BBx984GQY8Jej8qEGIYQQQghRxogRI9xfDDuOJDjPbGa7du2scePGbkaUmd9ffvnFdtxxRzdLyozpZ599ZnvttZczhB988MHF3KI5v9VWW9mHH37ojFg20sKdGuMWg9Lz5JNPullmZpCZxSUss77M4GJ88mxmbQHDmzDMQjOTTNihQ4fa+eef736/7bbbnEGbC+57zz33OIP35ptvts8//9ylg9lvno97N7PIcXg+s+M8+4wzzqhRo7cQMDsa382Zg82hPBv0bGUHbd7ezZTe+eZ4Z5Tc+/ZEZxBuvGJL2ycyFtC9+X7uo3+43w/ovaR9fsWq9tP1a9hLZ61gK0eG53d/zLX7357kDCYMhRteGufOYZA9eFIPF/aDi1exLVdr5Wbybn51XJmxkS99erWz185d0XbbsK37zuzss/9Zwa44eBk3m/bFL7Ps4fcnO4P36r8vY1/3W92+vmo1u+3oZZ2RzEzsBz8s7rY/PkrTtpFRddMR3e3o7ZayRpUYKOTVpiu3tHcuWNl+vG4NG3BiD5dOZgWZReQeX0XPHRI9/7TdOhl23ZCRc2zoH3Pc9eQRM4xvDpnh8o48JH/IU/KW/LnuxTHOqOOet0aGNPm+xSqtXB6+ff5K9sb/Vo6M42buPL8TLqRty0Z2/wk97Pv+q9t7F67s8h0+/WmWLbNUE3vjvJXKxZ1nvhwZgx4GP66LDEvKcq8ovwdfvqoNi+Low1O2DDJgUObila+muTQu2aqRm7GnLCiTc/ft4vLkicgg/T6bJ1Oi+nXDy2NdPA7efEn76NJVXbyfOLVn2fMeeLd0J/KwnG85cll7/6JV7OMovL8vvxEmH6qbT7WNDF8hhBBCiBqmSZMmboYUg5BNsfxMbZs2bWyTTTZxn5kRZVY3ZLXVVnOvR+rcubObNcb1GZdkwCjGY4+/zEADG2vtsMMOLizuxrgpc4/vv//eGZ4Y1himGMPMVm+00UYubPPmzZ2b8tZbb+2Mc4zZXDAb/M4777jZ3e23395dz2zKmmuu6Yxp7j127Nhs6EXgdt2/f39n+ONdWNcN33xoHGnSh2/TwRlgX4+YY2c88LubccRQ/PcunZxhABgCGG/MsJ6wQ0d3HoN45a7N7eSdO7kw73w33RnIv02cZ+99V2pg9t21s20aGaiExVX1lChsq2YN3WzbqMnzXZhCsU6PlvZuZLwMigyh3Tdo64wgnrtejxa29JJNncE6asriz9xt/bbO6N153SVcXDGoKuOIrTs6F+pGUf71jq5hNhLWX76FHbXtUta6eUNnjPfZZEnn4syzh40ubRvMhnvjibwjD4kneUrekscYet9FxufChRk3Ow2d2jVxs6bAszH47zhmWTt86w5ls6ae7dZuY5tHhjKzjEsv2cT+sWVpeyXvT9+ts5vZjscdV1+/S/ZzUR2gLHHvPWPPzs545RGbrdzKuaTzmZn/X8aVb+8h5OfHl65ir527km2SrQNct8mKrVxap88uiQze0uctiDJo1pxSI7prlE5fBqt3b+HKhnTusl7prP7cqAgx9JtGdbTTEqX1kPvuu1E7u/u45ezGKPzynfPbyKq6+VTblJa+EEIIIYSoMTAMmQ2dMWOGc0fecsst3Xt/OVjLm4v4rDLrilkfCxiXc+bMca7DfoZ2vfXWc8/y4FKMG/Xw4cPd+lxcl5nhhYsvvrgsDhy4WuNeDRi/uSA+uHfj8n3EEUe4tb1cz5rjr776KhtqcbiGAYD6Qq5dnS/Zf+lsiFJ41ctJO5XOTA76YaYzKpgFZjbYgwsvxhtG7ZYX/FhuBvnEe35zYSZOXxgdC5xBO3bqfGsXGUtrdi9dG+xZNzJOv756NTeDuNoy5X+rLhhLrZs3sje+mW77XzfcVvl3afx6nfODW8ubi1YtGqZ2Qw1xhlfWImkVPR+3Ww952jCoz/B7lIfe6Cfvwrwkb8lj8vrXKM6sNWY2mvvgmr3tRcPcO2kZWOjRsZlbU8y6YAYwKqIxN4ggbhh1njDuHtzXh40pnYldZ7mW1nmJ8nV+tW7NrXPbJs4wHjMl9+vOiHuLJg3tkQ8m286X/2QrnVKaxt2u/NldG9KuVWPbao3SJResKd7lip/c38+Hz7Q1l23h0rlmZATDKpExvka3Zs5wPvD6X9xrre5/d5JNjO65+aqtXdiObcqvN8+XNPn0Z1AHoiCEEEIIUX9Ybrnl3F/clznyBdfmPfbYw7kJs472hBNOsJNOOsm22267bIiqwfpcDsCtuiKYUcZYBoxVZpCTDtyYc8GsMbPXJ598sn399ddu7S/pwK3aG+V/NTZduZWtvVypYYHBstM6S5QzpqbNKS0fZoKX79jUbagUP7p1aOwMhJIoKIYbJkTM5qtRWFN60r0j7cyH/7Bho+baTusu4TbLwn24TWTc1hX8jCU2FrOMSXnJ0SwyGuHgzdrbTUcs68JiMN/7zkQ77NZfbYMzv7OzHxnl0l1I8MJYmL0lxl68DNkAqmV2rTKzn7lgZvuA64dbv+fGurW2rNOmPPbYoO1is+rUtdN262zn9+nqBkx+HD3XrZU+6IYonf/9zm58ebxbmwu4HN95bA/bt1c79x338oseH21/u3iYbXXBj/b+j8W7C33dqcVCCCGEEPUA3HkxMJll9etm47z44ovOsPSuwxiLrOHFLflf//qXex0Suz6zOzPuwtWhUaNG7gDcjCsiDHvqqae6dcFJx7HHHuvCJIG7Nu7SrDVmjfAll1zi0sE1GPR/NVhz+vTgKfblr7Pdd9w474+Mq3D9ZpPsTNiKXZvZ46f2dGtt48eTp61gK3ZpZg2j4iE4V3Pv2oL1u6zjxS33kX8vb9cf1t3Nep+6ayfr1KbuzNT7/GGW+ZpDl0nMSw7WNLvwUdgd12nj1hRzXHXIMm69L4MLj34w2a55fmzZ5lOFAI8L3HuhJLpvvAzZVdq7X8d3qg55ZvBU57KNG/1z/13BroziTXkc87elEnfNZifmf2zZ3j66ZBV7NUr//yIjeO3lmmfXPI+zhwZNzoYsdZsnH764cjV7+OQedlx0TzbDYmDglHt/i55bOjhWbLhcC11ihBBCCCFEbnr06OHWzWLM4kYcNzaZBWYNLBCOWVBmWlnDC8wYhzOz7KpcHTp06FC2G/QXX3zhZpw8uDazRpj3Cn/55ZflwrLJFbtGh3BtZfHBxZo0s06ZHas9XFfdtNRHfhw9x+54bYIzsFiXyezoS1+yMdH0bAgrczP9cRQ7FJfPcwgNL2YmcYVlw6Ihv5U3QFjnule/n22fa352ux3jWtqyeakeP25q+fW3VIM0djPrdzEG27dqbF3aLTJ0OVdSmxZ4Jfj8YdaXfI4brcTXR5c08Uqdq6IDo6701Uzt7N7jezj3dPho2EyblnKjsIpgbfJKXUrd0IeMnLWYW/J3v89xruwMMHRpl9tD4/dJpbPBXaP0tmtROlgFSeVKHWQzthteGu8GXlbo1Mz+GRnBA//d080Qw/s/THdu2E9HBjVhn/xkitv9eeMVWxmvi3r81OVd3hLfz/Pc3Kq+4G1dzfgKIYQQQqSAjab++c9/OuOVWdyrrrrKRo0a5d6xywww749k1pTfd999d+c2zAZQ3tj96KOPbMqUKc5I/PTTT927cKsDa4f9rDEzsOwAzb1xaWZmFgOXuC255JIu7G677ebCMivNa4l4VRGMHz/eLr/8cjd7O3nyotmhOH6dLjPZuDrzLO5BOr755hv3218FZnVvfqV012Be03LJAUvb37O7PN/48riyzaA2W7WV2+gIY+1/j45yGxsRBvdTXiGz06XD7LnP2KzM3OZMvC8V7nxjQmQ8zXbn2cUZN91vRs5xhk+HNo3dZk3+1TCvfz3dGcbcF0PogsdHOeM5Ts+OpRsXERZjkDhwTZOsbTVq8jz7bPhsd45nMlvIK3nqCuTP9muXbpT0f4Mm2d1vTnAGHXnE+3qPvG2EnRvlsXuPc/Q/uyPzrl/C+Vl4ymH89NKy6dC6/LriQsDmYBi23/4+1/o9P9blI/GjLHmfMHnbe+VWtnxkoObCewlgKFOeXE95Xf3cmMXKdW6ULnZ5ps49+WkkW0qTabMjI3jyjNKBOWZ0ec3UmKhO3vfOJDcg8E0UHw+7YPM+Zh7bsQ7N8BcSGb5CCCGEEClhrS4zqXDXXXe599ayqzHv1uU78Ds7LEPbtm1t//33d8bv008/beuvv757D/ABBxxQ9t5dNr6Kv84oX/bZZx83q4vhzW7N3Jvdly+88EJndOOG7Gd6ibt/zdHtt9/u1vTils3mV8SdGd34THDIZptt5naWxm2bza14FvfglUnsDg0Y0fWdXK8z4vCvNGJWl1lH1u7+a/uObn3lgZu1t2Uj44x1lve+NdHNSHZp28TO37erW1+JMXzoTb+6zYpW7TvUjrljpFvrOWHGfGfQsl6T3Yp53Q6G3J79htuKUdj1z/zeueayRvTknUp3Cmat5/6bLumezwZUO2U3Qdr58p/dLtNZ26kcbK7EdRhRO172k+0ahZ0UGUdsAsUzcY09/u6R7j4887EPJ5dtXhWfVf4zIH9O3Kmje3UOects7pqnD3V5tP0lP9m7381wM6rzox+ZwTx7764ufwa8O8nWOWOobX3hMNvwzO8io3myO3/ENh3dLtKFZNWlm9u/d+3k4kqZkY/Ej7KkTMnns/bu4tyTc8GO1tQXBlX2vfYXd/0W5/9oH/44q6xcx00rLQ88CtgBHNi8a+0oP0jnRmd/b+/9MNPd5+9bdHDxYWBm2zVbu/vufc1wF6b3eT/YPtFnZnu3XqO19V4l9xr/+kxhS1kIIYQQ4i8ABiwbOvH+3F122cXNpALuv/vtt587z+9+lhdXO14XdMcddzgDGQjL+2x5by6zwrwfmJnjqsD1zNayY7S/P7sv77TTTvb44487o9i7+2Gc/ve//7X777/fGex+12h2Zz7nnHPc5ltLL11+5+IQXLdvuukm96ok0k0auQ/rlvfcc08XZtiwYRUaz8UABiwzbMyu7bJ+W7fBFeBOe0JkmGGcPDxokn3yU+lgBi6lT5+xgnNBbd+6dHoVw4vXwTz67+XtsC07lBk0vJLm0VN6utcXYbQAYXGlfvL0nrZNdgdf2GLV1tb/sO5l60UJx/rW249etux1SiEb9mxpZ+/TxRnp0K51Q+N1Pjzz7uN6uGu5B3FhjShuwdutVTrDyvtp/6xX0YR0aN3Ybj1qObvswKVL10VHceXgM+fYzGqJbNox8p797wrudT649rLrc+PIkOf7Y6f2dL8XGpraIZu1t4dPWd6VL/kJlCVlStmS3xWxRrfm9tDJPZyBj8HKweeHTupha2R3+/4q6w3A8/6xRXtXj3gekE7y4JDoPOnkfoCRf+Phy5blHbPHvIuZz2yOdcNhyxZ8IKCu0CASkG4ynN0A/ZoQNlsQQgghhBCivuJcXYUQRQFrp9PAkhM/2MeGfnzWjK8QQgghhBBCiKJGhq8QQgghhBBCiKKmzPD1U8FCCCGEEEIIIUR9J7RxNeMrhBBCCCGEEKKokeErhBBCCCGEEKKokauzEEIIIYQQQoiio5yrswxeIYQQQgghhBDFCjZvOVdnGcFCCCGEEEIIIeo7cdtWa3yFEEIIIYQQQhQ1MnyFEEIIIYQQQhQ1zvANp4Hl7iyEEEIIIYQQor6SZN9qxlcIIYQQQgghRFEjw1cIIYQQQgghRFGj1xkJIYQQQgghhChqymZ8ZQALIYQQQgghhCgWQhtXrs5CCCGEEKLoaKg5HSGKgkK1ZRm+QgghhBCi6GjcKPtBCFGvKVRbLmf4yt1ZCCGEEEIUA40aNtCsrxD1HNoxR1WI27aa8RVCCCGEEEVJ08ayfIWozzQpoOdGuc2tGjZsqFlfIYQQQghRFKDWtmjawBnAzBpJzxWibkMbpa3SZmm7VW2y3Cc83LkbbrghE+G+8LekpMT69u3rvgshhBBCCCGEEPWJ/v37L2b4ytVZCCGEEEIIIURRI8NXCCGEEEIIIURR4wxfP/0rhBBCCCGEEEIUE9i7mvEVQgghhBBCCFHUNNRsrxBCCCGEEEKIYsXN+CYZvrNmzcp+EkIIIYQQQggh6ge5bNnF1vjyLt/x48dnvwkhhBBCCCGEEPUDbFnsW2/j+r/l1vj6k19//bX7K4QQQgghhBBC1BdCW7bcBG/2b7mTP/30k40YMSL7TQghhBBCCCGEqNv8/PPPzpYNWWzGNzR84e2337aFCxdmvwkhhBBCCCGEEHWT+fPn21tvvZX9Vkpo4za4+eabMxHGUVJS4v4Cn2G11VazFVdc0bp27WqtWrVy54QQQgghhBBCiD+TmTNn2ujRo90s79ChQ52hm3Swj9Vihi/EP4M3iv0B/nP4W3jef4/fK/zNf/bfc5Hr9/B8ZfcQQgghhBBCCJEejMckMCwrw1/rw4b3Cg1UiIflrzNcg3D+AP8b+PPhPfz3crH3AULiN/GHJ+l8/Hv8weFnH1H/N35URhguvC7fQwghhBBCCCGKnSRbKJ/DGY0JdmISua71n/35pLDgnxOGSbrOk+s8hL+Bm/Hlg5919bOzfhYX/N9cM7sQ/hae8/jP/rdcfyHpnCfpXEhlvwshhBBCCCGEqJi4IRkn1++LGZyxz0nf/bn457jRGzeM/e/g/zZq1Mj99deVXZPL8A0/g/+9ovPxA5KMX08Yzv+FpHP5EA+f9nohhBBCCCGE+KvijcckKvrNE4bxn3P9BT6H58PP3mCN/wUftsyoTbhusTBxw9cf/lxoCIe/pTF+498h/AxJYUKSzucKm0SasEIIIYQQQgjxV8AbjfmQFDbX9f58aLB6/G/h3/Czvyb+F8KwZUZt9lwYrvy1Zv8PDFsxziKWmVMAAAAASUVORK5CYII=\"/></center>\n",
        "\n",
        "Este notebook é seguro e pode escolher \"Executar mesmo assim\".\n",
        "\n",
        "\n",
        "**Nota**: para o primeiro ficheiro vai demorar alguns 2 ou 3 minutos a mais (no Passo 2), pois na primeira execução precisa de fazer o download do modelo, 3.09 GB.\n",
        "\n"
      ],
      "metadata": {
        "id": "2wV-abtpVWO2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Início:"
      ],
      "metadata": {
        "id": "HJNN0wJEoDUT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Passo 0: Instala as bibliotecas Python necessárias\n",
        "\n",
        "#@markdown Escolher um Tempo de Execução T4 (GPU NVidia Tesla T4)\n",
        "\n",
        "#@markdown Só é necessário correr esta célula para o primeiro ficheiro, na sessão atual. Tempo de Execução necessário: aprox/ 1 minuto.\n",
        "\n",
        "#@markdown NOTA: Se fechar esta página e não a abrir dentro de 10 a 15 minutos, todos os ficheiros e parametrizações serão apagados.\n",
        "\n",
        "%%capture cap\n",
        "!pip install faster-whisper pysubs2\n",
        "#!apt install libcublas11\n",
        "!apt-get install -y libcudnn9-cuda-12 # Install cuDNN 9.5 for CUDA 12\n"
      ],
      "metadata": {
        "id": "6rM5c-a8CDxE",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Passo 1: Fazer o upload do ficheiro audio ou vídeo a transcrever\n",
        "\n",
        "from google.colab import files\n",
        "\n",
        "uploaded = files.upload()\n",
        "\n",
        "for fn in uploaded.keys():\n",
        "  file_name = fn"
      ],
      "metadata": {
        "id": "HCVFFXOr3z8U",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Passo 2: Transcrição em frases na língua original do vídeo\n",
        "#@markdown Irá fazer o download automático do ficheiro das legendas (.srt), quando terminar.\n",
        "\n",
        "#@markdown Para transcrever mais vídeos ou ficheiros aúdio, basta repetir o Passo 1 acima e este, novamente.\n",
        "\n",
        "from faster_whisper import WhisperModel\n",
        "import pysubs2\n",
        "import os\n",
        "\n",
        "model_size = \"large-v2\"\n",
        "\n",
        "# Run on GPU with FP16\n",
        "model = WhisperModel(model_size, device=\"cuda\", compute_type=\"float16\")\n",
        "\n",
        "# or run on GPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cuda\", compute_type=\"int8_float16\")\n",
        "# or run on CPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cpu\", compute_type=\"int8\")\n",
        "\n",
        "segments, info = model.transcribe(file_name, beam_size=5)\n",
        "\n",
        "\n",
        "# Caso esteja em PT e se queira traduzir para Inglês (comentar a instrução anterior)\n",
        "# segments, info = model.transcribe(file_name, beam_size=5, language=\"en\")\n",
        "\n",
        "\n",
        "print(\"Língua detectada: '%s' -- probabilidade de %f\"% (info.language, info.language_probability))\n",
        "\n",
        "results= []\n",
        "results2= []\n",
        "\n",
        "for segment in segments:\n",
        "    try:\n",
        "      print(\"[%.2fs -> %.2fs] %s\" % (segment.start, segment.end, segment.text))\n",
        "      segment_dict = {'start':segment.start,'end':segment.end,'text':segment.text}\n",
        "      results.append(segment_dict)\n",
        "    except Exception as e:\n",
        "        print(f\"Oops... erro: {e}. Sem problema, vou avançar para a próxima frase.\")\n",
        "\n",
        "# Separar o nome do arquivo da sua extensão\n",
        "base_name, extension = os.path.splitext(file_name)\n",
        "\n",
        "# Altera o nome do ficheiro para conter o idioma reconhecido e a extensão .srt\n",
        "srt_file_name = f\"{base_name}.{info.language}.srt\"\n",
        "\n",
        "subs = pysubs2.load_from_whisper(results)\n",
        "#save srt file\n",
        "subs.save(srt_file_name)\n",
        "\n",
        "import time\n",
        "from google.colab import files\n",
        "files.download(srt_file_name)"
      ],
      "metadata": {
        "id": "uFmxZ_kQOb2Y",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title ... passo 2 alternativo: Caso acima não tenha detectado Português como o idioma, força-o para Português\n",
        "#@markdown Irá fazer o download automático do ficheiro das legendas, quando terminar.\n",
        "\n",
        "from faster_whisper import WhisperModel\n",
        "import pysubs2\n",
        "import os\n",
        "\n",
        "model_size = \"large-v2\"\n",
        "\n",
        "# Run on GPU with FP16\n",
        "model = WhisperModel(model_size, device=\"cuda\", compute_type=\"float16\")\n",
        "\n",
        "# or run on GPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cuda\", compute_type=\"int8_float16\")\n",
        "# or run on CPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cpu\", compute_type=\"int8\")\n",
        "\n",
        "segments, info = model.transcribe(file_name, beam_size=5, language=\"pt\")\n",
        "\n",
        "\n",
        "# Caso esteja em PT e se queira traduzir para Inglês (comentar a instrução anterior)\n",
        "# segments, info = model.transcribe(file_name, beam_size=5, language=\"en\")\n",
        "\n",
        "\n",
        "print(\"Língua detectada: '%s' -- probabilidade de %f\"% (info.language, info.language_probability))\n",
        "\n",
        "results= []\n",
        "results2= []\n",
        "\n",
        "for segment in segments:\n",
        "    try:\n",
        "      print(\"[%.2fs -> %.2fs] %s\" % (segment.start, segment.end, segment.text))\n",
        "      segment_dict = {'start':segment.start,'end':segment.end,'text':segment.text}\n",
        "      results.append(segment_dict)\n",
        "    except Exception as e:\n",
        "        print(f\"Oops... erro: {e}. Sem problema, vou avançar para a próxima frase.\")\n",
        "\n",
        "# Separar o nome do arquivo da sua extensão\n",
        "base_name, extension = os.path.splitext(file_name)\n",
        "\n",
        "# Altera o nome do ficheiro para conter o idioma reconhecido e a extensão .srt\n",
        "srt_file_name = f\"{base_name}.{info.language}.srt\"\n",
        "\n",
        "subs = pysubs2.load_from_whisper(results)\n",
        "#save srt file\n",
        "subs.save(srt_file_name)\n",
        "\n",
        "import time\n",
        "from google.colab import files\n",
        "files.download(srt_file_name)"
      ],
      "metadata": {
        "cellView": "form",
        "id": "Ksc9GzIneJqr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Extra 1: vídeo com legendas \"queimadas\"**\n",
        "#### Nota, na célula \"Passo 1\", o ficheiro carregado deverá ter sido um vídeo (.mp4, .mov, .avi, .mkv, ...) e não apenas um ficheiro aúdio (.mp3, .wav, .ac3, .aac, ...).\n",
        "\n",
        "##### Caso seja necessário editar/corrigir as legendas, abrir o respectivo ficheiro .srt (clicar no icon \\\"ficheiros\\\" do lado esquerdo, por debaixo da \\\"chave\\\"), fazer as necessárias alterações e fechar ( = gravar) antes de correr a célula do Passo 1 abaixo."
      ],
      "metadata": {
        "id": "iOMJUBbfnFaA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Extra 1 - Passo 0: Instala biblioteca necessária\n",
        "#@markdown Só é necessário ser corrida da primeira vez / primeiro ficheiro em cada sessão.\n",
        "%%capture\n",
        "!pip install ffmpeg-python"
      ],
      "metadata": {
        "id": "TEcM1yiAnP2r",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Extra 1 - Passo 1: Gera Vídeo legendado\n",
        "#@markdown Esperar que processe o vídeo -- irá fazer o download automático do ficheiro legendado.\n",
        "\n",
        "#@markdown Se aparecer uma notificação que o browser está a tentar fazer o download ao mesmo tempo de vários ficheiros, escolher \"Permitir\" / \"Allow\".\n",
        "\n",
        "import ffmpeg\n",
        "\n",
        "def queimar_legendas(caminho_video, caminho_legendas, caminho_saida):\n",
        "    # Carregar o vídeo\n",
        "    video_entrada = ffmpeg.input(caminho_video)\n",
        "    # Incorporar as legendas no vídeo\n",
        "    video_saida = ffmpeg.output(video_entrada, caminho_saida, vf='subtitles=' + caminho_legendas)\n",
        "    # Executar o comando\n",
        "    video_saida.run()\n",
        "\n",
        "\n",
        "# Separar o nome do arquivo da sua extensão\n",
        "base_name, extension = os.path.splitext(file_name)\n",
        "\n",
        "# Adiciona \"legendado\" antes da extensão\n",
        "new_file_name = f\"{base_name}_legendado_{info.language}{extension}\"\n",
        "\n",
        "# Chamar a função com os seus caminhos de vídeo, legendas e saída\n",
        "queimar_legendas(file_name, srt_file_name, new_file_name)\n",
        "files.download(new_file_name)\n"
      ],
      "metadata": {
        "id": "pym18XqwnE2v",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Extra 2: Ficheiro de legendas em Inglês**"
      ],
      "metadata": {
        "id": "PpTUCCK8vNnG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Transcrição em frases, tradução automática para Inglês caso o idioma original não seja EN.\n",
        "#@markdown Para se gerar um ficheiro vídeo com as legendas em Inglês \"queimadas\" na imagem, basta executar a célula anterior (\"Extra 1 - Passo 1: Gera Vídeo legendado\"), após esta.\n",
        "\n",
        "from faster_whisper import WhisperModel\n",
        "import pysubs2\n",
        "\n",
        "model_size = \"large-v2\"\n",
        "\n",
        "# Run on GPU with FP16\n",
        "model = WhisperModel(model_size, device=\"cuda\", compute_type=\"float16\")\n",
        "\n",
        "# or run on GPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cuda\", compute_type=\"int8_float16\")\n",
        "# or run on CPU with INT8\n",
        "# model = WhisperModel(model_size, device=\"cpu\", compute_type=\"int8\")\n",
        "\n",
        "# segments, info = model.transcribe(file_name, beam_size=5)\n",
        "\n",
        "\n",
        "# Caso esteja em PT e se queira traduzir para Inglês (comentar a instrução anterior)\n",
        "segments, info = model.transcribe(file_name, beam_size=5, language=\"en\")\n",
        "\n",
        "\n",
        "print(\"Língua detectada: '%s' -- probabilidade de %f\"% (info.language, info.language_probability))\n",
        "\n",
        "results= []\n",
        "results2= []\n",
        "\n",
        "for segment in segments:\n",
        "    try:\n",
        "      print(\"[%.2fs -> %.2fs] %s\" % (segment.start, segment.end, segment.text))\n",
        "      segment_dict = {'start':segment.start,'end':segment.end,'text':segment.text}\n",
        "      results.append(segment_dict)\n",
        "    except Exception as e:\n",
        "        print(f\"Oops... erro: {e}. Sem problema, vou avançar para a próxima frase.\")\n",
        "\n",
        "# Separar o nome do arquivo da sua extensão\n",
        "base_name, extension = os.path.splitext(file_name)\n",
        "\n",
        "# Altera o nome do ficheiro para conter o idioma reconhecido e a extensão .srt\n",
        "srt_file_name = f\"{base_name}.{info.language}.srt\"\n",
        "\n",
        "subs = pysubs2.load_from_whisper(results)\n",
        "#save srt file\n",
        "subs.save(srt_file_name)\n",
        "\n",
        "import time\n",
        "from google.colab import files\n",
        "files.download(srt_file_name)"
      ],
      "metadata": {
        "id": "LK7IVEOLvk_L",
        "cellView": "form"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "---\n",
        "---\n",
        "***\n",
        "\n",
        "> Para fazer a legendagem de mais vídeos, basta voltar a executar a célula \"Passo 1: Fazer o upload do ficheiro audio ou vídeo a transcrever\" e seguintes pela ordem anterior, sem necessitar de executar os que são \"Passo 0\".\n",
        "\n"
      ],
      "metadata": {
        "id": "CIhfCwxix-fc"
      }
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}