From d5d8458093816be7b3caf4303b239e3c240582e3 Mon Sep 17 00:00:00 2001
From: Eren Golge <erogol@hotmail.com>
Date: Wed, 19 Dec 2018 12:52:30 +0100
Subject: [PATCH] Update README.md

notebook example
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 918e3d42..d579b372 100644
--- a/README.md
+++ b/README.md
@@ -62,7 +62,7 @@ TTS provides a generic dataloder easy to use for new datasets. You need to write
 - [Nancy](http://www.cstr.ed.ac.uk/projects/blizzard/2011/lessac_blizzard2011/)
 
 ## Training and Fine-tuning LJ-Speech
-[Click Here](https://colab.research.google.com/drive/1yUUBzdlavefXj3WRgK1SuWyJf92ME9_m?authuser=1#folderId=0AIBvuHEQn-h3Uk9PVA&updateTitle=true) for hands on **CoLab example**, training LJSpeech.
+[Click Here](https://gist.github.com/erogol/8f39174c3f0475221c8978aeb10d4fdc) for hands on **Notebook example**, training LJSpeech.
 
 Split ```metadata.csv``` into train and validation subsets respectively ```metadata_train.csv``` and ```metadata_val.csv```. Note that having a validation split does not work well as oppose to other ML problems since at the validation time model generates spectrogram slices without "Teacher-Forcing" and that leads misalignment between the ground-truth and the prediction. Therefore, validation loss does not really show the model performance. Rather, you might use the all data for training and check the model performance by relying on human inspection.