Build against stt

[jtojnar]

It was dropped by Mozilla so some of the responsible employees launched a startup.
Currently targeting 0.9.3, which is API compatible with the original Deepspeech.

[jtojnar]

Also builds with 1.3.0.

Verified that the example still works. Though had to use the official sample because autoaudiosrc clearly could not understand me:

--- a/examples/python/print_speech.py
+++ b/examples/python/print_speech.py
@@ -19,7 +19,7 @@ if __name__ == "__main__":
     Gst.init(None)
     loop = GLib.MainLoop()
 
-    pipeline = Gst.parse_launch("autoaudiosrc ! audioconvert ! audiorate ! audioresample ! deepspeech silence-length=20 ! fakesink")
+    pipeline = Gst.parse_launch("filesrc location=examples/python/8455-210777-0068.wav ! wavparse ! audioconvert ! audiorate ! audioresample ! deepspeech silence-length=20 ! fakesink")
     
     bus = pipeline.get_bus()
     bus.add_signal_watch()