wip: add tts debug

pyproject.toml

@@ -18,6 +18,10 @@ dependencies = [
     "pyinstaller>=6.11.0",
     "twitchwebsocket>=1.2.1",
     "loguru>=0.7.2",
+    "ffmpeg>=1.4",
+    "ffmpeg-python>=0.2.0",
+    "simpleaudio>=1.0.4",
+    "coqui-tts>=0.24.3",
 ]
 
 [tool.uv]
@@ -44,7 +48,9 @@ module = [
     "nltk.tokenize.treebank",
     "nltk.tokenize.destructive",
     "TwitchWebsocket",
-    "tokenizer"
+    "tokenizer",
+    "TTS.api",
+    "simpleaudio"
 ]
 ignore_missing_imports = true
 

src/markovbot_gui/main.py

@@ -8,6 +8,7 @@ from kivy.uix.widget import Widget
 
 from src.markovbot_gui.bot_runner import BotRunner
 from src.markovbot_gui.config_window import ConfigWindow
+from src.markovbot_gui.tts_window import TTSWindow
 
 
 class BotApp(App):
@@ -51,6 +52,29 @@ class BotApp(App):
 
         popup.open()
 
+    def run_tts(self, instance):
+        """Create a TTS window and open it"""
+        tts_window = TTSWindow(config_path=self.config_path)
+        popup = Popup(
+            title=f"TTS configuration, available at {self.config_path}",
+            content=tts_window,
+            size_hint=(None, None),
+            size=(dp(600), dp(400)),
+            auto_dismiss=False,
+        )
+
+        # Add close button
+        close_button = Button(
+            text="Close",
+            size_hint=(None, None),
+            size=(dp(100), dp(40)),
+            pos_hint={"center_x": 0.5},
+        )
+        close_button.bind(on_release=popup.dismiss)
+        tts_window.add_widget(close_button)
+
+        popup.open()
+
     def build(self):
         widget = Widget()
 
@@ -60,6 +84,10 @@ class BotApp(App):
         run_button.bind(on_release=self.run_bot)
         layout.add_widget(run_button)
 
+        tts_button = Button(text="TTS button")
+        tts_button.bind(on_release=self.run_tts)
+        layout.add_widget(tts_button)
+
         config_button = Button(text="Open config")
         config_button.bind(on_release=self.run_config)
         layout.add_widget(config_button)

src/markovbot_gui/tts_window.py

@@ -0,0 +1,48 @@
+from pathlib import Path
+
+from kivy.metrics import dp
+from kivy.uix.boxlayout import BoxLayout
+from kivy.uix.button import Button
+from kivy.uix.textinput import TextInput
+from TTS.api import TTS
+
+
+class TTSWindow(BoxLayout):
+    def __init__(self, config_path: Path, **kwargs):
+        super().__init__(**kwargs)
+        self.config_path = config_path
+        self.orientation = "vertical"
+
+        self.spacing = dp(10)
+        self.padding = dp(20)
+
+        # create a text input for the text to be converted to audio,
+        # and a button to execute the conversion
+
+        text_layout = BoxLayout(orientation="vertical")
+        self.add_widget(text_layout)
+
+        self.text_input = TextInput(multiline=True)
+        text_layout.add_widget(self.text_input)
+
+        tts_button = Button(text="TTS button")
+        tts_button.bind(on_release=self.run_tts)
+        text_layout.add_widget(tts_button)
+
+    def run_tts(self, instance):
+        device = "cpu"
+        model_name = "tts_models/multilingual/multi-dataset/xtts_v2"
+        tts = TTS(model_name).to(device)
+        tts.download_model_by_name(model_name)
+
+        file = "output.wav"
+        tts.tts_to_file(
+            text=self.text_input.text.strip(),
+            speaker="Rosemary Okafor",
+            language="es",
+            file_path=file,
+        )
+
+        import simpleaudio as sa
+
+        sa.WaveObject.from_wave_file(file).play()