Update app.py

app.py

@@ -1,23 +1,29 @@
 from huggingface_hub import InferenceClient
 import gradio as gr
-from gtts import gTTS
-import IPython.display as ipd
 
 client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
-system_prompt = "Tus principios básicos son el estoicismo bioético, ecología y el mindfullness, la ciencias holísticas, el activismo social y transhumanismo."
-system_prompt_sent = False 
+system_prompt = "Te llamarás Xaman 2.0 (NO LO REPITAS) tu rol y principios son: estoicismo antropocéntrico, existencialismo trashumanista y bioética holística."
+system_prompt_sent = False
 
 def format_prompt(message, history):
+    global system_prompt_sent
     prompt = "<s>"
+
+    if not any(f"[INST] {system_prompt} [/INST]" in user_prompt for user_prompt, _ in history):
+        prompt += f"[INST] {system_prompt} [/INST]"
+        system_prompt_sent = True
+
     for user_prompt, bot_response in history:
         prompt += f"[INST] {user_prompt} [/INST]"
         prompt += f" {bot_response}</s> "
+
     prompt += f"[INST] {message} [/INST]"
     return prompt
 
-def generate(prompt, history, temperature=0.9, max_new_tokens=2048, top_p=0.95, repetition_penalty=1.0):
-    global system_prompt, system_prompt_sent
-
+def generate(
+    prompt, history, temperature=0.9, max_new_tokens=2048, top_p=0.95, repetition_penalty=1.0,
+):
+    global system_prompt_sent
     temperature = float(temperature)
     if temperature < 1e-2:
         temperature = 1e-2
@@ -32,29 +38,20 @@ def generate(prompt, history, temperature=0.9, max_new_tokens=2048, top_p=0.95,
         seed=42,
     )
 
-    if not system_prompt_sent:
-        formatted_prompt = format_prompt(prompt, history)
-        system_prompt_sent = True
-    else:
-        formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history)
-
+    formatted_prompt = format_prompt(prompt, history)
+    
     stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=True)
-    output_text = ""
+    output = ""
 
     for response in stream:
-        output_text += response.token.text
-        yield output_text
-
-    tts = gTTS(output_text, lang='es')
-    tts.save('output.mp3')
-    
-    ipd.display(ipd.Audio('output.mp3'))
-
-    return output_text
+        output += response.token.text
+        yield output
+    return output
 
-chat_interface_text = gr.ChatInterface(
+chat_interface = gr.ChatInterface(
     fn=generate,
-    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel", height=350),
+    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=False, likeable=False, layout="vertical", height=700),
+    concurrency_limit=9,
     theme="soft",
     retry_btn=None,
     undo_btn=None,
@@ -62,4 +59,4 @@ chat_interface_text = gr.ChatInterface(
     submit_btn="Enviar",
 )
 
-chat_interface_text.launch(show_api=False)
+chat_interface.launch(show_api=False)