Update app.py

app.py

@@ -81,8 +81,8 @@ def generate(audio_text, history, temperature=None, max_new_tokens=512, top_p=0.
     return response, audio_file
 
 def text_to_speech(text):
-    inputs = processor(text=text, return_tensors="pt")
-    
+    inputs = processor(text=text, return_tensors="pt", target_language="es")
+
     embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
     speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
     speech = model.generate_speech(inputs.input_ids, speaker_embeddings, vocoder=vocoder)