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 b047174401685c7e57d44c33d30335fa4f3d880a..86fa152b129950486e23440cf53930f47acc2f8a 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"
@@ -16,13 +17,19 @@
 #include "media/audio/audio_logging.h"
 #include "net/url_request/url_request_context_getter.h"
 
+namespace base {
+  class Thread;

[tommi (sloooow) - chröme, 2014/10/16 21:20:23]

no indent

[burnik, 2014/10/17 10:48:02]

Done.

+}
+
 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
@@ -45,6 +52,8 @@ class CONTENT_EXPORT SpeechRecognizerImpl
                        int session_id,
                        bool continuous,
                        bool provisional_results,
+                       bool using_audio_receiver,
+                       const base::ProcessHandle process_handle,
                        SpeechRecognitionEngine* engine);
 
   virtual void StartRecognition(const std::string& device_id) override;
@@ -116,6 +125,12 @@ 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);
+
+  // Called when the session ends due to error or when being stopped.
+  void ResetAudioReceiver();
+
   // Returns the time span of captured audio samples since the start of capture.
   int GetElapsedTimeMs() const;
 
@@ -154,6 +169,8 @@ 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 +180,10 @@ class CONTENT_EXPORT SpeechRecognizerImpl
   // output format.
   scoped_ptr<SpeechRecognizerImpl::OnDataConverter> audio_converter_;
 
+  // Used when getting audio data from the renderer.
+  scoped_ptr<base::Thread> audio_thread_;
+  scoped_ptr<SpeechRecognitionAudioReceiver> audio_receiver_;
+
   DISALLOW_COPY_AND_ASSIGN(SpeechRecognizerImpl);
 };