diff --git a/Blather.py b/Blather.py
index 70f2626..b2d5737 100755
--- a/Blather.py
+++ b/Blather.py
@@ -6,7 +6,7 @@
 
 import sys
 import signal
-import gobject
+from gi.repository import GObject as gobject
 import os.path
 import subprocess #used to execute commands
 import shutil #used to copy plugins directory
@@ -46,7 +46,7 @@ def __init__(self, opts):
 		self.continuous_listen = opts.continuous
 		self.commands = {}
 		self.read_commands()
-		self.recognizer = Recognizer(lang_file, dic_file, opts.microphone )
+		self.recognizer = Recognizer(lang_file, dic_file, strings_file, opts.microphone)
 		self.recognizer.connect('finished',self.recognizer_finished)
 		self.matchTime = 0
 		self.keywordTimeLimit = opts.keytime #set to 0 to always speak the keyword
@@ -85,18 +85,16 @@ def read_commands(self):
 		self.commands = {}
 		self.keywords = []
 		for line in file_lines:
-				print line
 				#trim the white spaces
 				line = line.strip()
 				#if the line has length and the first char isn't a hash
 				if len(line) and line[0]!="#":
 						#this is a parsible line
 						(key,value) = line.split(":",1)
-						print key, value
+						print key, ":", value
 						#get the keyword out of the commands file
 						if value == "keyword" and key.strip().lower() not in self.keywords:
 							self.keywords.append(key.strip().lower())
-							continue
 						self.commands[key.strip().lower()] = value.strip()
 						strings.write( key.strip()+"\n")
 
@@ -119,7 +117,7 @@ def log_history(self,text):
 
 	def recognizer_finished(self, recognizer, text):
 		#split the words spoken into an array
-		t = text.lower()
+		t = text.strip().lower().replace("-", "")
 		textWords = t.split(" ")
 
 		#get the keys array for all commands
@@ -132,7 +130,7 @@ def recognizer_finished(self, recognizer, text):
 		biggestKey = ret['biggestKey']
 		biggestKeySet = ret['biggestKeySet']
 		biggestKeyCount = ret['biggestKeyCount']
-
+		
 		#find the match percentage
 		percentMatch = self.calculate_match_percentage(biggestKeySet, biggestKeyCount)
 
@@ -215,7 +213,6 @@ def load_resource(self,string):
 		return False
 
 	def search_for_matches(self, textWords):
-		#TODO: https://github.com/ajbogh/blather/issues/1
 		ret = {'biggestKey':'', 'biggestKeySet':{}, 'biggestKeyCount':0}
 		currentTime = time.time()
 		matchLimit = 1
diff --git a/GtkUI.py b/GtkUI.py
index 05a5cd2..2647e56 100755
--- a/GtkUI.py
+++ b/GtkUI.py
@@ -2,10 +2,12 @@
 # -- this code is licensed GPLv3
 # Copyright 2013 Jezra
 import sys
-import gobject
+from gi.repository import GObject as gobject
 #Gtk
-import pygtk
-import gtk
+import gi
+gi.require_version('Gtk', '3.0')
+from gi.repository import Gtk as gtk
+from gi.repository import Gdk as gdk
 
 class UI(gobject.GObject):
 	__gsignals__ = {
@@ -16,7 +18,7 @@ def __init__(self,args, continuous):
 		gobject.GObject.__init__(self)
 		self.continuous = continuous
 		#make a window
-		self.window = gtk.Window(gtk.WINDOW_TOPLEVEL)
+		self.window = gtk.Window(gtk.WindowType.TOPLEVEL)
 		self.window.set_default_size(325, 55)
 		self.window.connect("delete_event", self.delete_event)
 		#give the window a name
@@ -43,7 +45,7 @@ def __init__(self,args, continuous):
 		#create an accellerator group for this window
 		accel = gtk.AccelGroup()
 		#add the ctrl+q to quit
-		accel.connect_group(gtk.keysyms.q, gtk.gdk.CONTROL_MASK, gtk.ACCEL_VISIBLE, self.accel_quit )
+		accel.connect(gdk.KEY_q, gdk.ModifierType.CONTROL_MASK, gtk.AccelFlags.VISIBLE, self.accel_quit )
 		#lock the group
 		accel.lock()
 		#add the group to the window
@@ -91,5 +93,5 @@ def finished(self, text):
 		self.label.set_text(text)
 
 	def set_icon(self, icon):
-		gtk.window_set_default_icon_from_file(icon)
+		gtk.Window.set_icon_from_file(self.window, icon)
 
diff --git a/README_Ubuntu.md b/README_Ubuntu.md
new file mode 100644
index 0000000..1041d22
--- /dev/null
+++ b/README_Ubuntu.md
@@ -0,0 +1,8 @@
+Ubuntu with Python 2.7 installation steps:
+
+```
+sudo apt-get install --reinstall python-gi python-psutil python-gst-1.0 python-gst-1.0-dbg gstreamer1.0-pocketsphinx python-pyaudio swig python-pip python-dev build-essential
+
+pip install PocketSphinx
+```
+ 
\ No newline at end of file
diff --git a/Recognizer.py b/Recognizer.py
index 26ccd80..9f811cf 100755
--- a/Recognizer.py
+++ b/Recognizer.py
@@ -2,49 +2,68 @@
 # -- this code is licensed GPLv3
 # Copyright 2013 Jezra
 
-import pygst
-pygst.require('0.10')
-import gst
+import gi
 import os.path
-import gobject
+from gi.repository import GObject as gobject
+import speech_recognition as sr
+
+r = sr.Recognizer()
+m = sr.Microphone()
 
 #define some global variables
 this_dir = os.path.dirname( os.path.abspath(__file__) )
 
-
 class Recognizer(gobject.GObject):
-	__gsignals__ = {
-		'finished' : (gobject.SIGNAL_RUN_LAST, gobject.TYPE_NONE, (gobject.TYPE_STRING,))
-	}
-	def __init__(self, language_file, dictionary_file, src = None):
-		gobject.GObject.__init__(self)
-		self.commands = {}
-		if src:
-			audio_src = 'alsasrc device="hw:%d,0"' % (src)
-		else:
-			audio_src = 'autoaudiosrc'
-
-		#build the pipeline
-		cmd = audio_src+' ! audioconvert ! audioresample ! vader name=vad ! pocketsphinx name=asr ! appsink sync=false'
-		self.pipeline=gst.parse_launch( cmd )
-		#get the Auto Speech Recognition piece
-		asr=self.pipeline.get_by_name('asr')
-		asr.connect('result', self.result)
-		asr.set_property('lm', language_file)
-		asr.set_property('dict', dictionary_file)
-		asr.set_property('configured', True)
-		#get the Voice Activity DEtectoR
-		self.vad = self.pipeline.get_by_name('vad')
-		self.vad.set_property('auto-threshold',True)
-
-	def listen(self):
-		self.pipeline.set_state(gst.STATE_PLAYING)
-
-	def pause(self):
-		self.vad.set_property('silent', True)
-		self.pipeline.set_state(gst.STATE_PAUSED)
-
-	def result(self, asr, text, uttid):
-		#emit finished
-		self.emit("finished", text)
+    __gsignals__ = {
+        'finished' : (gobject.SIGNAL_RUN_LAST, gobject.TYPE_NONE, (gobject.TYPE_STRING,))
+    }
+    def __init__(self, language_file, dictionary_file, strings_file, src = None):
+        gobject.GObject.__init__(self)
+
+        self.keywords = self.read_keywords(strings_file);
+
+        print("A moment of silence, please...")
+        with m as source: r.adjust_for_ambient_noise(source)
+        print("Set minimum energy threshold to {}".format(r.energy_threshold))
+
+    def read_keywords(self, strings_file):
+        #read the.commands file
+        strings = open(strings_file)
+        keywords = []
+        for line in strings:
+            line = line.strip()
+            keywords.append([line, 1])
+        
+        #close the strings file
+        strings.close()
+        return keywords
+
+    def callback(self, recognizer, audio):
+        print("Got it! Now to recognize it...")
+        # recognize speech using Sphinx
+        try:
+            text = recognizer.recognize_sphinx(audio, "en-US", self.keywords, None)
+            print("Sphinx thinks you said " + text)
+            self.result(text)
+        except sr.UnknownValueError:
+            print("Sphinx could not understand audio")
+        except sr.RequestError as e:
+            print("Sphinx error; {0}".format(e))
+
+    def listen(self):
+        print("Say something!")
+        self.stop_listening = r.listen_in_background(m, self.callback)
+        return
+
+    def pause(self):
+        try:
+            self.stop_listening()
+            print("No longer listening")
+        except Exception:
+            pass        
+        return
+
+    def result(self, text):
+        #emit finished
+        self.emit("finished", text)
 
diff --git a/language_updater.sh b/language_updater.sh
index ec5c868..e37fe2e 100755
--- a/language_updater.sh
+++ b/language_updater.sh
@@ -25,6 +25,8 @@ curl -C - -O $(cat $tempfile).dic
 curl -C - -O $(cat $tempfile).lm
 
 # mv em to the right name/place
+mkdir --parents $langdir
+
 mv *.dic $langdir/dic
 mv *.lm $langdir/lm
 
diff --git a/sr_test.py b/sr_test.py
new file mode 100644
index 0000000..9c0719c
--- /dev/null
+++ b/sr_test.py
@@ -0,0 +1,29 @@
+#!/usr/bin/env python2
+import speech_recognition as sr
+
+r = sr.Recognizer()
+m = sr.Microphone()
+
+try:
+    print("A moment of silence, please...")
+    with m as source: r.adjust_for_ambient_noise(source)
+    print("Set minimum energy threshold to {}".format(r.energy_threshold))
+    while True:
+        print("Say something!")
+        with m as source: audio = r.listen(source)
+        print("Got it! Now to recognize it...")
+        try:
+            # recognize speech using Google Speech Recognition
+            value = r.recognize_google(audio)
+
+            # we need some special handling here to correctly print unicode characters to standard output
+            if str is bytes:  # this version of Python uses bytes for strings (Python 2)
+                print(u"You said {}".format(value).encode("utf-8"))
+            else:  # this version of Python uses unicode for strings (Python 3+)
+                print("You said {}".format(value))
+        except sr.UnknownValueError:
+            print("Oops! Didn't catch that")
+        except sr.RequestError as e:
+            print("Uh oh! Couldn't request results from Google Speech Recognition service; {0}".format(e))
+except KeyboardInterrupt:
+    pass
\ No newline at end of file