diff --git a/ipod-shuffle-4g.py b/ipod-shuffle-4g.py
index 8225a56..de440c7 100755
--- a/ipod-shuffle-4g.py
+++ b/ipod-shuffle-4g.py
@@ -97,7 +97,7 @@ def group_tracks_by_id3_template(tracks, template):
     return sorted(grouped_tracks_dict.items())
 
 class Text2Speech(object):
-    valid_tts = {'pico2wave': True, 'RHVoice': True, 'espeak': True, 'say': True}
+    valid_tts = {'pico2wave': True, 'RHVoice': True, 'espeak': True, 'say': True, 'gtts': True}
 
     @staticmethod
     def check_support():
@@ -124,6 +124,14 @@ def check_support():
         else:
             voiceoverAvailable = True
 
+        # Check for gtts-cli voiceover
+        # https://github.com/pndurette/gTTS
+        if not exec_exists_in_path("gtts-cli"):
+            Text2Speech.valid_tts['gtts-cli'] = False
+            print("Warning: gtts-cli not found, voicever won't be generated using it.")
+        else:
+            voiceoverAvailable = True
+
         # Check for Russian RHVoice voiceover
         if not exec_exists_in_path("RHVoice"):
             Text2Speech.valid_tts['RHVoice'] = False
@@ -156,6 +164,8 @@ def text2speech(out_wav_path, text):
                 return True
             elif Text2Speech.say(out_wav_path, text):
                 return True
+            elif Text2Speech.gtts(out_wav_path, text):
+                return True
             else:
                 return False
 
@@ -181,6 +191,19 @@ def say(out_wav_path, unicodetext):
         subprocess.call(["say", "-o", out_wav_path, '--data-format=LEI16', '--file-format=WAVE', '--', unicodetext])
         return True
 
+    @staticmethod
+    def gtts(out_wav_path, unicodetext):
+        if not Text2Speech.valid_tts['gtts']:
+            return False
+        tmp_mp3_file = tempfile.NamedTemporaryFile(suffix=".mp3", delete=False)
+        tmp_mp3_file.close()
+        tmp_wav_file = tempfile.NamedTemporaryFile(suffix=".wav", delete=False)
+        tmp_wav_file.close()
+        subprocess.call(["gtts-cli", "--lang", "en", unicodetext, "--output", tmp_mp3_file.name])
+        subprocess.call(["ffmpeg", "-y", "-loglevel", "error", "-hide_banner", "-nostats", "-i", tmp_mp3_file.name, tmp_wav_file.name])
+        subprocess.call(["mv", tmp_wav_file.name, out_wav_path])
+        return True
+
     @staticmethod
     def espeak(out_wav_path, unicodetext):
         if not Text2Speech.valid_tts['espeak']: