App.py

# Import libraries
import whisper
import requests
from gtts import gTTS
import os
import gradio as gr

# Load Whisper model for transcription
model = whisper.load_model("base")

# Function to transcribe audio input using Whisper
def transcribe_audio(audio_file):
    result = model.transcribe(audio_file)
    return result['text']

# Function to get response from Groq's LLM API (replace with actual API details)
def get_llm_response(text_input):
    # Example Groq API call (you'll need to modify with actual endpoint and key)
    response = requests.post("https://api.groq.com/llm", json={"prompt": text_input})
    return response.json()["output"]  # Adjust this based on the actual response format

# Function to convert text response to speech using gTTS
def text_to_speech(response_text):
    tts = gTTS(text=response_text, lang='en')
    tts.save("response_audio.mp3")
    return "response_audio.mp3"

# Gradio pipeline function combining all steps
def chatbot_pipeline(audio):
    # Step 1: Transcribe audio input
    transcription = transcribe_audio(audio)
    
    # Step 2: Get LLM response from Groq API
    llm_response = get_llm_response(transcription)
    
    # Step 3: Convert LLM response to audio using gTTS
    audio_output = text_to_speech(llm_response)
    
    return llm_response, audio_output

# Gradio interface for real-time interaction
interface = gr.Interface(
    fn=chatbot_pipeline,
    inputs=gr.Audio(type="filepath"),  # Accepts audio file input, records if mic
    outputs=[gr.Textbox(label="Transcription"), gr.Audio(label="Response Audio")]
)

# Launch Gradio interface
interface.launch()