TTS_tf notebook update

tf/notebooks/Benchmark-TTS_tf.ipynb

@@ -6,18 +6,24 @@
     "Collapsed": "false"
    },
    "source": [
-    "This is to test TTS models with benchmark sentences for speech synthesis.\n",
+    "This is to test TTS tensorflow models with benchmark sentences.\n",
     "\n",
     "Before running this script please DON'T FORGET: \n",
     "- to set file paths.\n",
-    "- to download related model files.\n",
+    "- to download related models.\n",
+    "    - Sample TF model: https://www.dropbox.com/sh/3b1fat5oxqab6yn/AADDlNs-9-r7ASbVnFYx3RHHa?dl=0\n",
     "- download or clone related repos, linked below.\n",
     "- setup the repositories. ```python setup.py install```\n",
     "- to checkout right commit versions (given next to the model in the models page).\n",
     "- to set the file paths below.\n",
     "\n",
     "Repositories:\n",
-    "- TTS: https://github.com/mozilla/TTS"
+    "- TTS: https://github.com/mozilla/TTS\n",
+    "- PWGAN: https://github.com/erogol/ParallelWaveGAN (if you like to use a vocoder model)\n",
+    "\n",
+    "Known Issues:\n",
+    "- To load the model second time you need to restart the notebook kernel. \n",
+    "- Some of the advance methods are not yet implemented for Tensorflow."
    ]
   },
   {
@@ -63,7 +69,7 @@
     "import IPython\n",
     "from IPython.display import Audio\n",
     "\n",
-    "%matplotlib agg"
+    "%matplotlib inline"
    ]
   },
   {
@@ -112,7 +118,7 @@
    "outputs": [],
    "source": [
     "# Set constants\n",
-    "ROOT_PATH = '../tf_model/'\n",
+    "ROOT_PATH = '../torch_model/'\n",
     "MODEL_PATH = ROOT_PATH + '/tts_tf_checkpoint_360000.pkl'\n",
     "CONFIG_PATH = ROOT_PATH + '/config.json'\n",
     "OUT_FOLDER = '/home/erogol/Dropbox/AudioSamples/benchmark_samples/'\n",