TTS: Add part of linux TTS

7 files changed, 326 insertions, 4 deletions

diff --git a/backends/text-to-speech/linux/linux-text-to-speech.cpp b/backends/text-to-speech/linux/linux-text-to-speech.cpp
index 7a8e899595..bbe6705f58 100644
--- a/backends/text-to-speech/linux/linux-text-to-speech.cpp
+++ b/backends/text-to-speech/linux/linux-text-to-speech.cpp
@@ -26,12 +26,177 @@
 #include "backends/text-to-speech/linux/linux-text-to-speech.h"
 
 #if defined(USE_LINUX_TTS)
+#include <speech-dispatcher/libspeechd.h>
+
 #include "common/translation.h"
+#include "common/debug.h"
+#include "common/system.h"
+SPDConnection *_connection;
+
+void speech_begin_callback(size_t msg_id, size_t client_id, SPDNotificationType state){
+	LinuxTextToSpeechManager *manager =
+		static_cast<LinuxTextToSpeechManager *> (g_system->getTextToSpeechManager());
+	manager->updateState(LinuxTextToSpeechManager::SPEAKING);
+}
+
+void speech_end_callback(size_t msg_id, size_t client_id, SPDNotificationType state){
+	LinuxTextToSpeechManager *manager =
+		static_cast<LinuxTextToSpeechManager *> (g_system->getTextToSpeechManager());
+	manager->updateState(LinuxTextToSpeechManager::READY);
+}
+
+void speech_cancel_callback(size_t msg_id, size_t client_id, SPDNotificationType state){
+	LinuxTextToSpeechManager *manager =
+		static_cast<LinuxTextToSpeechManager *> (g_system->getTextToSpeechManager());
+	manager->updateState(LinuxTextToSpeechManager::READY);
+}
+
+void speech_resume_callback(size_t msg_id, size_t client_id, SPDNotificationType state){
+	LinuxTextToSpeechManager *manager =
+		static_cast<LinuxTextToSpeechManager *> (g_system->getTextToSpeechManager());
+	manager->updateState(LinuxTextToSpeechManager::SPEAKING);
+}
+
+void speech_pause_callback(size_t msg_id, size_t client_id, SPDNotificationType state){
+	LinuxTextToSpeechManager *manager =
+		static_cast<LinuxTextToSpeechManager *> (g_system->getTextToSpeechManager());
+	manager->updateState(LinuxTextToSpeechManager::PAUSED);
+}
+
+LinuxTextToSpeechManager::LinuxTextToSpeechManager()
+	: _speechState(READY) {
+	_connection = spd_open("ScummVM", "main", NULL, SPD_MODE_THREADED);
+	if (_connection == 0) {
+		debug("couldn't open");
+		return;
+	}
 
-LinuxTextToSpeechManager::LinuxTextToSpeechManager() {
+	_connection->callback_begin = speech_begin_callback;
+	spd_set_notification_on(_connection, SPD_BEGIN);
+	_connection->callback_end = speech_end_callback;
+	spd_set_notification_on(_connection, SPD_END);
+	_connection->callback_cancel = speech_cancel_callback;
+	spd_set_notification_on(_connection, SPD_CANCEL);
+	_connection->callback_resume = speech_resume_callback;
+	spd_set_notification_on(_connection, SPD_RESUME);
+	_connection->callback_pause = speech_pause_callback;
+	spd_set_notification_on(_connection, SPD_PAUSE);
+
+	setLanguage(Common::String("en"));
+	updateVoices();
 }
 
 LinuxTextToSpeechManager::~LinuxTextToSpeechManager() {
+	spd_close(_connection);
+}
+
+void LinuxTextToSpeechManager::updateState(LinuxTextToSpeechManager::SpeechState state) {
+	_speechState = state;
+}
+
+bool LinuxTextToSpeechManager::say(Common::String str) {
+	if (isSpeaking())
+		stop();
+	return spd_say(_connection, SPD_MESSAGE, str.c_str()) == -1;
+
+}
+
+bool LinuxTextToSpeechManager::stop() {
+	if (_speechState == READY)
+		return false;
+	return spd_cancel(_connection) == -1;
+}
+
+bool LinuxTextToSpeechManager::pause() {
+	if (_speechState == READY || _speechState == PAUSED)
+		return false;
+	return spd_pause(_connection) == -1;
+}
+
+bool LinuxTextToSpeechManager::resume() {
+	if (_speechState == READY || _speechState == SPEAKING)
+		return false;
+	return spd_resume(_connection) == -1;
+}
+
+bool LinuxTextToSpeechManager::isSpeaking() {
+	if (_speechState == SPEAKING)
+		return true;
+	return false;
+}
+
+void LinuxTextToSpeechManager::setVoice(Common::TTSVoice *voice) {
+	assert(voice != nullptr && voice->getData() != nullptr);
+	spd_set_voice_type(_connection, *(SPDVoiceType *)(voice->getData()));
+	_ttsState->_activeVoice = voice;
+}
+
+void LinuxTextToSpeechManager::setRate(int rate) {
+	assert(rate >= -100 && rate <= 100);
+	spd_set_voice_rate(_connection, rate);
+	_ttsState->_rate = rate;
+}
+
+void LinuxTextToSpeechManager::setPitch(int pitch) {
+	assert(pitch >= -100 && pitch <= 100);
+	spd_set_voice_pitch(_connection, pitch);
+	_ttsState->_pitch = pitch;
+}
+
+void LinuxTextToSpeechManager::setVolume(int volume) {
+	assert(volume >= -100 && volume <= 100);
+	spd_set_volume(_connection, volume);
+	_ttsState->_volume = volume;
+}
+
+void LinuxTextToSpeechManager::setLanguage(Common::String language) {
+	spd_set_language(_connection, language.c_str());
+	_ttsState->_language = language;
+	if (_ttsState->_activeVoice)
+		setVoice(_ttsState->_activeVoice);
+}
+
+void LinuxTextToSpeechManager::updateVoices() {
+	/* just use these voices:
+	   SPD_MALE1, SPD_MALE2, SPD_MALE3,
+	   SPD_FEMALE1, SPD_FEMALE2, SPD_FEMALE3,
+	   SPD_CHILD_MALE, SPD_CHILD_FEMALE
+
+	   it depends on the user to map them to the right voices in speech-dispatcher
+	   configuration
+	*/
+
+	SPDVoiceType *type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_MALE1;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::MALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_MALE2;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::MALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_MALE3;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::MALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_FEMALE1;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::FEMALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_FEMALE2;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::FEMALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_FEMALE3;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::FEMALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_CHILD_MALE;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::MALE, (void *) type));
+
+	type = (SPDVoiceType *) malloc(sizeof(SPDVoiceType));
+	*type = SPD_CHILD_FEMALE;
+	_ttsState->_availaibleVoices.push_back(Common::TTSVoice(Common::TTSVoice::FEMALE, (void *) type));
 }
 
 #endif
diff --git a/backends/text-to-speech/linux/linux-text-to-speech.h b/backends/text-to-speech/linux/linux-text-to-speech.h
index ab9b2b6fe2..b0f71d33b6 100644
--- a/backends/text-to-speech/linux/linux-text-to-speech.h
+++ b/backends/text-to-speech/linux/linux-text-to-speech.h
@@ -28,11 +28,42 @@
 #if defined(USE_LINUX_TTS)
 
 #include "common/text-to-speech.h"
+#include "common/str.h"
 
 class LinuxTextToSpeechManager : public Common::TextToSpeechManager {
 public:
+	enum SpeechState {
+		READY,
+		PAUSED,
+		SPEAKING
+	};
+
 	LinuxTextToSpeechManager();
 	virtual ~LinuxTextToSpeechManager();
+
+	virtual bool say(Common::String str);
+
+	virtual bool stop();
+	virtual bool pause();
+	virtual bool resume();
+
+	virtual bool isSpeaking();
+	
+	virtual void setVoice(Common::TTSVoice *voice);
+
+	virtual void setRate(int rate);
+
+	virtual void setPitch(int pitch);
+
+	virtual void setVolume(int volume);
+
+	virtual void setLanguage(Common::String language);
+
+	void updateState(SpeechState state);
+
+private:
+	virtual void updateVoices();
+	SpeechState _speechState;
 };
 
 #endif
diff --git a/common/module.mk b/common/module.mk
index 46e32d72d9..56ed05fe72 100644
--- a/common/module.mk
+++ b/common/module.mk
@@ -99,5 +99,10 @@ MODULE_OBJS += \
 	lua/scummvm_file.o
 endif
 
+ifdef USE_TTS
+MODULE_OBJS += \
+	text-to-speech.o
+endif
+
 # Include common rules
 include $(srcdir)/rules.mk
diff --git a/common/text-to-speech.cpp b/common/text-to-speech.cpp
new file mode 100644
index 0000000000..d1b9539f93
--- /dev/null
+++ b/common/text-to-speech.cpp
@@ -0,0 +1,47 @@
+/* ScummVM - Graphic Adventure Engine
+ *
+ * ScummVM is the legal property of its developers, whose names
+ * are too numerous to list here. Please refer to the COPYRIGHT
+ * file distributed with this source distribution.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ */
+
+
+#include "common/text-to-speech.h"
+#if defined(USE_TTS)
+
+namespace Common {
+TextToSpeechManager::TextToSpeechManager() {
+	_ttsState = new TTSState;
+	_ttsState->_pitch = 0;
+	_ttsState->_volume = 0;
+	_ttsState->_rate = 0;
+	_ttsState->_activeVoice = nullptr;
+	_ttsState->_next = nullptr;
+}
+
+TextToSpeechManager::~TextToSpeechManager() {
+	TTSState *tmp = _ttsState;
+	while (tmp != nullptr) {
+		tmp = _ttsState->_next;
+		delete _ttsState;
+		_ttsState = tmp;
+	}
+}
+
+}
+#endif
diff --git a/common/text-to-speech.h b/common/text-to-speech.h
index e95050e3f5..62cc5e878b 100644
--- a/common/text-to-speech.h
+++ b/common/text-to-speech.h
@@ -23,19 +23,90 @@
 #ifndef BACKENDS_TEXT_TO_SPEECH_ABSTRACT_H
 #define BACKENDS_TEXT_TO_SPEECH_ABSTRACT_H
 
+#include "common/scummsys.h"
+
 #if defined(USE_TTS)
 
+#include "common/array.h"
+#include "common/debug.h"
 namespace Common {
 
+class TTSVoice {
+	friend class TextToSpeechManager;
+
+	public:
+		enum Gender {
+			MALE,
+			FEMALE,
+			UNKNOWN
+		};
+
+	public:
+		TTSVoice()
+			: _gender(UNKNOWN)
+			, _data(nullptr) {}
+		TTSVoice(Gender gender, void *data) 
+			: _gender(gender)
+			, _data(data) {}
+		~TTSVoice() { debug("%d", * (int *)_data); if (_data != nullptr) free(_data); }
+		Gender getGender() { return _gender; };
+		void setGender(Gender gender) { _gender = gender; };
+		void setData(void *data) { _data = data; };
+		void *getData() { return _data; };
+
+	protected:
+		Gender _gender;
+		void *_data;
+};
+
+struct TTSState {
+	int _rate;
+	int _pitch;
+	int _volume;
+	String _language;
+	TTSVoice *_activeVoice;
+	Array<TTSVoice> _availaibleVoices;
+	TTSState *_next;
+};
+
 /**
  * The TextToSpeechManager allows speech synthesis.
  *
  */
 class TextToSpeechManager {
 public:
-	TextToSpeechManager() {}
-	virtual ~TextToSpeechManager() {}
+	TextToSpeechManager();
+	virtual ~TextToSpeechManager();
+
+	virtual bool say(String str) { return false; }
+
+	virtual bool stop() { return false; }
+	virtual bool pause() { return false; }
+	virtual bool resume() { return false; }
+
+	virtual bool isSpeaking() { return false; }
+	
+	virtual void setVoice(TTSVoice *voice) {}
+	TTSVoice getVoice() { return *(_ttsState->_activeVoice); }
+
+	virtual void setRate(int rate) {}
+	int getRate() { return _ttsState->_rate; }
+
+	virtual void setPitch(int pitch) {}
+	int getPitch() { return _ttsState->_pitch; }
+
+	virtual void setVolume(int volume) {}
+	int getVolume() { return _ttsState->_volume; }
+
+	virtual void setLanguage(String language) {}
+	String getLanguage() { return _ttsState->_language; }
+
+	Array<TTSVoice> getVoicesArray() { return _ttsState->_availaibleVoices; }
+
+protected:
+	TTSState *_ttsState;
 
+	virtual void updateVoices() {};
 };
 
 } // End of namespace Common
diff --git a/configure b/configure
index 8cfe58473e..af65184a2e 100755
--- a/configure
+++ b/configure
@@ -5379,6 +5379,7 @@ else
 fi
 define_in_config_if_yes $_tts 'USE_TTS'
 define_in_config_if_yes $_linux_tts 'USE_LINUX_TTS'
+append_var LIBS '-lspeechd'
 
 #
 # Check whether to build Bink video support
diff --git a/gui/widget.h b/gui/widget.h
index f87816b2e3..bcab3b4fa6 100644
--- a/gui/widget.h
+++ b/gui/widget.h
@@ -31,6 +31,8 @@
 #include "graphics/surface.h"
 #include "gui/object.h"
 #include "gui/ThemeEngine.h"
+#include "common/text-to-speech.h"
+#include "common/system.h"
 
 namespace GUI {
 
@@ -218,7 +220,7 @@ public:
 
 	void handleMouseUp(int x, int y, int button, int clickCount);
 	void handleMouseDown(int x, int y, int button, int clickCount);
-	void handleMouseEntered(int button)	{ if (_duringPress) { setFlags(WIDGET_PRESSED); } else { setFlags(WIDGET_HILITED); } markAsDirty(); }
+	void handleMouseEntered(int button)	{ g_system->getTextToSpeechManager()->say(_label); if (_duringPress) { setFlags(WIDGET_PRESSED); } else { setFlags(WIDGET_HILITED); } markAsDirty(); }
 	void handleMouseLeft(int button)	{ clearFlags(WIDGET_HILITED | WIDGET_PRESSED); markAsDirty(); }
 
 	void setHighLighted(bool enable);