Update app.py

app.py

@@ -1,91 +1,39 @@
 import os
-import streamlit as st
+import gradio as gr
 import whisper
 from gtts import gTTS
-from dotenv import load_dotenv
+from groq import Groq
 import tempfile
-from streamlit_webrtc import webrtc_streamer, WebRtcMode, AudioProcessorBase, ClientSettings
-import numpy as np
-import wave
 
-# Load environment variables
-load_dotenv()
-
-# Load Whisper model
-st.write("Loading Whisper model...")
+# Load Whisper model for speech-to-text
+print("Loading Whisper model...")
 whisper_model = whisper.load_model("base")
 
-# API Key for LLM
-api_key = os.getenv("GROQ_API_KEY")
-if not api_key:
-    raise ValueError("API key is missing. Set GROQ_API_KEY in your .env file.")
+# Initialize Groq API
+print("Initializing Groq API...")
+client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
 
 # Function to transcribe audio to text
 def transcribe_audio(audio_file):
-    result = whisper_model.transcribe(audio_file)
-    return result['text']
+    try:
+        result = whisper_model.transcribe(audio_file)
+        return result["text"]
+    except Exception as e:
+        return f"Error in transcription: {e}"
 
-# Simulated response from LLM (Replace with actual API call if available)
+# Function to get response from LLM using Groq API
 def get_llm_response(user_input):
-    # Replace with your actual LLM call logic here
-    return f"Response to: {user_input}"
-
-# Convert text to speech
-def text_to_speech(text):
-    tts = gTTS(text)
-    temp_file = tempfile.NamedTemporaryFile(suffix=".mp3", delete=False)
-    tts.save(temp_file.name)
-    return temp_file.name
-
-# AudioProcessor class to capture audio from the microphone
-class AudioProcessor(AudioProcessorBase):
-    def __init__(self):
-        self.frames = []
-
-    def recv(self, frame):
-        self.frames.append(frame.to_ndarray().flatten())
-        return frame
-
-    def save_audio(self, path):
-        with wave.open(path, "wb") as wf:
-            wf.setnchannels(1)  # Mono
-            wf.setsampwidth(2)  # 16-bit samples
-            wf.setframerate(16000)  # 16kHz
-            wf.writeframes(np.concatenate(self.frames).tobytes())
-
-# Streamlit UI
-st.title("Real-Time Voice Chatbot")
-st.write("Interact with the chatbot using your voice.")
-
-webrtc_ctx = webrtc_streamer(
-    key="example",
-    mode=WebRtcMode.SENDONLY,
-    audio_processor_factory=AudioProcessor,
-    client_settings=ClientSettings(
-        rtc_configuration={"iceServers": [{"urls": ["stun:stun.l.google.com:19302"]}]},
-        media_stream_constraints={"audio": True, "video": False},
-    ),
-)
-
-if webrtc_ctx.audio_processor:
-    audio_processor = webrtc_ctx.audio_processor
-
-    if st.button("Process Microphone Input"):
-        if audio_processor:
-            audio_path = tempfile.NamedTemporaryFile(suffix=".wav", delete=False).name
-            audio_processor.save_audio(audio_path)
-
-            # Step 1: Transcribe
-            with st.spinner("Transcribing audio..."):
-                user_input = transcribe_audio(audio_path)
-                st.write(f"**You said:** {user_input}")
-
-            # Step 2: Get LLM response
-            with st.spinner("Generating response..."):
-                response = get_llm_response(user_input)
-                st.write(f"**Chatbot Response:** {response}")
-
-            # Step 3: Text to speech
-            with st.spinner("Converting response to audio..."):
-                response_audio = text_to_speech(response)
-                st.audio(response_audio, format="audio/mp3")
+    try:
+        chat_completion = client.chat.completions.create(
+            messages=[
+                {"role": "user", "content": user_input}
+            ],
+            model="llama3-8b-8192",
+            stream=False,
+        )
+        return chat_completion.choices[0].message.content
+    except Exception as e:
+        return f"Error in LLM interaction: {e}"
+
+# Function to convert text to speech
+def text_to_sp