fix(xtts): overwrite model_args in GPTTrainerConfig

This makes it possible to use --continue_path with XTTS

TTS/tts/layers/xtts/trainer/gpt_trainer.py

@@ -23,16 +23,6 @@
 logger = logging.getLogger(__name__)
 
 
-@dataclass
-class GPTTrainerConfig(XttsConfig):
-    lr: float = 5e-06
-    training_seed: int = 1
-    optimizer_wd_only_on_weights: bool = False
-    weighted_loss_attrs: dict = field(default_factory=lambda: {})
-    weighted_loss_multipliers: dict = field(default_factory=lambda: {})
-    test_sentences: list[dict] = field(default_factory=lambda: [])
-
-
 @dataclass
 class GPTArgs(XttsArgs):
     min_conditioning_length: int = 66150
@@ -51,6 +41,17 @@ class GPTArgs(XttsArgs):
     vocoder: str = ""  # overide vocoder key on the config to avoid json write issues
 
 
+@dataclass
+class GPTTrainerConfig(XttsConfig):
+    lr: float = 5e-06
+    training_seed: int = 1
+    optimizer_wd_only_on_weights: bool = False
+    weighted_loss_attrs: dict = field(default_factory=lambda: {})
+    weighted_loss_multipliers: dict = field(default_factory=lambda: {})
+    test_sentences: list[dict] = field(default_factory=lambda: [])
+    model_args: GPTArgs = field(default_factory=GPTArgs)
+
+
 def callback_clearml_load_save(operation_type, model_info):
     # return None means skip the file upload/log, returning model_info will continue with the log/upload
     # you can also change the upload destination file name model_info.upload_filename or check the local file size with Path(model_info.local_model_path).stat().st_size