Fix VITS multispeaker recipe

recipes/vctk/vits/train_vits.py

@@ -5,7 +5,7 @@ from TTS.trainer import Trainer, TrainingArgs
 from TTS.tts.configs.shared_configs import BaseDatasetConfig
 from TTS.tts.configs.vits_config import VitsConfig
 from TTS.tts.datasets import load_tts_samples
-from TTS.tts.models.vits import Vits
+from TTS.tts.models.vits import Vits, VitsArgs
 from TTS.tts.utils.speakers import SpeakerManager
 from TTS.utils.audio import AudioProcessor
 
@@ -31,10 +31,14 @@ audio_config = BaseAudioConfig(
     resample=True,
 )
 
+vitsArgs = VitsArgs(
+    use_speaker_embedding=True,
+)
+
 config = VitsConfig(
+    model_args=vitsArgs,
     audio=audio_config,
     run_name="vits_vctk",
-    use_speaker_embedding=True,
     batch_size=32,
     eval_batch_size=16,
     batch_group_size=5,