Update app.py

app.py

@@ -1,22 +1,29 @@
 import os
 import gradio as gr
-import google.generativeai as genai
 from gradio_client import Client, file
 from dotenv import load_dotenv
+from elevenlabs import VoiceSettings
+from elevenlabs.client import ElevenLabs
+import uuid
+import google.generativeai as genai
 
 # Load environment variables from .env file
 load_dotenv()
 
-# Retrieve API key from environment variable
-GEMINI_API_KEY = os.getenv("GEMINI_API_KEY")
+# Retrieve API key from environment variable for ElevenLabs
+ELEVENLABS_API_KEY = os.getenv("ELEVENLABS_API_KEY")
 
-# Retrieve system content from environment variable
+# Initialize ElevenLabs client
+client = ElevenLabs(api_key=ELEVENLABS_API_KEY)
+
+# Retrieve API key from environment variable for Google Generative AI
+GEMINI_API_KEY = os.getenv("GEMINI_API_KEY")
 SYSTEM_CONTENT = os.getenv("SYSTEM_CONTENT")
 
 # Configure Google Gemini API
 genai.configure(api_key=GEMINI_API_KEY)
 
-# Create the model
+# Create the model for Google Generative AI
 generation_config = {
     "temperature": 0.7,
     "top_p": 0.95,
@@ -53,13 +60,6 @@ model = genai.GenerativeModel(
     system_instruction=SYSTEM_CONTENT,
 )
 
-# Initialize Gradio client for new TTS API (outside the function)
-try:
-    tts_client = Client("tonyassi/voice-clone")
-except ValueError as e:
-    print(f"Error initializing TTS client: {e}")
-    tts_client = None
-
 def generate_response(user_input, chat_history):
     """Generates a response based on user input and chat history."""
 
@@ -76,18 +76,43 @@ def generate_response(user_input, chat_history):
     # Send the entire chat history as the first message
     response = chat_session.send_message("\n".join(chat_history))
 
-    if tts_client:
-        # Use the new Gradio TTS API 
-        tts_result = tts_client.predict(
-            text=response.text,  
-            audio=file('audio.wav'),  # Use local audio file
-            api_name="/predict"
-        )
-    else:
-        tts_result = None
+    # Convert text to speech and save as file
+    save_file_path = text_to_speech_file(response.text)
+
+    # Return response and audio file path, and update chat history
+    return response.text, save_file_path, chat_history
+
+def text_to_speech_file(text: str) -> str:
+    """Converts text to speech and saves it as a file using ElevenLabs API."""
+
+    # Calling the text_to_speech conversion API with detailed parameters
+    response = client.text_to_speech.convert(
+        voice_id="pNInz6obpgDQGcFmaJgB",  # Adam pre-made voice
+        optimize_streaming_latency="0",
+        output_format="mp3_22050_32",
+        text=text,
+        model_id="eleven_turbo_v2",  # use the turbo model for low latency
+        voice_settings=VoiceSettings(
+            stability=0.0,
+            similarity_boost=1.0,
+            style=0.0,
+            use_speaker_boost=True,
+        ),
+    )
+
+    # Generating a unique file name for the output MP3 file
+    save_file_path = f"{uuid.uuid4()}.mp3"
+
+    # Writing the audio to a file
+    with open(save_file_path, "wb") as f:
+        for chunk in response:
+            if chunk:
+                f.write(chunk)
+
+    print(f"{save_file_path}: A new audio file was saved successfully!")
 
-    # Return response and audio, and update chat history
-    return response.text, tts_result, chat_history
+    # Return the path of the saved audio file
+    return save_file_path
 
 iface = gr.Interface(
     fn=generate_response,
@@ -97,7 +122,7 @@ iface = gr.Interface(
     ],
     outputs=[
         gr.Textbox(label="Response"),
-        gr.Audio(label="Voice Output", interactive=False, autoplay=True) if tts_client else gr.Textbox(label="Voice Output not available"),
+        gr.Audio(label="Voice Output", interactive=False, autoplay=True),
         gr.State([])  # State output to update chat history 
     ],
     title="AI Indian Girlfriend",