fix

app.py

@@ -22,7 +22,7 @@ def process_inputs(image, audio):
         ]
     },]
 
-    input_ids = self.processor.apply_chat_template(
+    input_ids = processor.apply_chat_template(
         messages,
         add_generation_prompt=True,
         tokenize=True,
@@ -31,14 +31,14 @@ def process_inputs(image, audio):
     )
     input_len = input_ids["input_ids"].shape[-1]
 
-    input_ids = input_ids.to(self.model.device, dtype=model.dtype)
+    input_ids = input_ids.to(model.device, dtype=model.dtype)
     with torch.inference_mode:
-        outputs = self.model.generate(
+        outputs = model.generate(
             **input_ids,
             max_new_tokens=max_tokens,
             disable_compile=True
         )
-    text = self.processor.batch_decode(
+    text = processor.batch_decode(
         outputs[:, input_len:],
         skip_special_tokens=True,
         clean_up_tokenization_spaces=True