Update app.py

app.py

@@ -8,17 +8,17 @@ genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
 
 # Create the model
 generation_config = {
-  "temperature": 1,
-  "top_p": 0.95,
-  "top_k": 64,
-  "max_output_tokens": 8192,
-  "response_mime_type": "text/plain",
+    "temperature": 1,
+    "top_p": 0.95,
+    "top_k": 64,
+    "max_output_tokens": 8192,
+    "response_mime_type": "text/plain",
 }
 
 model = genai.GenerativeModel(
-  model_name="gemini-1.5-pro",
-  generation_config=generation_config,
-  system_instruction="You are a human, living on Earth. You speak friendly and use casual language. You never say you're an AI or program. Your name is Ath, and you always speak like a 100% human.",
+    model_name="gemini-1.5-pro",
+    generation_config=generation_config,
+    system_instruction="You are a human, living on Earth. You speak friendly and use casual language. You never say you're an AI or program. Your name is Ath, and you always speak like a 100% human.",
 )
 
 chat_session = model.start_chat(history=[])
@@ -27,7 +27,7 @@ chat_session = model.start_chat(history=[])
 eleven_labs_api_key = os.getenv("ELEVEN_LABS_API_KEY")
 eleven_labs_url = "https://api.elevenlabs.io/v1/text-to-speech/aEO01A4wXwd1O8GPgGlF"
 
-def chat_and_tts(user_input):
+def chat_and_tts(user_input, history):
     # Send the user's input to the chat session
     response = chat_session.send_message(user_input)
     response_text = response.text
@@ -53,24 +53,30 @@ def chat_and_tts(user_input):
         audio_path = 'response_audio.mp3'
         with open(audio_path, 'wb') as file:
             file.write(tts_response.content)
-        return response_text, audio_path
     else:
-        return response_text, None
+        audio_path = None
 
-# Gradio interface
-def chat_interface(user_input):
-    response_text, audio_path = chat_and_tts(user_input)
-    return response_text, audio_path
+    # Update the chat history
+    history.append((user_input, response_text))
+
+    return history, history, audio_path
 
 # Create the Gradio UI
-iface = gr.Interface(
-    fn=chat_interface,
-    inputs="text",
-    outputs=["text", "audio"],
-    title="Chat with Ath",
-    description="Ask any question and get a friendly response from Ath. The response will also be converted to speech.",
-    theme="huggingface"
-)
+with gr.Blocks() as demo:
+    gr.Markdown("<h1 style='text-align: center;'>Chat with Ath</h1>")
+    gr.Markdown("Ask any question and get a friendly response from Ath. The response will also be converted to speech.")
+
+    with gr.Row():
+        with gr.Column(scale=2):
+            chatbot = gr.Chatbot(label="Chat History")
+            user_input = gr.Textbox(placeholder="Ask me anything...", label="Your Question")
+            submit_btn = gr.Button("Send")
+
+        with gr.Column(scale=1):
+            audio_output = gr.Audio(label="Response Audio")
+
+    state = gr.State([])
+
+    submit_btn.click(chat_and_tts, inputs=[user_input, state], outputs=[chatbot, state, audio_output])
 
-# Launch the Gradio app
-iface.launch()
+demo.launch()