Update app.py

app.py

@@ -15,6 +15,9 @@ from uuid import uuid4
 import re
 import matplotlib.pyplot as plt
 import spaces
+from elevenlabs.client import ElevenLabs
+from elevenlabs import stream
+import base64
 
 from rdkit import Chem
 from rdkit.Chem import AllChem, QED
@@ -702,6 +705,30 @@ def clear_history():
   global chat_history
   chat_history = []
 
+def get_speech():
+   elita_text = "Hi, I'm Modrag! Let's design together!"
+
+  voice_settings = {
+          "stability": 0.37,
+          "similarity_boost": 0.90,
+          "style": 0.0,
+          "speed": 1.05
+      }
+
+  audio_stream = elevenlabs.text_to_speech.convert(
+      text = elita_text,    
+      voice_id = 'vxO9F6g9yqYJ4RsWvMbc',
+      model_id = 'eleven_multilingual_v2',
+      output_format='mp3_44100_128',
+      voice_settings=voice_settings
+  )
+
+  audio_converted = b"".join(audio_stream)
+  audio = base64.b64encode(audio_converted).decode("utf-8")
+  audio_player = f'<audio src="data:audio/mpeg;base64,{audio}" controls autoplay></audio>'
+
+  return audio_player
+
 with gr.Blocks() as forest:
   gr.Markdown('''
               # MoDrAg - the *Mo*dular *Dr*ug Design *Ag*ent!
@@ -743,10 +770,13 @@ with gr.Blocks() as forest:
   chat_btn = gr.Button(value = "Send")
   
   clear = gr.ClearButton([task])
+  modrags_voice = gr.HTML() 
   pic = gr.Image(label="Molecules (if needed)")
-  
+    
   chat_btn.click(DDAgent, inputs = [task], outputs = [task, chatbot, pic])
   task.submit(DDAgent, [task], [task, chatbot, pic])
   clear.click(clear_history)
+  modrags_voice.release(get_speech, outputs=[modrags_voice])
+  
 
 forest.launch(debug=False, mcp_server=True)