diff --git a/Readme.md b/Readme.md
index 49ae9c9..526e52e 100644
--- a/Readme.md
+++ b/Readme.md
@@ -6,9 +6,10 @@ This repository contains ROS packages for the CereProc CereVoice TTS engine.
 0.0 [Install ROS Hydro](http://wiki.ros.org/hydro/Installation/Ubuntu) (or maybe Indigo or J-Turtle in the future) if you haven't already.
 0.1 [Create a Catkin workspace](http://wiki.ros.org/catkin/Tutorials/create_a_workspace) if you don't already have one.
 
-1.  Unpack CereVoice into a directory (e.g. ~/cerevoice, ~/CereVoice, ~/Cerevoice, ~/cereproc/cerevoice, ~/cerevoice_sdk, /opt/cereproc/cerevoice or /opt/cerevoice (to add more possible install directories edit the [CMakeLists.txt](master/cerevoice_tts/CMakeLists.txt))) [Location Seafile](https://cloud.bitfactory.org/repo/a68f0d25-4e20-4ecb-86c0-1a598ddd9408/?p=%2FCereVoice%2F)
-2.  Copy your language and license files to a directory. (e.g. create a voices directory in your CereVoice directory)
-3.  Clone this repository:
+1.  Clone the CereVoice SDK:
+    git clone git@141.69.58.11:various/cerevoice_sdk.git
+
+2.  Clone this repository:
 
     $ cd ~/catkin_ws/src
     $ git clone git@141.69.58.11:ros/cerevoice_tts.git
@@ -67,4 +68,4 @@ Example:
 
 
 The path and the license of a voice are required. Lexicon and abbreviations file are optional.
-The optional string startup_sentence will be synthesized when the TTS node is ready.
\ No newline at end of file
+The optional string startup_sentence will be synthesized when the TTS node is ready.