summaryrefslogtreecommitdiffstats
path: root/content/browser/speech/speech_recognition_manager_impl.cc
blob: 8f9c3e120ea5ada0c31ca52dd92b9a5534581857 (plain)
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
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
// 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.

#include "content/browser/speech/speech_recognition_manager_impl.h"

#include "base/bind.h"
#include "content/browser/browser_main_loop.h"
#include "content/browser/renderer_host/media/media_stream_manager.h"
#include "content/browser/speech/google_one_shot_remote_engine.h"
#include "content/browser/speech/google_streaming_remote_engine.h"
#include "content/browser/speech/speech_recognition_engine.h"
#include "content/browser/speech/speech_recognizer.h"
#include "content/public/browser/browser_thread.h"
#include "content/public/browser/content_browser_client.h"
#include "content/public/browser/resource_context.h"
#include "content/public/browser/speech_recognition_event_listener.h"
#include "content/public/browser/speech_recognition_manager_delegate.h"
#include "content/public/browser/speech_recognition_session_config.h"
#include "content/public/browser/speech_recognition_session_context.h"
#include "content/public/common/speech_recognition_error.h"
#include "content/public/common/speech_recognition_result.h"
#include "media/audio/audio_manager.h"

using base::Callback;

namespace content {

SpeechRecognitionManager* SpeechRecognitionManager::manager_for_tests_;

namespace {

SpeechRecognitionManagerImpl* g_speech_recognition_manager_impl;

void ShowAudioInputSettingsOnFileThread() {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::FILE));
  BrowserMainLoop::GetAudioManager()->ShowAudioInputSettings();
}

}  // namespace

SpeechRecognitionManager* SpeechRecognitionManager::GetInstance() {
  if (manager_for_tests_)
    return manager_for_tests_;
  return SpeechRecognitionManagerImpl::GetInstance();
}

void SpeechRecognitionManager::SetManagerForTests(
    SpeechRecognitionManager* manager) {
  manager_for_tests_ = manager;
}

SpeechRecognitionManagerImpl* SpeechRecognitionManagerImpl::GetInstance() {
  return g_speech_recognition_manager_impl;
}

SpeechRecognitionManagerImpl::SpeechRecognitionManagerImpl()
    : primary_session_id_(kSessionIDInvalid),
      last_session_id_(kSessionIDInvalid),
      is_dispatching_event_(false),
      delegate_(GetContentClient()->browser()->
                    GetSpeechRecognitionManagerDelegate()),
      ALLOW_THIS_IN_INITIALIZER_LIST(weak_factory_(this)) {
  DCHECK(!g_speech_recognition_manager_impl);
  g_speech_recognition_manager_impl = this;
}

SpeechRecognitionManagerImpl::~SpeechRecognitionManagerImpl() {
  DCHECK(g_speech_recognition_manager_impl);
  g_speech_recognition_manager_impl = NULL;
  // Recognition sessions will be aborted by the corresponding destructors.
  sessions_.clear();
}

int SpeechRecognitionManagerImpl::CreateSession(
    const SpeechRecognitionSessionConfig& config) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));

  const int session_id = GetNextSessionID();
  DCHECK(!SessionExists(session_id));
  // Set-up the new session.
  Session& session = sessions_[session_id];
  session.id = session_id;
  session.config = config;
  session.context = config.initial_context;

  std::string hardware_info;
  bool can_report_metrics = false;
  if (delegate_.get())
    delegate_->GetDiagnosticInformation(&can_report_metrics, &hardware_info);

  SpeechRecognitionEngineConfig remote_engine_config;
  remote_engine_config.language = config.language;
  remote_engine_config.grammars = config.grammars;
  remote_engine_config.audio_sample_rate = SpeechRecognizer::kAudioSampleRate;
  remote_engine_config.audio_num_bits_per_sample =
     SpeechRecognizer::kNumBitsPerAudioSample;
  remote_engine_config.filter_profanities = config.filter_profanities;
  remote_engine_config.continuous = config.continuous;
  remote_engine_config.interim_results = config.interim_results;
  remote_engine_config.max_hypotheses = config.max_hypotheses;
  remote_engine_config.hardware_info = hardware_info;
  remote_engine_config.origin_url = can_report_metrics ? config.origin_url : "";

  SpeechRecognitionEngine* google_remote_engine;
  if (config.is_legacy_api) {
    google_remote_engine =
        new GoogleOneShotRemoteEngine(config.url_request_context_getter);
  } else {
    google_remote_engine =
        new GoogleStreamingRemoteEngine(config.url_request_context_getter);
  }

  google_remote_engine->SetConfig(remote_engine_config);

  // The legacy api cannot use continuous mode.
  DCHECK(!config.is_legacy_api || !config.continuous);

  session.recognizer = new SpeechRecognizer(
      this,
      session_id,
      !config.continuous,
      google_remote_engine);
  return session_id;
}

void SpeechRecognitionManagerImpl::StartSession(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  // If there is another active session, abort that.
  if (primary_session_id_ != kSessionIDInvalid &&
      primary_session_id_ != session_id) {
    AbortSession(primary_session_id_);
  }

  primary_session_id_ = session_id;

  if (delegate_.get()) {
    delegate_->CheckRecognitionIsAllowed(
        session_id,
        base::Bind(&SpeechRecognitionManagerImpl::RecognitionAllowedCallback,
                   weak_factory_.GetWeakPtr(),
                   session_id));
  }
}

void SpeechRecognitionManagerImpl::RecognitionAllowedCallback(int session_id,
                                                              bool ask_user,
                                                              bool is_allowed) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

#if defined(OS_IOS)
  // On iOS, voice search can only be initiated by clear user action and thus
  // it is always allowed.
  DCHECK(!ask_user && is_allowed);
#else
  if (ask_user) {
    SessionsTable::iterator iter = sessions_.find(session_id);
    DCHECK(iter != sessions_.end());
    SpeechRecognitionSessionContext& context = iter->second.context;
    context.label =
        BrowserMainLoop::GetMediaStreamManager()->MakeMediaAccessRequest(
            context.render_process_id,
            context.render_view_id,
            StreamOptions(MEDIA_DEVICE_AUDIO_CAPTURE, MEDIA_NO_SERVICE),
            GURL(context.context_name),
            base::Bind(
                &SpeechRecognitionManagerImpl::MediaRequestPermissionCallback,
                weak_factory_.GetWeakPtr()));

    return;
  }
#endif  // defined(OS_IOS)

  if (is_allowed) {
    MessageLoop::current()->PostTask(FROM_HERE,
        base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                   weak_factory_.GetWeakPtr(), session_id, EVENT_START));
  } else {
    OnRecognitionError(session_id, SpeechRecognitionError(
        SPEECH_RECOGNITION_ERROR_NOT_ALLOWED));
    MessageLoop::current()->PostTask(FROM_HERE,
        base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                   weak_factory_.GetWeakPtr(), session_id, EVENT_ABORT));
  }
}

void SpeechRecognitionManagerImpl::MediaRequestPermissionCallback(
    const std::string& label, const MediaStreamDevices& devices) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  for (SessionsTable::iterator iter = sessions_.begin();
       iter != sessions_.end(); ++iter) {
    if (iter->second.context.label == label) {
      bool is_allowed = false;
      if (!devices.empty()) {
        // Copy the approved devices array to the context for UI indication.
        iter->second.context.devices = devices;
        is_allowed = true;
      }

      // Clear the label to indicate the request has been done.
      iter->second.context.label.clear();

      // Notify the recognition about the request result.
      RecognitionAllowedCallback(iter->first, false, is_allowed);
      break;
    }
  }
}

void SpeechRecognitionManagerImpl::AbortSession(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

#if !defined(OS_IOS)
  const SpeechRecognitionSessionContext& context =
      GetSessionContext(session_id);
  if (!context.label.empty())
    BrowserMainLoop::GetMediaStreamManager()->CancelRequest(context.label);
#endif  // !defined(OS_IOS)

  MessageLoop::current()->PostTask(FROM_HERE,
      base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                 weak_factory_.GetWeakPtr(), session_id, EVENT_ABORT));
}

void SpeechRecognitionManagerImpl::StopAudioCaptureForSession(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

#if !defined(OS_IOS)
  const SpeechRecognitionSessionContext& context =
      GetSessionContext(session_id);
  if (!context.label.empty())
    BrowserMainLoop::GetMediaStreamManager()->CancelRequest(context.label);
#endif  // !defined(OS_IOS)

  MessageLoop::current()->PostTask(FROM_HERE,
      base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                 weak_factory_.GetWeakPtr(), session_id, EVENT_STOP_CAPTURE));
}

// Here begins the SpeechRecognitionEventListener interface implementation,
// which will simply relay the events to the proper listener registered for the
// particular session (most likely InputTagSpeechDispatcherHost) and to the
// catch-all listener provided by the delegate (if any).

void SpeechRecognitionManagerImpl::OnRecognitionStart(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

#if !defined(OS_IOS)
  const SpeechRecognitionSessionContext& context =
      GetSessionContext(session_id);
  if (!context.devices.empty()) {
    // Notify the UI the devices are being used.
    BrowserMainLoop::GetMediaStreamManager()->NotifyUIDevicesOpened(
        context.label, context.render_process_id, context.render_view_id,
        context.devices);
  }
#endif  // !defined(OS_IOS)

  DCHECK_EQ(primary_session_id_, session_id);
  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnRecognitionStart(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnRecognitionStart(session_id);
}

void SpeechRecognitionManagerImpl::OnAudioStart(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  DCHECK_EQ(primary_session_id_, session_id);
  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnAudioStart(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnAudioStart(session_id);
}

void SpeechRecognitionManagerImpl::OnEnvironmentEstimationComplete(
    int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  DCHECK_EQ(primary_session_id_, session_id);
  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnEnvironmentEstimationComplete(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnEnvironmentEstimationComplete(session_id);
}

void SpeechRecognitionManagerImpl::OnSoundStart(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  DCHECK_EQ(primary_session_id_, session_id);
  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnSoundStart(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnSoundStart(session_id);
}

void SpeechRecognitionManagerImpl::OnSoundEnd(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnSoundEnd(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnSoundEnd(session_id);
}

void SpeechRecognitionManagerImpl::OnAudioEnd(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnAudioEnd(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnAudioEnd(session_id);
  MessageLoop::current()->PostTask(FROM_HERE,
      base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                 weak_factory_.GetWeakPtr(), session_id, EVENT_AUDIO_ENDED));
}

void SpeechRecognitionManagerImpl::OnRecognitionResults(
    int session_id, const SpeechRecognitionResults& results) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnRecognitionResults(session_id, results);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnRecognitionResults(session_id, results);
}

void SpeechRecognitionManagerImpl::OnRecognitionError(
    int session_id, const SpeechRecognitionError& error) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnRecognitionError(session_id, error);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnRecognitionError(session_id, error);
}

void SpeechRecognitionManagerImpl::OnAudioLevelsChange(
    int session_id, float volume, float noise_volume) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnAudioLevelsChange(session_id, volume, noise_volume);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnAudioLevelsChange(session_id, volume, noise_volume);
}

void SpeechRecognitionManagerImpl::OnRecognitionEnd(int session_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  if (!SessionExists(session_id))
    return;
#if !defined(OS_IOS)
  const SpeechRecognitionSessionContext& context =
        GetSessionContext(session_id);
  if (!context.devices.empty()) {
    // Notify the UI the devices has been closed.
     BrowserMainLoop::GetMediaStreamManager()->NotifyUIDevicesClosed(
         context.render_process_id, context.render_view_id, context.devices);
  }
#endif  // !defined(OS_IOS)

  if (SpeechRecognitionEventListener* delegate_listener = GetDelegateListener())
    delegate_listener->OnRecognitionEnd(session_id);
  if (SpeechRecognitionEventListener* listener = GetListener(session_id))
    listener->OnRecognitionEnd(session_id);
  MessageLoop::current()->PostTask(FROM_HERE,
      base::Bind(&SpeechRecognitionManagerImpl::DispatchEvent,
                 weak_factory_.GetWeakPtr(),
                 session_id,
                 EVENT_RECOGNITION_ENDED));
}

int SpeechRecognitionManagerImpl::GetSession(
    int render_process_id, int render_view_id, int request_id) const {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  SessionsTable::const_iterator iter;
  for(iter = sessions_.begin(); iter != sessions_.end(); ++iter) {
    const int session_id = iter->first;
    const SpeechRecognitionSessionContext& context = iter->second.context;
    if (context.render_process_id == render_process_id &&
        context.render_view_id == render_view_id &&
        context.request_id == request_id) {
      return session_id;
    }
  }
  return kSessionIDInvalid;
}

SpeechRecognitionSessionContext
SpeechRecognitionManagerImpl::GetSessionContext(int session_id) const {
  return GetSession(session_id).context;
}

void SpeechRecognitionManagerImpl::AbortAllSessionsForListener(
    SpeechRecognitionEventListener* listener) {
  // This method gracefully destroys sessions for the listener. However, since
  // the listener itself is likely to be destroyed after this call, we avoid
  // dispatching further events to it, marking the |listener_is_active| flag.
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  for (SessionsTable::iterator it = sessions_.begin(); it != sessions_.end();
       ++it) {
    Session& session = it->second;
    if (session.config.event_listener == listener) {
      AbortSession(session.id);
      session.listener_is_active = false;
    }
  }
}

void SpeechRecognitionManagerImpl::AbortAllSessionsForRenderView(
    int render_process_id,
    int render_view_id) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  for (SessionsTable::iterator it = sessions_.begin(); it != sessions_.end();
       ++it) {
    Session& session = it->second;
    if (session.context.render_process_id == render_process_id &&
        session.context.render_view_id == render_view_id) {
      AbortSession(session.id);
    }
  }
}

// -----------------------  Core FSM implementation ---------------------------
void SpeechRecognitionManagerImpl::DispatchEvent(int session_id,
                                                 FSMEvent event) {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));

  // There are some corner cases in which the session might be deleted (due to
  // an EndRecognition event) between a request (e.g. Abort) and its dispatch.
  if (!SessionExists(session_id))
    return;

  const Session& session = GetSession(session_id);
  FSMState session_state = GetSessionState(session_id);
  DCHECK_LE(session_state, SESSION_STATE_MAX_VALUE);
  DCHECK_LE(event, EVENT_MAX_VALUE);

  // Event dispatching must be sequential, otherwise it will break all the rules
  // and the assumptions of the finite state automata model.
  DCHECK(!is_dispatching_event_);
  is_dispatching_event_ = true;
  ExecuteTransitionAndGetNextState(session, session_state, event);
  is_dispatching_event_ = false;
}

// This FSM handles the evolution of each session, from the viewpoint of the
// interaction with the user (that may be either the browser end-user which
// interacts with UI bubbles, or JS developer intracting with JS methods).
// All the events received by the SpeechRecognizer instances (one for each
// session) are always routed to the SpeechRecognitionEventListener(s)
// regardless the choices taken in this FSM.
void SpeechRecognitionManagerImpl::ExecuteTransitionAndGetNextState(
    const Session& session, FSMState session_state, FSMEvent event) {
  // Note: since we're not tracking the state of the recognizer object, rather
  // we're directly retrieving it (through GetSessionState), we see its events
  // (that are AUDIO_ENDED and RECOGNITION_ENDED) after its state evolution
  // (e.g., when we receive the AUDIO_ENDED event, the recognizer has just
  // completed the transition from CAPTURING_AUDIO to WAITING_FOR_RESULT, thus
  // we perceive the AUDIO_ENDED event in WAITING_FOR_RESULT).
  // This makes the code below a bit tricky but avoids a lot of code for
  // tracking and reconstructing asynchronously the state of the recognizer.
  switch (session_state) {
    case SESSION_STATE_IDLE:
      switch (event) {
        case EVENT_START:
          return SessionStart(session);
        case EVENT_ABORT:
          return SessionAbort(session);
        case EVENT_RECOGNITION_ENDED:
          return SessionDelete(session);
        case EVENT_STOP_CAPTURE:
          return SessionStopAudioCapture(session);
        case EVENT_AUDIO_ENDED:
          return;
      }
      break;
    case SESSION_STATE_CAPTURING_AUDIO:
      switch (event) {
        case EVENT_STOP_CAPTURE:
          return SessionStopAudioCapture(session);
        case EVENT_ABORT:
          return SessionAbort(session);
        case EVENT_START:
          return;
        case EVENT_AUDIO_ENDED:
        case EVENT_RECOGNITION_ENDED:
          return NotFeasible(session, event);
      }
      break;
    case SESSION_STATE_WAITING_FOR_RESULT:
      switch (event) {
        case EVENT_ABORT:
          return SessionAbort(session);
        case EVENT_AUDIO_ENDED:
          return ResetCapturingSessionId(session);
        case EVENT_START:
        case EVENT_STOP_CAPTURE:
          return;
        case EVENT_RECOGNITION_ENDED:
          return NotFeasible(session, event);
      }
      break;
  }
  return NotFeasible(session, event);
}

SpeechRecognitionManagerImpl::FSMState
SpeechRecognitionManagerImpl::GetSessionState(int session_id) const {
  const Session& session = GetSession(session_id);
  if (!session.recognizer.get() || !session.recognizer->IsActive())
    return SESSION_STATE_IDLE;
  if (session.recognizer->IsCapturingAudio())
    return SESSION_STATE_CAPTURING_AUDIO;
  return SESSION_STATE_WAITING_FOR_RESULT;
}

// ----------- Contract for all the FSM evolution functions below -------------
//  - Are guaranteed to be executed in the IO thread;
//  - Are guaranteed to be not reentrant (themselves and each other);

void SpeechRecognitionManagerImpl::SessionStart(const Session& session) {
  DCHECK_EQ(primary_session_id_, session.id);
  session.recognizer->StartRecognition();
}

void SpeechRecognitionManagerImpl::SessionAbort(const Session& session) {
  if (primary_session_id_ == session.id)
    primary_session_id_ = kSessionIDInvalid;
  DCHECK(session.recognizer.get());
  session.recognizer->AbortRecognition();
}

void SpeechRecognitionManagerImpl::SessionStopAudioCapture(
    const Session& session) {
  DCHECK(session.recognizer.get());
  session.recognizer->StopAudioCapture();
}

void SpeechRecognitionManagerImpl::ResetCapturingSessionId(
    const Session& session) {
  DCHECK_EQ(primary_session_id_, session.id);
  primary_session_id_ = kSessionIDInvalid;
}

void SpeechRecognitionManagerImpl::SessionDelete(const Session& session) {
  DCHECK(session.recognizer == NULL || !session.recognizer->IsActive());
  if (primary_session_id_ == session.id)
    primary_session_id_ = kSessionIDInvalid;
  sessions_.erase(session.id);
}

void SpeechRecognitionManagerImpl::NotFeasible(const Session& session,
                                               FSMEvent event) {
  NOTREACHED() << "Unfeasible event " << event
               << " in state " << GetSessionState(session.id)
               << " for session " << session.id;
}

int SpeechRecognitionManagerImpl::GetNextSessionID() {
  ++last_session_id_;
  // Deal with wrapping of last_session_id_. (How civilized).
  if (last_session_id_ <= 0)
    last_session_id_ = 1;
  return last_session_id_;
}

bool SpeechRecognitionManagerImpl::SessionExists(int session_id) const {
  return sessions_.find(session_id) != sessions_.end();
}

const SpeechRecognitionManagerImpl::Session&
SpeechRecognitionManagerImpl::GetSession(int session_id) const {
  DCHECK(BrowserThread::CurrentlyOn(BrowserThread::IO));
  SessionsTable::const_iterator iter = sessions_.find(session_id);
  DCHECK(iter != sessions_.end());
  return iter->second;
}

SpeechRecognitionEventListener* SpeechRecognitionManagerImpl::GetListener(
    int session_id) const {
  const Session& session = GetSession(session_id);
  return session.listener_is_active ? session.config.event_listener : NULL;
}

SpeechRecognitionEventListener*
SpeechRecognitionManagerImpl::GetDelegateListener() const {
  return delegate_.get() ? delegate_->GetEventListener() : NULL;
}

const SpeechRecognitionSessionConfig&
SpeechRecognitionManagerImpl::GetSessionConfig(int session_id) const {
  return GetSession(session_id).config;
}

bool SpeechRecognitionManagerImpl::HasAudioInputDevices() {
  return BrowserMainLoop::GetAudioManager()->HasAudioInputDevices();
}

bool SpeechRecognitionManagerImpl::IsCapturingAudio() {
  return BrowserMainLoop::GetAudioManager()->IsRecordingInProcess();
}

string16 SpeechRecognitionManagerImpl::GetAudioInputDeviceModel() {
  return BrowserMainLoop::GetAudioManager()->GetAudioInputDeviceModel();
}

void SpeechRecognitionManagerImpl::ShowAudioInputSettings() {
  // Since AudioManager::ShowAudioInputSettings can potentially launch external
  // processes, do that in the FILE thread to not block the calling threads.
  BrowserThread::PostTask(BrowserThread::FILE, FROM_HERE,
                          base::Bind(&ShowAudioInputSettingsOnFileThread));
}

SpeechRecognitionManagerImpl::Session::Session()
  : id(kSessionIDInvalid),
    listener_is_active(true) {
}

SpeechRecognitionManagerImpl::Session::~Session() {
}

}  // namespace content