Binding media stream audio track to speech recognition [browser]

content/browser/speech/speech_recognizer_impl.h

Index: content/browser/speech/speech_recognizer_impl.h
diff --git a/content/browser/speech/speech_recognizer_impl.h b/content/browser/speech/speech_recognizer_impl.h
index bd869ee570799fb86ad45c16b463c96e7106e4c9..31af59dbeab6ad924f5d3dd1c9f67ba07368ba2a 100644
--- a/content/browser/speech/speech_recognizer_impl.h
+++ b/content/browser/speech/speech_recognizer_impl.h
@@ -7,6 +7,7 @@
 
 #include "base/basictypes.h"
 #include "base/memory/scoped_ptr.h"
+#include "base/process/process_handle.h"
 #include "content/browser/speech/endpointer/endpointer.h"
 #include "content/browser/speech/speech_recognition_engine.h"
 #include "content/browser/speech/speech_recognizer.h"
@@ -19,10 +20,12 @@
 namespace media {
 class AudioBus;
 class AudioManager;
+class AudioParameters;
 }
 
 namespace content {
 
+class SpeechRecognitionAudioReceiver;
 class SpeechRecognitionEventListener;
 
 // Handles speech recognition for a session (identified by |session_id|), taking
@@ -40,11 +43,11 @@ class CONTENT_EXPORT SpeechRecognizerImpl
   static const int kEndpointerEstimationTimeMs;
 
   static void SetAudioManagerForTesting(media::AudioManager* audio_manager);
-
-  SpeechRecognizerImpl(SpeechRecognitionEventListener* listener,
-                       int session_id,
-                       bool continuous,
-                       bool provisional_results,
+  // TODO(burnik): abstract out the process_handle. Pass differently.
+  SpeechRecognizerImpl(SpeechRecognitionEventListener* listener, int session_id,

[burnik, 2014/10/15 14:52:27]

Will move each param on it's own line.

+                       bool continuous, bool provisional_results,
+                       bool using_audio_receiver,
+                       base::ProcessHandle process_handle,
                        SpeechRecognitionEngine* engine);
 
   virtual void StartRecognition(const std::string& device_id) OVERRIDE;
@@ -116,6 +119,9 @@ class CONTENT_EXPORT SpeechRecognizerImpl
   FSMState DoNothing(const FSMEventArgs& event_args) const;
   FSMState NotFeasible(const FSMEventArgs& event_args);
 
+  // Called from |StartRecording()| when using |audio_receiver_|.
+  bool StartRecordingUsingAudioReceiver(const media::AudioParameters& params);
+
   // Returns the time span of captured audio samples since the start of capture.
   int GetElapsedTimeMs() const;
 
@@ -154,6 +160,8 @@ class CONTENT_EXPORT SpeechRecognizerImpl
   float audio_level_;
   bool is_dispatching_event_;
   bool provisional_results_;
+  bool using_audio_receiver_;
+  base::ProcessHandle peer_process_handle_;
   FSMState state_;
   std::string device_id_;
 
@@ -163,6 +171,9 @@ class CONTENT_EXPORT SpeechRecognizerImpl
   // output format.
   scoped_ptr<SpeechRecognizerImpl::OnDataConverter> audio_converter_;
 
+  // Used when getting audio data from the renderer.
+  scoped_ptr<SpeechRecognitionAudioReceiver> audio_receiver_;
+
   DISALLOW_COPY_AND_ASSIGN(SpeechRecognizerImpl);
 };