Update app.py

app.py

@@ -33,9 +33,9 @@ def synthesise(text):
     inputs = tokenizer(text=text, return_tensors="pt")
     # speech = model.generate_speech(inputs["input_ids"].to(device), speaker_embeddings.to(device), vocoder=vocoder)
     with torch.no_grad():
-        outputs = model(inputs['input_ids'])
+        tts_output = model(inputs['input_ids'].to(device))
 
-    speech = outputs['waveform']
+    speech = tts_output['waveform']
     return speech.cpu()