1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
|
// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#ifndef CONTENT_BROWSER_SPEECH_SPEECH_RECOGNIZER_IMPL_H_
#define CONTENT_BROWSER_SPEECH_SPEECH_RECOGNIZER_IMPL_H_
#include "base/basictypes.h"
#include "base/memory/ref_counted.h"
#include "base/memory/scoped_ptr.h"
#include "content/browser/speech/endpointer/endpointer.h"
#include "content/browser/speech/speech_recognition_engine.h"
#include "content/public/browser/speech_recognizer.h"
#include "content/public/common/speech_recognition_error.h"
#include "content/public/common/speech_recognition_result.h"
#include "media/audio/audio_input_controller.h"
#include "net/url_request/url_request_context_getter.h"
namespace content {
class SpeechRecognitionEventListener;
}
namespace media {
class AudioManager;
}
namespace speech {
// TODO(primiano) Next CL: Remove the Impl suffix and the exported
// /content/public/browser/speech_recognizer.h interface since this class should
// not be visible outside (currently we need it for speech input extension API).
// Handles speech recognition for a session (identified by |session_id|), taking
// care of audio capture, silence detection/endpointer and interaction with the
// SpeechRecognitionEngine.
class CONTENT_EXPORT SpeechRecognizerImpl
: public NON_EXPORTED_BASE(content::SpeechRecognizer),
public media::AudioInputController::EventHandler,
public NON_EXPORTED_BASE(SpeechRecognitionEngineDelegate) {
public:
static const int kAudioSampleRate;
static const ChannelLayout kChannelLayout;
static const int kNumBitsPerAudioSample;
static const int kNoSpeechTimeoutMs;
static const int kEndpointerEstimationTimeMs;
SpeechRecognizerImpl(
content::SpeechRecognitionEventListener* listener,
int session_id,
SpeechRecognitionEngine* engine);
// content::SpeechRecognizer methods.
virtual void StartRecognition() OVERRIDE;
virtual void AbortRecognition() OVERRIDE;
virtual void StopAudioCapture() OVERRIDE;
virtual bool IsActive() const OVERRIDE;
virtual bool IsCapturingAudio() const OVERRIDE;
const SpeechRecognitionEngine& recognition_engine() const;
protected:
virtual ~SpeechRecognizerImpl();
private:
friend class SpeechRecognizerImplTest;
enum FSMState {
STATE_IDLE = 0,
STATE_STARTING,
STATE_ESTIMATING_ENVIRONMENT,
STATE_WAITING_FOR_SPEECH,
STATE_RECOGNIZING,
STATE_WAITING_FINAL_RESULT,
STATE_MAX_VALUE = STATE_WAITING_FINAL_RESULT
};
enum FSMEvent {
EVENT_ABORT = 0,
EVENT_START,
EVENT_STOP_CAPTURE,
EVENT_AUDIO_DATA,
EVENT_ENGINE_RESULT,
EVENT_ENGINE_ERROR,
EVENT_AUDIO_ERROR,
EVENT_MAX_VALUE = EVENT_AUDIO_ERROR
};
struct FSMEventArgs {
explicit FSMEventArgs(FSMEvent event_value);
~FSMEventArgs();
FSMEvent event;
int audio_error_code;
scoped_refptr<AudioChunk> audio_data;
content::SpeechRecognitionResult engine_result;
content::SpeechRecognitionError engine_error;
};
// Entry point for pushing any new external event into the recognizer FSM.
void DispatchEvent(const FSMEventArgs& event_args);
// Defines the behavior of the recognizer FSM, selecting the appropriate
// transition according to the current state and event.
FSMState ExecuteTransitionAndGetNextState(const FSMEventArgs& args);
// Process a new audio chunk in the audio pipeline (endpointer, vumeter, etc).
void ProcessAudioPipeline(const AudioChunk& raw_audio);
// The methods below handle transitions of the recognizer FSM.
FSMState StartRecording(const FSMEventArgs& event_args);
FSMState StartRecognitionEngine(const FSMEventArgs& event_args);
FSMState WaitEnvironmentEstimationCompletion(const FSMEventArgs& event_args);
FSMState DetectUserSpeechOrTimeout(const FSMEventArgs& event_args);
FSMState StopCaptureAndWaitForResult(const FSMEventArgs& event_args);
FSMState ProcessIntermediateResult(const FSMEventArgs& event_args);
FSMState ProcessFinalResult(const FSMEventArgs& event_args);
FSMState Abort(const FSMEventArgs& event_args);
FSMState AbortWithError(const content::SpeechRecognitionError* error);
FSMState AbortWithError(const content::SpeechRecognitionError& error);
FSMState DetectEndOfSpeech(const FSMEventArgs& event_args);
FSMState DoNothing(const FSMEventArgs& event_args) const;
FSMState NotFeasible(const FSMEventArgs& event_args);
// Returns the time span of captured audio samples since the start of capture.
int GetElapsedTimeMs() const;
// Calculates the input volume to be displayed in the UI, triggering the
// OnAudioLevelsChange event accordingly.
void UpdateSignalAndNoiseLevels(const float& rms, bool clip_detected);
void CloseAudioControllerAsynchronously();
void SetAudioManagerForTesting(media::AudioManager* audio_manager);
// Callback called on IO thread by audio_controller->Close().
void OnAudioClosed(media::AudioInputController*);
// AudioInputController::EventHandler methods.
virtual void OnCreated(media::AudioInputController* controller) OVERRIDE {}
virtual void OnRecording(media::AudioInputController* controller) OVERRIDE {}
virtual void OnError(media::AudioInputController* controller,
int error_code) OVERRIDE;
virtual void OnData(media::AudioInputController* controller,
const uint8* data, uint32 size) OVERRIDE;
// SpeechRecognitionEngineDelegate methods.
virtual void OnSpeechRecognitionEngineResult(
const content::SpeechRecognitionResult& result) OVERRIDE;
virtual void OnSpeechRecognitionEngineError(
const content::SpeechRecognitionError& error) OVERRIDE;
content::SpeechRecognitionEventListener* listener_;
media::AudioManager* testing_audio_manager_;
scoped_ptr<SpeechRecognitionEngine> recognition_engine_;
Endpointer endpointer_;
scoped_refptr<media::AudioInputController> audio_controller_;
int session_id_;
int num_samples_recorded_;
float audio_level_;
bool is_dispatching_event_;
FSMState state_;
DISALLOW_COPY_AND_ASSIGN(SpeechRecognizerImpl);
};
} // namespace speech
#endif // CONTENT_BROWSER_SPEECH_SPEECH_RECOGNIZER_IMPL_H_
|
