Move VoiceEngineObserver into AudioSendStream so that we detect typing noises and return properly i…

talk/media/base/fakemediaengine.h

Index: talk/media/base/fakemediaengine.h
diff --git a/talk/media/base/fakemediaengine.h b/talk/media/base/fakemediaengine.h
index a6fa960dee9dbc4476f7220ee8a80b3142fefb46..ae1b9baaf12315615ce057d7ba443a5582db0e8a 100644
--- a/talk/media/base/fakemediaengine.h
+++ b/talk/media/base/fakemediaengine.h
@@ -733,7 +733,9 @@ class FakeVoiceEngine : public FakeBaseEngine {
   }
   bool Init(rtc::Thread* worker_thread) { return true; }
   void Terminate() {}
-  webrtc::VoiceEngine* GetVoE() { return nullptr; }
+  rtc::linked_ptr<webrtc::AudioState> GetAudioState() const {
+    return rtc::linked_ptr<webrtc::AudioState>();
+  }
   AudioOptions GetOptions() const {
     return options_;
   }