Update app.py

app.py

@@ -5,17 +5,7 @@ from transformers import BarkModel
 from optimum.bettertransformer import BetterTransformer
 
 bark_model = BarkModel.from_pretrained("suno/bark", torch_dtype=torch.float16)
-device = "cuda:0" if torch.cuda.is_available() else "cpu"
-bark_model = bark_model.to(device)
 
-from transformers import AutoProcessor
-processor = AutoProcessor.from_pretrained("suno/bark")
-
-# Use bettertransform for flash attention
-bark_model = BetterTransformer.transform(bark_model, keep_original_model=False)
-
-# Enable CPU offload
-bark_model.enable_cpu_offload()
 
 from TTS.tts.configs.bark_config import BarkConfig
 from TTS.tts.models.bark import Bark