add app files

app.py

@@ -46,9 +46,10 @@ whisper_model = WhisperModel("large-v3", device="cuda", compute_type="float16")
 
 # Load Mistral LLM
 print("Loading Mistral LLM")
-llm_model_name="mistral-7b-instruct-v0.1.Q5_K_M.gguf"
-hf_hub_download(repo_id="TheBloke/Mistral-7B-Instruct-v0.1-GGUF", local_dir=".", filename=llm_model_name)
-mistral_llm = Llama(model_path=f"./{llm_model_name}",n_gpu_layers=35,max_new_tokens=256, context_window=4096, n_ctx=4096,n_batch=128,verbose=False)
+hf_hub_download(repo_id="TheBloke/Mistral-7B-Instruct-v0.1-GGUF", local_dir=".", filename="mistral-7b-instruct-v0.1.Q5_K_M.gguf")
+mistral_model_path="./mistral-7b-instruct-v0.1.Q5_K_M.gguf"
+mistral_llm = Llama(model_path=mistral_model_path,n_gpu_layers=35,max_new_tokens=256, context_window=4096, n_ctx=4096,n_batch=128,verbose=False)
+
 
 # Load XTTS Model
 print("Loading XTTS model")
@@ -177,4 +178,4 @@ with gr.Blocks(title="Voice chat with LLM") as demo:
             - iOS (Iphone/Ipad) devices may not experience voice due to autoplay being disabled on these devices by Vendor"""
     gr.Markdown(FOOTNOTE)
     demo.load(block=None, fn=generate_speech, inputs=[chatbot,chatbot_voice, gr.State(value=True)], outputs=[sentence, chatbot, audio_playback])
-demo.queue().launch(debug=True,share=True)
+demo.queue().launch(debug=True,share=True)