Update app.py

app.py

@@ -20,7 +20,7 @@ vq_model = RQBottleneckTransformer.load_model(
         "whisper-vq-stoks-medium-en+pl-fixed.model"
     ).to(device)
 # tts = TTSProcessor('cpu')
-use_8bit = False    
+use_8bit = True   
 llm_path = "akjindal53244/Llama-3.1-Storm-8B"
 tokenizer = AutoTokenizer.from_pretrained(llm_path)
 model_kwargs = {}