Update app.py

app.py

@@ -1,55 +1,54 @@
-import streamlit as st
-from transformers import pipeline
-
-# Title of the application
-st.title("AI Code Bot")
-
-# Introduction
-st.write("""
-### Generate Code Snippets with AI
-Enter a programming-related prompt, and the AI will generate a code snippet with explanatory comments.
-""")
-
-# Input box for user query
-user_input = st.text_area("Enter your prompt (e.g., 'Write a Python function to reverse a list'):")
-
-# Load the Hugging Face pipeline (code generation)
-@st.cache_resource
-def load_model():
-    # Authenticate automatically in Hugging Face Spaces
-    model = pipeline(
-        "text-generation",
-        model="EleutherAI/gpt-neo-1.3B"  # The model to use
+import os
+import whisper
+from gtts import gTTS
+from groq import Groq
+import gradio as gr
+from tempfile import NamedTemporaryFile
+
+# Set up Groq client
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
+
+# Load OpenAI Whisper model
+whisper_model = whisper.load_model("base")
+
+def voice_to_voice_chat(audio_file):
+    # Step 1: Transcribe Audio Input
+    transcription = whisper_model.transcribe(audio_file)["text"]
+
+    # Step 2: Query Groq's LLM
+    chat_completion = client.chat.completions.create(
+        messages=[
+            {"role": "user", "content": transcription}
+        ],
+        model="llama3-8b-8192",
+        stream=False,
     )
-    return model
-
-model = load_model()
-
-# When the user clicks the button
-if st.button("Generate Code"):
-    if user_input.strip() == "":
-        st.warning("Please enter a prompt.")
-    else:
-        with st.spinner("Generating code..."):
-            try:
-                # Generate code using the model
-                result = model(user_input, max_length=200, num_return_sequences=1)
-                generated_text = result[0]["generated_text"]
-                
-                # Split comments and code for better presentation
-                lines = generated_text.split("\n")
-                comments = [line for line in lines if line.strip().startswith("#")]
-                code = [line for line in lines if not line.strip().startswith("#")]
-
-                # Display comments as markdown
-                if comments:
-                    st.subheader("Comments")
-                    for comment in comments:
-                        st.markdown(comment)
-
-                # Display code as Python syntax-highlighted block
-                if code:
-                    st.subheader("Code")
-                    st.code("\n".join(code), language="python")
-            except Exception as e:
-                st.error(f"An error occurred: {e}")
+    llm_response = chat_completion.choices[0].message.content
+
+    # Step 3: Convert LLM Response to Audio
+    tts = gTTS(llm_response)
+    audio_output = NamedTemporaryFile(delete=False, suffix=".mp3")
+    tts.save(audio_output.name)
+
+    return llm_response, audio_output.name
+
+# Gradio Interface
+def chatbot_interface(audio_input):
+    response, audio_response_file = voice_to_voice_chat(audio_input)
+    return response, audio_response_file
+
+# Build Gradio App
+interface = gr.Interface(
+    fn=chatbot_interface,
+    inputs=gr.Audio(source="microphone", type="filepath", label="Speak into the Microphone"),
+    outputs=[
+        gr.Textbox(label="Chatbot Response"),
+        gr.Audio(type="filepath", label="Voice Output")
+    ],
+    title="Real-Time Voice-to-Voice Chatbot",
+    description="Speak to the chatbot and get a spoken response!",
+)
+
+# Launch Interface
+if __name__ == "__main__":
+    interface.launch()