add definitions

app.py

@@ -749,10 +749,11 @@ with gr.Blocks(title=title) as demo:
     gr.Markdown(
         """
 This Space demonstrates how to speak to a chatbot, based solely on open-source models.
-It relies on 3 models:
-1. [Whisper-large-v2](https://sanchit-gandhi-whisper-large-v2.hf.space/) as an ASR model, to transcribe recorded audio to text. It is called through a [gradio client](https://www.gradio.app/docs/client).
-2. [Zephyr-7b-alpha](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha) as the chat model. GGUF Q5_K_M quantized version used locally via llama_cpp from [huggingface.co/TheBloke](https://huggingface.co/TheBloke/zephyr-7B-alpha-GGUF).
-3. [Coqui's XTTS](https://huggingface.co/spaces/coqui/xtts) as a TTS model, to generate the chatbot answers. This time, the model is hosted locally.
+It relies on 3 stage models:
+Speech to Text : [Whisper-large-v2](https://sanchit-gandhi-whisper-large-v2.hf.space/) as an ASR model, to transcribe recorded audio to text. It is called through a [gradio client](https://www.gradio.app/docs/client).
+LLM Model      : [Mistral-7b-instruct](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) as the chat model, GGUF Q5_K_M quantized version used locally via llama_cpp[huggingface_hub](TheBloke/Mistral-7B-Instruct-v0.1-GGUF).
+                 With LLM_MODEL="zephyr" it can use  [Zephyr-7b-alpha](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha) as the chat model. GGUF Q5_K_M quantized version used locally via llama_cpp from [huggingface.co/TheBloke](https://huggingface.co/TheBloke/zephyr-7B-alpha-GGUF).
+Text to Speech : [Coqui's XTTS](https://huggingface.co/spaces/coqui/xtts) as a Multilingual TTS model, to generate the chatbot answers. This time, the model is hosted locally.
 
 Note:
 - By using this demo you agree to the terms of the Coqui Public Model License at https://coqui.ai/cpml