app.py

 import os
import whisper
from gtts import gTTS
from groq import Groq
import gradio as gr

# Set your Groq API key (added directly for simplicity, ensure you keep it secure)
os.environ["GROQ_API_KEY"] = "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"

# Initialize Whisper model
model = whisper.load_model("base")

# Initialize Groq API client
client = Groq(api_key=os.environ.get("GROQ_API_KEY"))

# Step 1: Transcribe Audio (Speech-to-Text using Whisper)
def transcribe_audio(audio_path):
    result = model.transcribe(audio_path)
    return result['text']

# Step 2: Interact with LLM (Groq API)
def interact_with_llm(user_input):
    chat_completion = client.chat.completions.create(
        messages=[
            {
                "role": "user",
                "content": user_input,
            }
        ],
        model="llama3-8b-8192",
        stream=False,
    )
    response = chat_completion.choices[0].message.content
    return response

# Step 3: Convert Text to Speech using gTTS
def text_to_speech(text):
    tts = gTTS(text, lang="en")
    audio_file = "response.mp3"
    tts.save(audio_file)
    return audio_file

# Combined workflow: Transcribe -> Interact with LLM -> Convert to Speech
def chatbot(audio):
    # Step 1: Transcribe Audio to Text
    transcription = transcribe_audio(audio)
    
    # Step 2: Get LLM response based on transcription
    llm_response = interact_with_llm(transcription)
    
    # Step 3: Convert LLM response to audio (text-to-speech)
    audio_output = text_to_speech(llm_response)
    
    return transcription, llm_response, audio_output

# Gradio Interface setup
interface = gr.Interface(
    fn=chatbot,
    inputs=gr.Audio(type="filepath", label="Speak into the microphone"),
    outputs=[
        "text",  # Transcription output
        "text",  # LLM response output
        gr.Audio(type="filepath", label="Response Audio")  # Final audio output
    ],
    live=True,
    title="Real-Time Voice-to-Voice Chatbot",
    description="Talk to an AI in real-time! Speak into the microphone, get a response, and hear it back.",
)

# Launch Gradio app
if __name__ == "__main__":
    interface.launch()