1 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
2 // Use of this source code is governed by a BSD-style license that can be
3 // found in the LICENSE file.
5 #ifndef CHROME_BROWSER_SPEECH_TTS_CONTROLLER_H_
6 #define CHROME_BROWSER_SPEECH_TTS_CONTROLLER_H_
13 #include "base/memory/scoped_ptr.h"
14 #include "base/memory/singleton.h"
15 #include "base/memory/weak_ptr.h"
19 class TtsPlatformImpl;
29 // Events sent back from the TTS engine indicating the progress.
36 TTS_EVENT_INTERRUPTED,
49 // Returns true if this event type is one that indicates an utterance
50 // is finished and can be destroyed.
51 bool IsFinalTtsEventType(TtsEventType event_type);
53 // The continuous parameters that apply to a given utterance.
54 struct UtteranceContinuousParameters {
55 UtteranceContinuousParameters();
62 // Information about one voice.
70 std::string extension_id;
71 std::set<TtsEventType> events;
73 // If true, the synthesis engine is a remote network resource.
74 // It may be higher latency and may incur bandwidth costs.
77 // If true, this is implemented by this platform's subclass of
78 // TtsPlatformImpl. If false, this is implemented by an extension.
80 std::string native_voice_identifier;
83 // Interface that delegates TTS requests to user-installed extensions.
84 class TtsEngineDelegate {
86 virtual ~TtsEngineDelegate() {}
88 // Return a list of all available voices registered.
89 virtual void GetVoices(content::BrowserContext* browser_context,
90 std::vector<VoiceData>* out_voices) = 0;
92 // Speak the given utterance by sending an event to the given TTS engine.
93 virtual void Speak(Utterance* utterance, const VoiceData& voice) = 0;
95 // Stop speaking the given utterance by sending an event to the target
96 // associated with this utterance.
97 virtual void Stop(Utterance* utterance) = 0;
99 // Pause in the middle of speaking this utterance.
100 virtual void Pause(Utterance* utterance) = 0;
102 // Resume speaking this utterance.
103 virtual void Resume(Utterance* utterance) = 0;
105 // Load the built-in component extension for ChromeOS.
106 virtual bool LoadBuiltInTtsExtension(
107 content::BrowserContext* browser_context) = 0;
110 // Class that wants to receive events on utterances.
111 class UtteranceEventDelegate {
113 virtual ~UtteranceEventDelegate() {}
114 virtual void OnTtsEvent(Utterance* utterance,
115 TtsEventType event_type,
117 const std::string& error_message) = 0;
120 // Class that wants to be notified when the set of
121 // voices has changed.
122 class VoicesChangedDelegate {
124 virtual ~VoicesChangedDelegate() {}
125 virtual void OnVoicesChanged() = 0;
128 // One speech utterance.
131 // Construct an utterance given a profile and a completion task to call
132 // when the utterance is done speaking. Before speaking this utterance,
133 // its other parameters like text, rate, pitch, etc. should all be set.
134 explicit Utterance(content::BrowserContext* browser_context);
137 // Sends an event to the delegate. If the event type is TTS_EVENT_END
138 // or TTS_EVENT_ERROR, deletes the utterance. If |char_index| is -1,
139 // uses the last good value.
140 void OnTtsEvent(TtsEventType event_type,
142 const std::string& error_message);
144 // Finish an utterance without sending an event to the delegate.
147 // Getters and setters for the text to speak and other speech options.
148 void set_text(const std::string& text) { text_ = text; }
149 const std::string& text() const { return text_; }
151 void set_options(const base::Value* options);
152 const base::Value* options() const { return options_.get(); }
154 void set_src_extension_id(const std::string& src_extension_id) {
155 src_extension_id_ = src_extension_id;
157 const std::string& src_extension_id() { return src_extension_id_; }
159 void set_src_id(int src_id) { src_id_ = src_id; }
160 int src_id() { return src_id_; }
162 void set_src_url(const GURL& src_url) { src_url_ = src_url; }
163 const GURL& src_url() { return src_url_; }
165 void set_voice_name(const std::string& voice_name) {
166 voice_name_ = voice_name;
168 const std::string& voice_name() const { return voice_name_; }
170 void set_lang(const std::string& lang) {
173 const std::string& lang() const { return lang_; }
175 void set_gender(TtsGenderType gender) {
178 TtsGenderType gender() const { return gender_; }
180 void set_continuous_parameters(const UtteranceContinuousParameters& params) {
181 continuous_parameters_ = params;
183 const UtteranceContinuousParameters& continuous_parameters() {
184 return continuous_parameters_;
187 void set_can_enqueue(bool can_enqueue) { can_enqueue_ = can_enqueue; }
188 bool can_enqueue() const { return can_enqueue_; }
190 void set_required_event_types(const std::set<TtsEventType>& types) {
191 required_event_types_ = types;
193 const std::set<TtsEventType>& required_event_types() const {
194 return required_event_types_;
197 void set_desired_event_types(const std::set<TtsEventType>& types) {
198 desired_event_types_ = types;
200 const std::set<TtsEventType>& desired_event_types() const {
201 return desired_event_types_;
204 const std::string& extension_id() const { return extension_id_; }
205 void set_extension_id(const std::string& extension_id) {
206 extension_id_ = extension_id;
209 UtteranceEventDelegate* event_delegate() const {
210 return event_delegate_.get();
212 void set_event_delegate(
213 base::WeakPtr<UtteranceEventDelegate> event_delegate) {
214 event_delegate_ = event_delegate;
217 // Getters and setters for internal state.
218 content::BrowserContext* browser_context() const { return browser_context_; }
219 int id() const { return id_; }
220 bool finished() const { return finished_; }
223 // The BrowserContext that initiated this utterance.
224 content::BrowserContext* browser_context_;
226 // The extension ID of the extension providing TTS for this utterance, or
227 // empty if native TTS is being used.
228 std::string extension_id_;
230 // The unique ID of this utterance, used to associate callback functions
234 // The id of the next utterance, so we can associate requests with
236 static int next_utterance_id_;
238 // The text to speak.
241 // The full options arg passed to tts.speak, which may include fields
242 // other than the ones we explicitly parse, below.
243 scoped_ptr<base::Value> options_;
245 // The extension ID of the extension that called speak() and should
247 std::string src_extension_id_;
249 // The source extension's ID of this utterance, so that it can associate
250 // events with the appropriate callback.
253 // The URL of the page where the source extension called speak.
256 // The delegate to be called when an utterance event is fired.
257 base::WeakPtr<UtteranceEventDelegate> event_delegate_;
259 // The parsed options.
260 std::string voice_name_;
262 TtsGenderType gender_;
263 UtteranceContinuousParameters continuous_parameters_;
265 std::set<TtsEventType> required_event_types_;
266 std::set<TtsEventType> desired_event_types_;
268 // The index of the current char being spoken.
271 // True if this utterance received an event indicating it's done.
275 // Singleton class that manages text-to-speech for the TTS and TTS engine
276 // extension APIs, maintaining a queue of pending utterances and keeping
277 // track of all state.
278 class TtsController {
280 // Get the single instance of this class.
281 static TtsController* GetInstance();
283 // Returns true if we're currently speaking an utterance.
284 virtual bool IsSpeaking() = 0;
286 // Speak the given utterance. If the utterance's can_enqueue flag is true
287 // and another utterance is in progress, adds it to the end of the queue.
288 // Otherwise, interrupts any current utterance and speaks this one
290 virtual void SpeakOrEnqueue(Utterance* utterance) = 0;
292 // Stop all utterances and flush the queue. Implies leaving pause mode
294 virtual void Stop() = 0;
296 // Pause the speech queue. Some engines may support pausing in the middle
298 virtual void Pause() = 0;
301 virtual void Resume() = 0;
303 // Handle events received from the speech engine. Events are forwarded to
304 // the callback function, and in addition, completion and error events
305 // trigger finishing the current utterance and starting the next one, if
307 virtual void OnTtsEvent(int utterance_id,
308 TtsEventType event_type,
310 const std::string& error_message) = 0;
312 // Return a list of all available voices, including the native voice,
313 // if supported, and all voices registered by extensions.
314 virtual void GetVoices(content::BrowserContext* browser_context,
315 std::vector<VoiceData>* out_voices) = 0;
317 // Called by the extension system or platform implementation when the
318 // list of voices may have changed and should be re-queried.
319 virtual void VoicesChanged() = 0;
321 // Add a delegate that wants to be notified when the set of voices changes.
322 virtual void AddVoicesChangedDelegate(VoicesChangedDelegate* delegate) = 0;
324 // Remove delegate that wants to be notified when the set of voices changes.
325 virtual void RemoveVoicesChangedDelegate(VoicesChangedDelegate* delegate) = 0;
327 // Set the delegate that processes TTS requests with user-installed
329 virtual void SetTtsEngineDelegate(TtsEngineDelegate* delegate) = 0;
331 // Get the delegate that processes TTS requests with user-installed
333 virtual TtsEngineDelegate* GetTtsEngineDelegate() = 0;
336 virtual void SetPlatformImpl(TtsPlatformImpl* platform_impl) = 0;
337 virtual int QueueSize() = 0;
340 virtual ~TtsController() {}
343 #endif // CHROME_BROWSER_SPEECH_TTS_CONTROLLER_H_