Merge pull request #175 from yjg30737/Dev

v1.6.0

Author: yjg30737

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "pyqt-openai"
-version = "1.5.0"
+version = "1.6.0"
 description = "Python multipurpose chatbot that user can use GPT, other AI models altogether (Release Name: VividNode)"
 authors = [{ name = "Jung Gyu Yoon", email = "yjg30737@gmail.com" }]
 license = { text = "MIT" }
@@ -16,6 +16,7 @@ dependencies = [
     "requests",
     "pyaudio",
     "pillow",
+    "psutil",
 
     "openai",
     "anthropic",
@@ -24,7 +25,9 @@ dependencies = [
     "replicate",
 
     "g4f",
-    "curl_cffi"
+    "curl_cffi",
+
+    "edge-tts"
 ]
 keywords = ['openai', 'pyqt', 'pyqt5', 'pyqt6', 'pyside6', 'desktop', 'app', 'chatbot', 'gpt', 'replicate', 'gemini', 'claude', 'llama', 'llm', 'gpt4free']
 

pyqt_openai/__init__.py

@@ -23,7 +23,7 @@
 # For the sake of following the PEP8 standard, we will declare module-level dunder names.
 # PEP8 standard about dunder names: https://peps.python.org/pep-0008/#module-level-dunder-names
 
-__version__ = "1.5.0"
+__version__ = "1.6.0"
 __author__ = "Jung Gyu Yoon"
 
 # Constants
@@ -37,6 +37,7 @@
 # For Windows
 AUTOSTART_REGISTRY_KEY = r"Software\Microsoft\Windows\CurrentVersion\Run"
 
+
 # Check if the application is frozen (compiled with PyInstaller)
 # If this is main.py, it will return False, that means it is not frozen.
 def is_frozen():
@@ -81,6 +82,7 @@ def get_config_directory():
 # The default updater path (relative to the application's root directory) - For Windows
 UPDATER_PATH = os.path.join(UPDATE_DIR, "Updater.exe")
 
+
 # Move the Updater.exe to the config folder
 def move_updater():
     original_updater_path = os.path.join(ROOT_DIR, "Updater.exe")
@@ -199,6 +201,8 @@ def move_updater():
 ICON_PAYPAL = os.path.join(ICON_PATH, "paypal.png")
 ICON_KOFI = os.path.join(ICON_PATH, "kofi.png")
 ICON_PATREON = os.path.join(ICON_PATH, "patreon.svg")
+ICON_SHORTCUT = os.path.join(ICON_PATH, "shortcut.svg")
+ICON_REALTIME_API = os.path.join(ICON_PATH, "realtime_api.svg")
 
 ## CUSTOMIZE
 DEFAULT_ICON_SIZE = (24, 24)
@@ -243,7 +247,6 @@ def move_updater():
 DEFAULT_SHORTCUT_PROMPT_ENDING = "Ctrl+E"
 DEFAULT_SHORTCUT_SUPPORT_PROMPT_COMMAND = "Ctrl+Shift+P"
 DEFAULT_SHORTCUT_STACK_ON_TOP = "Ctrl+Shift+S"
-DEFAULT_SHORTCUT_SHOW_TOOLBAR = "Ctrl+T"
 DEFAULT_SHORTCUT_SHOW_SECONDARY_TOOLBAR = "Ctrl+Shift+T"
 DEFAULT_SHORTCUT_FOCUS_MODE = "F10"
 DEFAULT_SHORTCUT_FULL_SCREEN = "F11"
@@ -288,7 +291,6 @@ def move_updater():
 PROMPT_JSON_KEY_NAME = "prompt_json"
 PROMPT_MAIN_KEY_NAME = "prompt_main"
 PROMPT_END_KEY_NAME = "prompt_ending"
-PROMPT_NAME_REGEX = "^[a-zA-Z_0-9]+$"
 INDENT_SIZE = 4
 NOTIFIER_MAX_CHAR = 100
 
@@ -355,8 +357,21 @@ def move_updater():
 WHISPER_TTS_VOICE_TYPE = ["alloy", "echo", "fable", "onyx", "nova", "shimmer"]
 WHISPER_TTS_VOICE_SPEED_RANGE = 0.25, 4.0
 WHISPER_TTS_MODEL = "tts-1"
-WHISPER_TTS_DEFAULT_VOICE = "alloy"
-WHISPER_TTS_DEFAULT_SPEED = 1.0
+
+## EDGE-TTS (TTS)
+EDGE_TTS_VOICE_TYPE = [
+    "en-GB-SoniaNeural",
+    "en-US-GuyNeural",
+    "en-US-JennyNeural",
+    "en-US-AvaMultilingualNeural",
+]
+
+# TTS in general
+TTS_DEFAULT_PROVIDER = "OpenAI"
+TTS_DEFAULT_VOICE = "alloy"
+TTS_DEFAULT_SPEED = 1.0
+TTS_DEFAULT_AUTO_PLAY = False
+TTS_DEFAULT_AUTO_STOP_SILENCE_DURATION = 3
 
 STT_MODEL = "whisper-1"
 
@@ -413,6 +428,8 @@ def move_updater():
 
 G4F_USE_CHAT_HISTORY = True
 
+G4F_DEFAULT_IMAGE_MODEL = "flux"
+
 # Dictionary that stores the platform and model pairs
 PROVIDER_MODEL_DICT = {
     "OpenAI": ["gpt-4o", "gpt-4o-mini"] + O1_MODELS,
@@ -628,7 +645,6 @@ def move_updater():
         "notify_finish": True,
         "temperature": 1,
         "max_tokens": -1,
-        "show_toolbar": True,
         "show_secondary_toolbar": True,
         "top_p": 1,
         "chat_column_to_show": ["id", "name", "insert_dt", "update_dt"],
@@ -662,16 +678,21 @@ def move_updater():
         "llama_index_directory": "",
         "apply_user_defined_styles": False,
         "focus_mode": False,
-        "voice": WHISPER_TTS_DEFAULT_VOICE,
-        "voice_speed": WHISPER_TTS_DEFAULT_SPEED,
         "OPENAI_API_KEY": "",
         "GEMINI_API_KEY": "",
         "CLAUDE_API_KEY": "",
         "LLAMA_API_KEY": "",
+        "show_realtime_api": False,
         # G4F
         "g4f_model": DEFAULT_LLM,
         "provider": G4F_PROVIDER_DEFAULT,
         "g4f_use_chat_history": G4F_USE_CHAT_HISTORY,
+        # STT and TTS settings
+        "voice_provider": TTS_DEFAULT_PROVIDER,
+        "voice": TTS_DEFAULT_VOICE,
+        "voice_speed": TTS_DEFAULT_SPEED,
+        "auto_play_voice": TTS_DEFAULT_AUTO_PLAY,
+        "auto_stop_silence_duration": TTS_DEFAULT_AUTO_STOP_SILENCE_DURATION,
     },
     "DALLE": {
         "quality": "standard",
@@ -682,7 +703,6 @@ def move_updater():
         "width": 1024,
         "height": 1024,
         "prompt_type": 1,
-
         "show_history": True,
         "show_setting": True,
         "prompt": "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
@@ -698,7 +718,6 @@ def move_updater():
         "model": "stability-ai/sdxl:39ed52f2a78e934b3ba6e2a89f5b1c712de7dfea535525255b1aa35c5565e08b",
         "width": 768,
         "height": 768,
-
         "show_history": True,
         "show_setting": True,
         "prompt": "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
@@ -708,12 +727,11 @@ def move_updater():
         "number_of_images_to_create": 2,
         "save_prompt_as_text": True,
         "show_prompt_on_image": False,
-
         "negative_prompt": "ugly, deformed, noisy, blurry, distorted",
     },
     "G4F_IMAGE": {
-        "model": "flux",
-
+        "model": G4F_DEFAULT_IMAGE_MODEL,
+        "provider": G4F_PROVIDER_DEFAULT,
         "show_history": True,
         "show_setting": True,
         "prompt": "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
@@ -723,11 +741,11 @@ def move_updater():
         "number_of_images_to_create": 2,
         "save_prompt_as_text": True,
         "show_prompt_on_image": False,
-
         "negative_prompt": "ugly, deformed, noisy, blurry, distorted",
-    }
+    },
 }
 
+
 # Dynamically add the API keys to the configuration data
 def update_general_config_with_api_keys(config_data, api_configs):
     for config in api_configs:
@@ -737,7 +755,7 @@ def update_general_config_with_api_keys(config_data, api_configs):
 update_general_config_with_api_keys(CONFIG_DATA, DEFAULT_API_CONFIGS)
 
 # Set the default llama index cache directory for preventing any issues such as PermissionError
-os.environ['NLTK_DATA'] = os.path.join(get_config_directory(), "llama_index_cache")
+os.environ["NLTK_DATA"] = os.path.join(get_config_directory(), "llama_index_cache")
 
 # Update the __all__ list with the PEP8 standard dunder names
 __all__ = ["__version__", "__author__"]

pyqt_openai/chat_widget/center/aiChatUnit.py

@@ -110,13 +110,15 @@ def __speak(self, f):
         if f:
             text = self._lbl.toPlainText()
             if text:
+                voice_provider = CONFIG_MANAGER.get_general_property("voice_provider")
+
                 args = {
                     "model": WHISPER_TTS_MODEL,
                     "voice": CONFIG_MANAGER.get_general_property("voice"),
                     "input": text,
                     "speed": CONFIG_MANAGER.get_general_property("voice_speed"),
                 }
-                self.thread = stream_to_speakers(args)
+                self.thread = stream_to_speakers(voice_provider, args)
                 self.thread.finished.connect(self.__on_thread_complete)
                 self.thread.errorGenerated.connect(
                     lambda x: QMessageBox.critical(self, "Error", x)

pyqt_openai/chat_widget/center/commandSuggestionWidget.py

@@ -1,9 +1,12 @@
+from PySide6.QtCore import Signal
 from PySide6.QtWidgets import QListWidget, QWidget, QVBoxLayout, QLabel
 
 from pyqt_openai.lang.translations import LangClass
 
 
 class CommandSuggestionWidget(QWidget):
+    toggleCommandSuggestion = Signal(bool)
+
     def __init__(self, parent=None):
         super().__init__(parent)
         self.__initUi()
@@ -29,3 +32,9 @@ def onCountChanged(self):
         scrollbarHeight = self.__commandList.verticalScrollBar().sizeHint().height()
         totalHeight = contentHeight + itemHeight
         self.setMaximumHeight(totalHeight + scrollbarHeight)
+
+    def showEvent(self, event):
+        self.toggleCommandSuggestion.emit(True)
+
+    def hideEvent(self, event):
+        self.toggleCommandSuggestion.emit(False)

pyqt_openai/chat_widget/center/prompt.py

@@ -47,6 +47,7 @@
     STTThread,
 )
 from pyqt_openai.widgets.button import Button
+from pyqt_openai.widgets.jsonEditor import JSONEditor
 from pyqt_openai.widgets.toolButton import ToolButton
 
 
@@ -86,6 +87,9 @@ def __initUi(self):
 
         # Create the command suggestion list
         self.__suggestionWidget = CommandSuggestionWidget()
+        self.__suggestionWidget.toggleCommandSuggestion.connect(
+            self.__setCommandSuggestionEnabled
+        )
         self.__suggestion_list = self.__suggestionWidget.getCommandList()
 
         self.__uploadedImageFileWidget = UploadedImageFileWidget()
@@ -246,6 +250,13 @@ def __getEveryPromptCommands(self, get_name_only=False):
             return [obj["name"] for obj in self.__p_grp]
         return self.__p_grp
 
+    def __setCommandSuggestionEnabled(self, f):
+        for w in self.__textEditGroup.getGroup().values():
+            if isinstance(w, JSONEditor):
+                pass
+            else:
+                w.setCommandSuggestionEnabled(f)
+
     def __updateSuggestions(self):
         w = self.__textEditGroup.getCurrentTextEdit()
         if w and self.__commandEnabled:
@@ -256,7 +267,6 @@ def __updateSuggestions(self):
                 input_text_chunk = input_text_chunk[-1]
                 starts_with_f = input_text_chunk.startswith("/")
                 self.__suggestionWidget.setVisible(starts_with_f)
-                w.setCommandSuggestionEnabled(starts_with_f)
                 if starts_with_f:
                     command_word = input_text_chunk[1:]
                     # Set every prompt commands first
@@ -299,7 +309,7 @@ def showWidgetInPromptDuringResponse(self, f):
         self.__controlWidgetDuringGeneration.setVisible(f)
 
     def executeCommand(self, item):
-        self.__textEditGroup.executeCommand(item, self.__p_grp)
+        self.__textEditGroup.showPromptContent(item, self.__p_grp)
 
     def updateHeight(self):
         overallHeight = self.__textEditGroup.adjustHeight()
@@ -338,7 +348,6 @@ def __showEnding(self, f):
 
     def __supportPromptCommand(self, f):
         self.__commandEnabled = f
-        self.__textEditGroup.setCommandEnabled(f)
 
     def __readingFiles(self):
         filenames = QFileDialog.getOpenFileNames(

pyqt_openai/chat_widget/center/realtimeApiWidget.py

@@ -0,0 +1,22 @@
+from PySide6.QtWidgets import QWidget, QVBoxLayout, QLabel
+from PySide6.QtGui import QFont
+from PySide6.QtCore import Qt
+
+from pyqt_openai import LARGE_LABEL_PARAM
+
+
+class RealtimeApiWidget(QWidget):
+    def __init__(self, parent=None):
+        super().__init__(parent)
+        self.initUI()
+
+    def initUI(self):
+        lay = QVBoxLayout()
+
+        title = QLabel("Coming Soon...", self)
+        title.setFont(QFont(*LARGE_LABEL_PARAM))
+        title.setAlignment(Qt.AlignmentFlag.AlignCenter)
+
+        lay.addWidget(title)
+
+        self.setLayout(lay)

pyqt_openai/chat_widget/center/textEditPromptGroup.py

@@ -50,7 +50,10 @@ def __initUi(self):
         lay = QVBoxLayout()
         for w in self.__textGroup.values():
             # Connect every group text edit widget to the signal
-            w.textChanged.connect(self.onUpdateSuggestion)
+            if isinstance(w, JSONEditor):
+                pass
+            else:
+                w.textChanged.connect(self.onUpdateSuggestion)
             w.textChanged.connect(self.textChanged)
             w.moveCursorToOtherPrompt.connect(self.__moveCursorToOtherPrompt)
 
@@ -70,7 +73,7 @@ def __initUi(self):
         self.__textGroup[PROMPT_MAIN_KEY_NAME].installEventFilter(self)
         self.__textGroup[PROMPT_MAIN_KEY_NAME].handleDrop.connect(self.handleDrop)
 
-    def executeCommand(self, item, grp):
+    def showPromptContent(self, item, grp):
         command_key = item.text()
         command = ""
         for i in range(len(grp)):
@@ -81,7 +84,7 @@ def executeCommand(self, item, grp):
         if w:
             cursor = w.textCursor()
             cursor.deletePreviousChar()
-            cursor.select(QTextCursor.WordUnderCursor)
+            cursor.select(QTextCursor.SelectionType.WordUnderCursor)
             w.setTextCursor(cursor)
             w.insertPlainText(command)
 
@@ -92,11 +95,6 @@ def getCurrentTextEdit(self):
             if w.hasFocus():
                 return w
 
-    def setCommandEnabled(self, f: bool):
-        for w in self.__textGroup.values():
-            if isinstance(w, TextEditPrompt):
-                w.setCommandSuggestionEnabled(f)
-
     def adjustHeight(self) -> int:
         """
         Adjust overall height of text edit group based on their contents and return adjusted height