Add functions to export and import the model state for a given voice (#129)

pocket_tts/__init__.py

@@ -3,7 +3,10 @@ from beartype.claw import beartype_this_package
 
 beartype_this_package(conf=BeartypeConf(is_color=False))
 
-from pocket_tts.models.tts_model import TTSModel  # noqa: E402
+from pocket_tts.models.tts_model import (  # noqa: E402
+    TTSModel,
+    export_model_state,
+)
 
 # Public methods:
 # TTSModel.device
@@ -13,4 +16,4 @@ from pocket_tts.models.tts_model import TTSModel  # noqa: E402
 # TTSModel.generate_audio_stream
 # TTSModel.get_state_for_audio_prompt
 
-__all__ = ["TTSModel"]
+__all__ = ["TTSModel", "export_model_state"]