Update app.py

app.py

@@ -20,7 +20,7 @@ bark_processor = BarkProcessor.from_pretrained("suno/bark")
 
 
 def translate(audio):
-    inputs = processor(audio, sampling_rate=SAMPLE_RATE, return_tensors="pt")
+    inputs = asr_processor(audio, sampling_rate=SAMPLE_RATE, return_tensors="pt")
     generated_ids = asr_model.generate(inputs["input_features"],attention_mask=inputs["attention_mask"],
     forced_bos_token_id=asr_processor.tokenizer.lang_code_to_id["it"],)
     translation = processor.batch_decode(generated_ids, skip_special_tokens=True)