diff --git a/src/fenrir.py b/src/fenrir.py
index 41e0a541..75863e20 100755
--- a/src/fenrir.py
+++ b/src/fenrir.py
@@ -11,11 +11,14 @@ import speech.es as es
 
 runtime = {
 'columns': 0,
+'lines': 0,
 'screenDriver': '/dev/vcsa3',
 'delta': '',
+'oldCursor':{'x':0,'y':0},
 'oldContentBytes': b'',
 'oldContentText': '',
 'oldContentAttrib': b'',
+'newCursor':{'x':0,'y':0},
 'newContentBytes': b'',
 'newContentText': '',
 'newContentAttrib': b'',
@@ -24,15 +27,23 @@ runtime = {
 }
 
 while(True):
+  # read screen
   vcsa = open(runtime['screenDriver'],'rb')
   runtime['newContentBytes'] = vcsa.read()
   vcsa.close()
+
+  # get metadata like cursor or screensize
+  runtime['lines'] = int( runtime['newContentBytes'][0])
   runtime['columns'] = int( runtime['newContentBytes'][1])
-  runtime['newContentText'] = str(runtime['newContentBytes'][::2].decode('cp1252').encode('utf-8'))[7:]
-  runtime['newContentAttrib'] = runtime['newContentBytes'][1::2][7:]
-  print(runtime['newContentBytes'][9])
-  #runtime['newContentString'] = str(runtime['newContentBytes'].decode('cp1252').encode('utf-8'))
+  runtime['newCursor']['x'] = int( runtime['newContentBytes'][2])
+  runtime['newCursor']['y'] = int( runtime['newContentBytes'][3])
+
+  # analyze content
+  runtime['newContentText'] = str(runtime['newContentBytes'][4:][::2].decode('cp1252').encode('utf-8'))
+  runtime['newContentAttrib'] = runtime['newContentBytes'][5:][::2]
   runtime['newContentText'] = '\n'.join(textwrap.wrap(runtime['newContentText'], runtime['columns'])) 
+  
+  # changes on the screen
   if runtime['oldContentBytes'] != runtime['newContentBytes']:
     runtime['speechDriver'].stop()
     print("tty3 changed")
@@ -40,10 +51,12 @@ while(True):
     diff = difflib.ndiff(runtime['oldContentText'], runtime['newContentText'])
     runtime['delta'] = ''.join(x[2:] for x in diff if x.startswith('+ '))
     
-    #print(runtime['delta'])
-    #runtime['speechDriver'].speak(runtime['delta'])
+    runtime['speechDriver'].speak(runtime['delta'])
 
+    # set new "old" values
     runtime['oldContentBytes'] = runtime['newContentBytes']
     runtime['oldContentText'] = runtime['newContentText']
     runtime['oldContentTextAttrib'] = runtime['newContentAttrib']
+    runtime['oldCursor']['x'] = runtime['newCursor']['x']
+    runtime['oldCursor']['y'] = runtime['newCursor']['y']
 
diff --git a/src/speech/es.py b/src/speech/es.py
index 4095d06a..4b7ab75a 100644
--- a/src/speech/es.py
+++ b/src/speech/es.py
@@ -6,20 +6,44 @@ class speech():
     def __init__(self, ):
         self.es = None
         self.isInitialized = False
-#        try:
-        from espeak import espeak 
-        self.es = espeak
-        self.isInitialized = True
-#        except:
-#            self.initialized = False
+        try:
+            from espeak import espeak 
+            self.es = espeak
+            self.isInitialized = True
+        except:
+            self.initialized = False
 
 
     def speak(self,text, queueable=True):
+        if not self.isInitialized:
+            return False
         if queueable == False: self.stop()
         self.es.synth(text)
+        return True
 
     def stop(self):
+        if not self.isInitialized:
+            return False
         self.es.cancel()
+        return True
 
     def clear_buffer(self):
-        pass
+        if not self.isInitialized:
+            return False
+        return True
+
+    def setVoice(self, voice):
+        if not self.isInitialized:
+            return False
+        return es.set_voice('de')
+
+    def setPitch(self, pitch):
+        if not self.isInitialized:
+            return False
+        return es.set_parameter(espeak.Parameter.Pitch, pitch) 
+
+    def setSpeed(self, speed):
+        if not self.isInitialized:
+            return False
+        return es.set_parameter(espeak.Parameter.Rate, speed) 
+