Update handler.py

handler.py

@@ -11,7 +11,7 @@ class EndpointHandler():
         self.model = AutoModelForSpeechSeq2Seq.from_pretrained(
             model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
         )
-        model.to(device)
+        self.model.to(device)
 
         processor = AutoProcessor.from_pretrained(model_id)