Bugfix fixtts (#4902)

* changing configs names to tts_provider

* accidently triggered circular importing.

* added config to places it needs other than the logger

* got it to work on windows

* did all the formatting stuff

---------

Co-authored-by: James Collins <collijk@uw.edu>

autogpt/agent/agent.py

@@ -165,7 +165,7 @@ class Agent:
                         assistant_reply_json, assistant_reply, self.config
                     )
                     if self.config.speak_mode:
-                        say_text(f"I want to execute {command_name}")
+                        say_text(f"I want to execute {command_name}", self.config)
 
                     arguments = self._resolve_pathlike_command_args(arguments)
 

autogpt/logs.py

@@ -16,7 +16,6 @@ if TYPE_CHECKING:
 
 from autogpt.log_cycle.json_handler import JsonFileHandler, JsonFormatter
 from autogpt.singleton import Singleton
-from autogpt.speech import say_text
 
 
 class Logger(metaclass=Singleton):
@@ -87,13 +86,16 @@ class Logger(metaclass=Singleton):
         self.json_logger.setLevel(logging.DEBUG)
 
         self.speak_mode = False
+        self.config = None
         self.chat_plugins = []
 
     def typewriter_log(
         self, title="", title_color="", content="", speak_text=False, level=logging.INFO
     ):
+        from autogpt.speech import say_text
+
         if speak_text and self.speak_mode:
-            say_text(f"{title}. {content}")
+            say_text(f"{title}. {content}", self.config)
 
         for plugin in self.chat_plugins:
             plugin.report(f"{title}. {content}")
@@ -265,6 +267,8 @@ def print_assistant_thoughts(
     assistant_reply_json_valid: object,
     config: Config,
 ) -> None:
+    from autogpt.speech import say_text
+
     assistant_thoughts_reasoning = None
     assistant_thoughts_plan = None
     assistant_thoughts_speak = None

autogpt/main.py

@@ -58,6 +58,9 @@ def run_auto_gpt(
     logger.speak_mode = speak
 
     config = ConfigBuilder.build_config_from_env()
+    # HACK: This is a hack to allow the config into the logger without having to pass it around everywhere
+    # or import it directly.
+    logger.config = config
 
     # TODO: fill in llm values here
     check_openai_api_key(config)

autogpt/speech/base.py

@@ -45,7 +45,7 @@ class VoiceBase(AbstractSingleton):
             return self._speech(text, voice_index)
 
     @abc.abstractmethod
-    def _setup(self) -> None:
+    def _setup(self, config: Config) -> None:
         """
         Setup the voices, API key, etc.
         """

autogpt/speech/gtts.py

@@ -4,13 +4,14 @@ import os
 import gtts
 from playsound import playsound
 
+from autogpt.config import Config
 from autogpt.speech.base import VoiceBase
 
 
 class GTTSVoice(VoiceBase):
     """GTTS Voice."""
 
-    def _setup(self) -> None:
+    def _setup(self, config: Config) -> None:
         pass
 
     def _speech(self, text: str, _: int = 0) -> bool:

autogpt/speech/macos_tts.py

@@ -1,13 +1,14 @@
 """ MacOS TTS Voice. """
 import os
 
+from autogpt.config import Config
 from autogpt.speech.base import VoiceBase
 
 
 class MacOSTTS(VoiceBase):
     """MacOS TTS Voice."""
 
-    def _setup(self) -> None:
+    def _setup(self, config: Config) -> None:
         pass
 
     def _speech(self, text: str, voice_index: int = 0) -> bool:

autogpt/speech/say.py

@@ -41,10 +41,10 @@ def _get_voice_engine(config: Config) -> tuple[VoiceBase, VoiceBase]:
     if tts_provider == "elevenlabs":
         voice_engine = ElevenLabsSpeech(config)
     elif tts_provider == "macos":
-        voice_engine = MacOSTTS()
+        voice_engine = MacOSTTS(config)
     elif tts_provider == "streamelements":
-        voice_engine = StreamElementsSpeech()
+        voice_engine = StreamElementsSpeech(config)
     else:
-        voice_engine = GTTSVoice()
+        voice_engine = GTTSVoice(config)
 
-    return GTTSVoice(), voice_engine
+    return GTTSVoice(config), voice_engine

autogpt/speech/stream_elements_speech.py

@@ -4,13 +4,14 @@ import os
 import requests
 from playsound import playsound
 
+from autogpt.config import Config
 from autogpt.speech.base import VoiceBase
 
 
 class StreamElementsSpeech(VoiceBase):
     """Streamelements speech module for autogpt"""
 
-    def _setup(self) -> None:
+    def _setup(self, config: Config) -> None:
         """Setup the voices, API key, etc."""
 
     def _speech(self, text: str, voice: str, _: int = 0) -> bool: