diff options
Diffstat (limited to 'platform/osx')
| -rw-r--r-- | platform/osx/SCsub | 1 | ||||
| -rw-r--r-- | platform/osx/display_server_osx.h | 11 | ||||
| -rw-r--r-- | platform/osx/display_server_osx.mm | 41 | ||||
| -rw-r--r-- | platform/osx/tts_osx.h | 66 | ||||
| -rw-r--r-- | platform/osx/tts_osx.mm | 266 |
5 files changed, 385 insertions, 0 deletions
diff --git a/platform/osx/SCsub b/platform/osx/SCsub index d72a75af04..3a4c95613d 100644 --- a/platform/osx/SCsub +++ b/platform/osx/SCsub @@ -18,6 +18,7 @@ files = [ "key_mapping_osx.mm", "godot_main_osx.mm", "dir_access_osx.mm", + "tts_osx.mm", "joypad_osx.cpp", "vulkan_context_osx.mm", "gl_manager_osx_legacy.mm", diff --git a/platform/osx/display_server_osx.h b/platform/osx/display_server_osx.h index fa3091ff81..fcb3a62bec 100644 --- a/platform/osx/display_server_osx.h +++ b/platform/osx/display_server_osx.h @@ -137,6 +137,8 @@ private: Vector<KeyEvent> key_event_buffer; int key_event_pos = 0; + id tts = nullptr; + Point2i im_selection; String im_text; @@ -264,6 +266,15 @@ public: virtual void global_menu_remove_item(const String &p_menu_root, int p_idx) override; virtual void global_menu_clear(const String &p_menu_root) override; + virtual bool tts_is_speaking() const override; + virtual bool tts_is_paused() const override; + virtual Array tts_get_voices() const override; + + virtual void tts_speak(const String &p_text, const String &p_voice, int p_volume = 50, float p_pitch = 1.f, float p_rate = 1.f, int p_utterance_id = 0, bool p_interrupt = false) override; + virtual void tts_pause() override; + virtual void tts_resume() override; + virtual void tts_stop() override; + virtual Error dialog_show(String p_title, String p_description, Vector<String> p_buttons, const Callable &p_callback) override; virtual Error dialog_input_text(String p_title, String p_description, String p_partial, const Callable &p_callback) override; diff --git a/platform/osx/display_server_osx.mm b/platform/osx/display_server_osx.mm index d209c90d87..17a44a3fbd 100644 --- a/platform/osx/display_server_osx.mm +++ b/platform/osx/display_server_osx.mm @@ -37,6 +37,8 @@ #include "key_mapping_osx.h" #include "os_osx.h" +#include "tts_osx.h" + #include "core/io/marshalls.h" #include "core/math/geometry_2d.h" #include "core/os/keyboard.h" @@ -702,6 +704,7 @@ bool DisplayServerOSX::has_feature(Feature p_feature) const { case FEATURE_NATIVE_ICON: //case FEATURE_KEEP_SCREEN_ON: case FEATURE_SWAP_BUFFERS: + case FEATURE_TEXT_TO_SPEECH: return true; default: { } @@ -1458,6 +1461,41 @@ void DisplayServerOSX::global_menu_clear(const String &p_menu_root) { } } +bool DisplayServerOSX::tts_is_speaking() const { + ERR_FAIL_COND_V(!tts, false); + return [tts isSpeaking]; +} + +bool DisplayServerOSX::tts_is_paused() const { + ERR_FAIL_COND_V(!tts, false); + return [tts isPaused]; +} + +Array DisplayServerOSX::tts_get_voices() const { + ERR_FAIL_COND_V(!tts, Array()); + return [tts getVoices]; +} + +void DisplayServerOSX::tts_speak(const String &p_text, const String &p_voice, int p_volume, float p_pitch, float p_rate, int p_utterance_id, bool p_interrupt) { + ERR_FAIL_COND(!tts); + [tts speak:p_text voice:p_voice volume:p_volume pitch:p_pitch rate:p_rate utterance_id:p_utterance_id interrupt:p_interrupt]; +} + +void DisplayServerOSX::tts_pause() { + ERR_FAIL_COND(!tts); + [tts pauseSpeaking]; +} + +void DisplayServerOSX::tts_resume() { + ERR_FAIL_COND(!tts); + [tts resumeSpeaking]; +} + +void DisplayServerOSX::tts_stop() { + ERR_FAIL_COND(!tts); + [tts stopSpeaking]; +} + Error DisplayServerOSX::dialog_show(String p_title, String p_description, Vector<String> p_buttons, const Callable &p_callback) { _THREAD_SAFE_METHOD_ @@ -3121,6 +3159,9 @@ DisplayServerOSX::DisplayServerOSX(const String &p_rendering_driver, WindowMode // Register to be notified on displays arrangement changes. CGDisplayRegisterReconfigurationCallback(_displays_arrangement_changed, nullptr); + // Init TTS + tts = [[TTS_OSX alloc] init]; + NSMenuItem *menu_item; NSString *title; diff --git a/platform/osx/tts_osx.h b/platform/osx/tts_osx.h new file mode 100644 index 0000000000..2cf6d21c18 --- /dev/null +++ b/platform/osx/tts_osx.h @@ -0,0 +1,66 @@ +/*************************************************************************/ +/* tts_osx.h */ +/*************************************************************************/ +/* This file is part of: */ +/* GODOT ENGINE */ +/* https://godotengine.org */ +/*************************************************************************/ +/* Copyright (c) 2007-2022 Juan Linietsky, Ariel Manzur. */ +/* Copyright (c) 2014-2022 Godot Engine contributors (cf. AUTHORS.md). */ +/* */ +/* Permission is hereby granted, free of charge, to any person obtaining */ +/* a copy of this software and associated documentation files (the */ +/* "Software"), to deal in the Software without restriction, including */ +/* without limitation the rights to use, copy, modify, merge, publish, */ +/* distribute, sublicense, and/or sell copies of the Software, and to */ +/* permit persons to whom the Software is furnished to do so, subject to */ +/* the following conditions: */ +/* */ +/* The above copyright notice and this permission notice shall be */ +/* included in all copies or substantial portions of the Software. */ +/* */ +/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */ +/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */ +/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/ +/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */ +/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */ +/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */ +/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ +/*************************************************************************/ + +#ifndef TTS_OSX_H +#define TTS_OSX_H + +#include "core/string/ustring.h" +#include "core/templates/list.h" +#include "core/templates/map.h" +#include "core/variant/array.h" +#include "servers/display_server.h" + +#include <AVFAudio/AVSpeechSynthesis.h> +#include <AppKit/AppKit.h> + +@interface TTS_OSX : NSObject <AVSpeechSynthesizerDelegate> { + // AVSpeechSynthesizer + bool speaking; + Map<id, int> ids; + + // NSSpeechSynthesizer + bool paused; + bool have_utterance; + int last_utterance; + + id synth; // NSSpeechSynthesizer or AVSpeechSynthesizer + List<DisplayServer::TTSUtterance> queue; +} + +- (void)pauseSpeaking; +- (void)resumeSpeaking; +- (void)stopSpeaking; +- (bool)isSpeaking; +- (bool)isPaused; +- (void)speak:(const String &)text voice:(const String &)voice volume:(int)volume pitch:(float)pitch rate:(float)rate utterance_id:(int)utterance_id interrupt:(bool)interrupt; +- (Array)getVoices; +@end + +#endif // TTS_OSX_H diff --git a/platform/osx/tts_osx.mm b/platform/osx/tts_osx.mm new file mode 100644 index 0000000000..e6a5236cd9 --- /dev/null +++ b/platform/osx/tts_osx.mm @@ -0,0 +1,266 @@ +/*************************************************************************/ +/* tts_osx.mm */ +/*************************************************************************/ +/* This file is part of: */ +/* GODOT ENGINE */ +/* https://godotengine.org */ +/*************************************************************************/ +/* Copyright (c) 2007-2022 Juan Linietsky, Ariel Manzur. */ +/* Copyright (c) 2014-2022 Godot Engine contributors (cf. AUTHORS.md). */ +/* */ +/* Permission is hereby granted, free of charge, to any person obtaining */ +/* a copy of this software and associated documentation files (the */ +/* "Software"), to deal in the Software without restriction, including */ +/* without limitation the rights to use, copy, modify, merge, publish, */ +/* distribute, sublicense, and/or sell copies of the Software, and to */ +/* permit persons to whom the Software is furnished to do so, subject to */ +/* the following conditions: */ +/* */ +/* The above copyright notice and this permission notice shall be */ +/* included in all copies or substantial portions of the Software. */ +/* */ +/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */ +/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */ +/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/ +/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */ +/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */ +/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */ +/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ +/*************************************************************************/ + +#include "tts_osx.h" + +@implementation TTS_OSX + +- (id)init { + self = [super init]; + self->speaking = false; + self->have_utterance = false; + self->last_utterance = -1; + self->paused = false; + if (@available(macOS 10.14, *)) { + self->synth = [[AVSpeechSynthesizer alloc] init]; + [self->synth setDelegate:self]; + print_verbose("Text-to-Speech: AVSpeechSynthesizer initialized."); + } else { + self->synth = [[NSSpeechSynthesizer alloc] init]; + [self->synth setDelegate:self]; + print_verbose("Text-to-Speech: NSSpeechSynthesizer initialized."); + } + return self; +} + +// AVSpeechSynthesizer callback (macOS 10.14+) + +- (void)speechSynthesizer:(AVSpeechSynthesizer *)av_synth willSpeakRangeOfSpeechString:(NSRange)characterRange utterance:(AVSpeechUtterance *)utterance API_AVAILABLE(macosx(10.14)) { + NSString *string = [utterance speechString]; + + // Convert from UTF-16 to UTF-32 position. + int pos = 0; + for (NSUInteger i = 0; i < MIN(characterRange.location, string.length); i++) { + unichar c = [string characterAtIndex:i]; + if ((c & 0xfffffc00) == 0xd800) { + i++; + } + pos++; + } + + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_BOUNDARY, ids[utterance], pos); +} + +// AVSpeechSynthesizer callback (macOS 10.14+) + +- (void)speechSynthesizer:(AVSpeechSynthesizer *)av_synth didCancelSpeechUtterance:(AVSpeechUtterance *)utterance API_AVAILABLE(macosx(10.14)) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_CANCELED, ids[utterance]); + ids.erase(utterance); + speaking = false; + [self update]; +} + +// AVSpeechSynthesizer callback (macOS 10.14+) + +- (void)speechSynthesizer:(AVSpeechSynthesizer *)av_synth didFinishSpeechUtterance:(AVSpeechUtterance *)utterance API_AVAILABLE(macosx(10.14)) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_ENDED, ids[utterance]); + ids.erase(utterance); + speaking = false; + [self update]; +} + +// NSSpeechSynthesizer callback (macOS 10.4+) + +- (void)speechSynthesizer:(NSSpeechSynthesizer *)ns_synth willSpeakWord:(NSRange)characterRange ofString:(NSString *)string { + if (!paused && have_utterance) { + // Convert from UTF-16 to UTF-32 position. + int pos = 0; + for (NSUInteger i = 0; i < MIN(characterRange.location, string.length); i++) { + unichar c = [string characterAtIndex:i]; + if ((c & 0xfffffc00) == 0xd800) { + i++; + } + pos++; + } + + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_BOUNDARY, last_utterance, pos); + } +} + +- (void)speechSynthesizer:(NSSpeechSynthesizer *)ns_synth didFinishSpeaking:(BOOL)success { + if (!paused && have_utterance) { + if (success) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_ENDED, last_utterance); + } else { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_CANCELED, last_utterance); + } + have_utterance = false; + } + speaking = false; + [self update]; +} + +- (void)update { + if (!speaking && queue.size() > 0) { + DisplayServer::TTSUtterance &message = queue.front()->get(); + + if (@available(macOS 10.14, *)) { + AVSpeechSynthesizer *av_synth = synth; + AVSpeechUtterance *new_utterance = [[AVSpeechUtterance alloc] initWithString:[NSString stringWithUTF8String:message.text.utf8().get_data()]]; + [new_utterance setVoice:[AVSpeechSynthesisVoice voiceWithIdentifier:[NSString stringWithUTF8String:message.voice.utf8().get_data()]]]; + if (message.rate > 1.f) { + [new_utterance setRate:Math::range_lerp(message.rate, 1.f, 10.f, AVSpeechUtteranceDefaultSpeechRate, AVSpeechUtteranceMaximumSpeechRate)]; + } else if (message.rate < 1.f) { + [new_utterance setRate:Math::range_lerp(message.rate, 0.1f, 1.f, AVSpeechUtteranceMinimumSpeechRate, AVSpeechUtteranceDefaultSpeechRate)]; + } + [new_utterance setPitchMultiplier:message.pitch]; + [new_utterance setVolume:(Math::range_lerp(message.volume, 0.f, 100.f, 0.f, 1.f))]; + + ids[new_utterance] = message.id; + [av_synth speakUtterance:new_utterance]; + } else { + NSSpeechSynthesizer *ns_synth = synth; + [ns_synth setObject:nil forProperty:NSSpeechResetProperty error:nil]; + [ns_synth setVoice:[NSString stringWithUTF8String:message.voice.utf8().get_data()]]; + int base_pitch = [[ns_synth objectForProperty:NSSpeechPitchBaseProperty error:nil] intValue]; + [ns_synth setObject:[NSNumber numberWithInt:(base_pitch * (message.pitch / 2.f + 0.5f))] forProperty:NSSpeechPitchBaseProperty error:nullptr]; + [ns_synth setVolume:(Math::range_lerp(message.volume, 0.f, 100.f, 0.f, 1.f))]; + [ns_synth setRate:(message.rate * 200)]; + + last_utterance = message.id; + have_utterance = true; + [ns_synth startSpeakingString:[NSString stringWithUTF8String:message.text.utf8().get_data()]]; + } + queue.pop_front(); + + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_STARTED, message.id); + speaking = true; + } +} + +- (void)pauseSpeaking { + if (@available(macOS 10.14, *)) { + AVSpeechSynthesizer *av_synth = synth; + [av_synth pauseSpeakingAtBoundary:AVSpeechBoundaryImmediate]; + } else { + NSSpeechSynthesizer *ns_synth = synth; + [ns_synth pauseSpeakingAtBoundary:NSSpeechImmediateBoundary]; + } + paused = true; +} + +- (void)resumeSpeaking { + if (@available(macOS 10.14, *)) { + AVSpeechSynthesizer *av_synth = synth; + [av_synth continueSpeaking]; + } else { + NSSpeechSynthesizer *ns_synth = synth; + [ns_synth continueSpeaking]; + } + paused = false; +} + +- (void)stopSpeaking { + for (DisplayServer::TTSUtterance &message : queue) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_CANCELED, message.id); + } + queue.clear(); + if (@available(macOS 10.14, *)) { + AVSpeechSynthesizer *av_synth = synth; + [av_synth stopSpeakingAtBoundary:AVSpeechBoundaryImmediate]; + } else { + NSSpeechSynthesizer *ns_synth = synth; + if (have_utterance) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_CANCELED, last_utterance); + } + [ns_synth stopSpeaking]; + } + have_utterance = false; + speaking = false; + paused = false; +} + +- (bool)isSpeaking { + return speaking || (queue.size() > 0); +} + +- (bool)isPaused { + if (@available(macOS 10.14, *)) { + AVSpeechSynthesizer *av_synth = synth; + return [av_synth isPaused]; + } else { + return paused; + } +} + +- (void)speak:(const String &)text voice:(const String &)voice volume:(int)volume pitch:(float)pitch rate:(float)rate utterance_id:(int)utterance_id interrupt:(bool)interrupt { + if (interrupt) { + [self stopSpeaking]; + } + + if (text.is_empty()) { + DisplayServer::get_singleton()->tts_post_utterance_event(DisplayServer::TTS_UTTERANCE_CANCELED, utterance_id); + return; + } + + DisplayServer::TTSUtterance message; + message.text = text; + message.voice = voice; + message.volume = CLAMP(volume, 0, 100); + message.pitch = CLAMP(pitch, 0.f, 2.f); + message.rate = CLAMP(rate, 0.1f, 10.f); + message.id = utterance_id; + queue.push_back(message); + + if ([self isPaused]) { + [self resumeSpeaking]; + } else { + [self update]; + } +} + +- (Array)getVoices { + Array list; + if (@available(macOS 10.14, *)) { + for (AVSpeechSynthesisVoice *voice in [AVSpeechSynthesisVoice speechVoices]) { + NSString *voiceIdentifierString = [voice identifier]; + NSString *voiceLocaleIdentifier = [voice language]; + NSString *voiceName = [voice name]; + Dictionary voice_d; + voice_d["name"] = String::utf8([voiceName UTF8String]); + voice_d["id"] = String::utf8([voiceIdentifierString UTF8String]); + voice_d["language"] = String::utf8([voiceLocaleIdentifier UTF8String]); + list.push_back(voice_d); + } + } else { + for (NSString *voiceIdentifierString in [NSSpeechSynthesizer availableVoices]) { + NSString *voiceLocaleIdentifier = [[NSSpeechSynthesizer attributesForVoice:voiceIdentifierString] objectForKey:NSVoiceLocaleIdentifier]; + NSString *voiceName = [[NSSpeechSynthesizer attributesForVoice:voiceIdentifierString] objectForKey:NSVoiceName]; + Dictionary voice_d; + voice_d["name"] = String([voiceName UTF8String]); + voice_d["id"] = String([voiceIdentifierString UTF8String]); + voice_d["language"] = String([voiceLocaleIdentifier UTF8String]); + list.push_back(voice_d); + } + } + return list; +} + +@end |