Home | History | Annotate | Download | only in speech
      1 /*
      2  * Copyright (C) 2013 Apple Inc. All rights reserved.
      3  *
      4  * Redistribution and use in source and binary forms, with or without
      5  * modification, are permitted provided that the following conditions
      6  * are met:
      7  * 1. Redistributions of source code must retain the above copyright
      8  *    notice, this list of conditions and the following disclaimer.
      9  * 2. Redistributions in binary form must reproduce the above copyright
     10  *    notice, this list of conditions and the following disclaimer in the
     11  *    documentation and/or other materials provided with the distribution.
     12  *
     13  * THIS SOFTWARE IS PROVIDED BY APPLE COMPUTER, INC. ``AS IS'' AND ANY
     14  * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
     15  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
     16  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE COMPUTER, INC. OR
     17  * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
     18  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
     19  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
     20  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
     21  * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
     22  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
     23  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
     24  */
     25 
     26 #include "config.h"
     27 #include "modules/speech/SpeechSynthesis.h"
     28 
     29 #include "bindings/core/v8/ExceptionState.h"
     30 #include "core/dom/ExecutionContext.h"
     31 #include "modules/speech/SpeechSynthesisEvent.h"
     32 #include "platform/speech/PlatformSpeechSynthesisVoice.h"
     33 #include "wtf/CurrentTime.h"
     34 
     35 namespace blink {
     36 
     37 SpeechSynthesis* SpeechSynthesis::create(ExecutionContext* context)
     38 {
     39     return adoptRefCountedGarbageCollectedWillBeNoop(new SpeechSynthesis(context));
     40 }
     41 
     42 SpeechSynthesis::SpeechSynthesis(ExecutionContext* context)
     43     : ContextLifecycleObserver(context)
     44     , m_platformSpeechSynthesizer(PlatformSpeechSynthesizer::create(this))
     45     , m_isPaused(false)
     46 {
     47 }
     48 
     49 void SpeechSynthesis::setPlatformSynthesizer(PlatformSpeechSynthesizer* synthesizer)
     50 {
     51     m_platformSpeechSynthesizer = synthesizer;
     52 }
     53 
     54 ExecutionContext* SpeechSynthesis::executionContext() const
     55 {
     56     return ContextLifecycleObserver::executionContext();
     57 }
     58 
     59 void SpeechSynthesis::voicesDidChange()
     60 {
     61     m_voiceList.clear();
     62     if (executionContext() && !executionContext()->activeDOMObjectsAreStopped())
     63         dispatchEvent(Event::create(EventTypeNames::voiceschanged));
     64 }
     65 
     66 const HeapVector<Member<SpeechSynthesisVoice> >& SpeechSynthesis::getVoices()
     67 {
     68     if (m_voiceList.size())
     69         return m_voiceList;
     70 
     71     // If the voiceList is empty, that's the cue to get the voices from the platform again.
     72     const HeapVector<Member<PlatformSpeechSynthesisVoice> >& platformVoices = m_platformSpeechSynthesizer->voiceList();
     73     size_t voiceCount = platformVoices.size();
     74     for (size_t k = 0; k < voiceCount; k++)
     75         m_voiceList.append(SpeechSynthesisVoice::create(platformVoices[k].get()));
     76 
     77     return m_voiceList;
     78 }
     79 
     80 bool SpeechSynthesis::speaking() const
     81 {
     82     // If we have a current speech utterance, then that means we're assumed to be in a speaking state.
     83     // This state is independent of whether the utterance happens to be paused.
     84     return currentSpeechUtterance();
     85 }
     86 
     87 bool SpeechSynthesis::pending() const
     88 {
     89     // This is true if there are any utterances that have not started.
     90     // That means there will be more than one in the queue.
     91     return m_utteranceQueue.size() > 1;
     92 }
     93 
     94 bool SpeechSynthesis::paused() const
     95 {
     96     return m_isPaused;
     97 }
     98 
     99 void SpeechSynthesis::startSpeakingImmediately()
    100 {
    101     SpeechSynthesisUtterance* utterance = currentSpeechUtterance();
    102     ASSERT(utterance);
    103 
    104     utterance->setStartTime(monotonicallyIncreasingTime());
    105     m_isPaused = false;
    106     m_platformSpeechSynthesizer->speak(utterance->platformUtterance());
    107 }
    108 
    109 void SpeechSynthesis::speak(SpeechSynthesisUtterance* utterance, ExceptionState& exceptionState)
    110 {
    111     if (!utterance) {
    112         exceptionState.throwTypeError("Invalid utterance argument");
    113         return;
    114     }
    115 
    116     m_utteranceQueue.append(utterance);
    117 
    118     // If the queue was empty, speak this immediately.
    119     if (m_utteranceQueue.size() == 1)
    120         startSpeakingImmediately();
    121 }
    122 
    123 void SpeechSynthesis::cancel()
    124 {
    125     // Remove all the items from the utterance queue. The platform
    126     // may still have references to some of these utterances and may
    127     // fire events on them asynchronously.
    128     m_utteranceQueue.clear();
    129     m_platformSpeechSynthesizer->cancel();
    130 }
    131 
    132 void SpeechSynthesis::pause()
    133 {
    134     if (!m_isPaused)
    135         m_platformSpeechSynthesizer->pause();
    136 }
    137 
    138 void SpeechSynthesis::resume()
    139 {
    140     if (!currentSpeechUtterance())
    141         return;
    142     m_platformSpeechSynthesizer->resume();
    143 }
    144 
    145 void SpeechSynthesis::fireEvent(const AtomicString& type, SpeechSynthesisUtterance* utterance, unsigned long charIndex, const String& name)
    146 {
    147     if (executionContext() && !executionContext()->activeDOMObjectsAreStopped())
    148         utterance->dispatchEvent(SpeechSynthesisEvent::create(type, charIndex, (currentTime() - utterance->startTime()), name));
    149 }
    150 
    151 void SpeechSynthesis::handleSpeakingCompleted(SpeechSynthesisUtterance* utterance, bool errorOccurred)
    152 {
    153     ASSERT(utterance);
    154 
    155     bool shouldStartSpeaking = false;
    156     // If the utterance that completed was the one we're currently speaking,
    157     // remove it from the queue and start speaking the next one.
    158     if (utterance == currentSpeechUtterance()) {
    159         m_utteranceQueue.removeFirst();
    160         shouldStartSpeaking = !!m_utteranceQueue.size();
    161     }
    162 
    163     // Always fire the event, because the platform may have asynchronously
    164     // sent an event on an utterance before it got the message that we
    165     // canceled it, and we should always report to the user what actually
    166     // happened.
    167     fireEvent(errorOccurred ? EventTypeNames::error : EventTypeNames::end, utterance, 0, String());
    168 
    169     // Start the next utterance if we just finished one and one was pending.
    170     if (shouldStartSpeaking && !m_utteranceQueue.isEmpty())
    171         startSpeakingImmediately();
    172 }
    173 
    174 void SpeechSynthesis::boundaryEventOccurred(PlatformSpeechSynthesisUtterance* utterance, SpeechBoundary boundary, unsigned charIndex)
    175 {
    176     DEFINE_STATIC_LOCAL(const String, wordBoundaryString, ("word"));
    177     DEFINE_STATIC_LOCAL(const String, sentenceBoundaryString, ("sentence"));
    178 
    179     switch (boundary) {
    180     case SpeechWordBoundary:
    181         fireEvent(EventTypeNames::boundary, static_cast<SpeechSynthesisUtterance*>(utterance->client()), charIndex, wordBoundaryString);
    182         break;
    183     case SpeechSentenceBoundary:
    184         fireEvent(EventTypeNames::boundary, static_cast<SpeechSynthesisUtterance*>(utterance->client()), charIndex, sentenceBoundaryString);
    185         break;
    186     default:
    187         ASSERT_NOT_REACHED();
    188     }
    189 }
    190 
    191 void SpeechSynthesis::didStartSpeaking(PlatformSpeechSynthesisUtterance* utterance)
    192 {
    193     if (utterance->client())
    194         fireEvent(EventTypeNames::start, static_cast<SpeechSynthesisUtterance*>(utterance->client()), 0, String());
    195 }
    196 
    197 void SpeechSynthesis::didPauseSpeaking(PlatformSpeechSynthesisUtterance* utterance)
    198 {
    199     m_isPaused = true;
    200     if (utterance->client())
    201         fireEvent(EventTypeNames::pause, static_cast<SpeechSynthesisUtterance*>(utterance->client()), 0, String());
    202 }
    203 
    204 void SpeechSynthesis::didResumeSpeaking(PlatformSpeechSynthesisUtterance* utterance)
    205 {
    206     m_isPaused = false;
    207     if (utterance->client())
    208         fireEvent(EventTypeNames::resume, static_cast<SpeechSynthesisUtterance*>(utterance->client()), 0, String());
    209 }
    210 
    211 void SpeechSynthesis::didFinishSpeaking(PlatformSpeechSynthesisUtterance* utterance)
    212 {
    213     if (utterance->client())
    214         handleSpeakingCompleted(static_cast<SpeechSynthesisUtterance*>(utterance->client()), false);
    215 }
    216 
    217 void SpeechSynthesis::speakingErrorOccurred(PlatformSpeechSynthesisUtterance* utterance)
    218 {
    219     if (utterance->client())
    220         handleSpeakingCompleted(static_cast<SpeechSynthesisUtterance*>(utterance->client()), true);
    221 }
    222 
    223 SpeechSynthesisUtterance* SpeechSynthesis::currentSpeechUtterance() const
    224 {
    225     if (!m_utteranceQueue.isEmpty())
    226         return m_utteranceQueue.first().get();
    227     return 0;
    228 }
    229 
    230 const AtomicString& SpeechSynthesis::interfaceName() const
    231 {
    232     return EventTargetNames::SpeechSynthesis;
    233 }
    234 
    235 void SpeechSynthesis::trace(Visitor* visitor)
    236 {
    237     visitor->trace(m_platformSpeechSynthesizer);
    238     visitor->trace(m_voiceList);
    239     visitor->trace(m_utteranceQueue);
    240     PlatformSpeechSynthesizerClient::trace(visitor);
    241     EventTargetWithInlineData::trace(visitor);
    242 }
    243 
    244 } // namespace blink
    245