{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9d8dbc75-dacc-457f-890c-c3d70f1e24ff",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "from lhotse import CutSet\n",
    "from lhotse.cut import Cut, MonoCut\n",
    "from lhotse.recipes import (\n",
    "    download_voxceleb1,\n",
    "    download_voxceleb2,\n",
    "    hifitts,\n",
    "    libritts,\n",
    "    prepare_voxceleb,\n",
    ")\n",
    "from lhotse.dataset import (\n",
    "    SimpleCutSampler,\n",
    "    UnsupervisedDataset\n",
    ")\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "181a454e-1e74-40fd-8d5b-765b65bf6111",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Dataset root path\n",
    "root_dir = Path(\"datasets_cache\")\n",
    "\n",
    "# datasets paths\n",
    "hifitts_path = root_dir / \"hifitts\"\n",
    "libritts_path = root_dir / \"libritts\"\n",
    "librittsr_path = root_dir / \"librittsr\"\n",
    "\n",
    "num_jobs = os.cpu_count() - 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7e6c3a68-b4e9-46ea-89b2-4aeed61f4a64",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading LibriTTS parts:   0%|          | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d9a20b8128847dca5a17a4ef95f2fb5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "datasets_cache/libritts/train-clean-100.tar.gz:   0%|          | 0.00/7.72G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading LibriTTS parts: 100%|██████████| 1/1 [04:05<00:00, 245.05s/it]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "PosixPath('datasets_cache/libritts')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prepare the LibriTTS dataset\n",
    "libritts_root = libritts.download_libritts(\n",
    "    libritts_path,\n",
    "    dataset_parts=[\"train-clean-100\"],\n",
    ")\n",
    "libritts_root"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5a0ad3a3-ac47-4498-8f84-afe4d5188780",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading LibriTTS_R parts:   0%|          | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8a542acbc35f4550bc3f93d48393a48e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "datasets_cache/librittsr/train-clean-100.tar.gz:   0%|          | 0.00/8.13G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading LibriTTS_R parts: 100%|██████████| 1/1 [03:54<00:00, 234.67s/it]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "PosixPath('datasets_cache/librittsr')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prepare the LibriTTS-R dataset\n",
    "librittsr_root = libritts.download_librittsr(\n",
    "    librittsr_path,\n",
    "    dataset_parts=[\"train-clean-100\"],\n",
    ")\n",
    "librittsr_root"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ec51bc2-e8bc-4278-88b4-630bd0743a18",
   "metadata": {},
   "source": [
    "### Check restore pipes of voicefixer for original and restored versions of the libri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bfe4d675-29d8-404e-82dc-6e12ca215a4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from voicefixer import VoiceFixer\n",
    "import soundfile as sf\n",
    "import tempfile\n",
    "\n",
    "with tempfile.NamedTemporaryFile(suffix=\".wav\", delete=True) as out_file:\n",
    "    voicefixer.restore(input=input_file.name, # low quality .wav/.flac file\n",
    "        output=output_file.name, # save file path\n",
    "        cuda=True, # GPU acceleration\n",
    "        mode=0,\n",
    "    )\n",
    "    wav_vf, samplerate = sf.read(output_file.name)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "TTS-Framework",
   "language": "python",
   "name": "tts_framework"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}