app.py

import os
import sys

# Function to check and install required libraries
def install_libraries():
    try:
        import whisper
        import gtts
        import gradio as gr
        from groq import Groq
    except ImportError:
        print("Required libraries not found. Installing now...")
        os.system("pip install git+https://github.com/openai/whisper.git gtts gradio groq")
        # Try to import again after installation
        try:
            import whisper
            import gtts
            import gradio as gr
            from groq import Groq
        except ImportError:
            print("Failed to install required libraries. Please install them manually.")
            sys.exit(1)

# Call the install function
install_libraries()

# Now we can safely import the libraries
import whisper
import gtts
import gradio as gr
from groq import Groq

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

# Hard-code the Groq API key
Groq_api_key = "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"

# Check if the API key is set (even though it's hard-coded, we can keep this check)
if Groq_api_key is None:
    raise ValueError("API key for Groq not found. Please set the 'GROQ_API_KEY' environment variable.")

# Initialize the Groq client
client = Groq(api_key=Groq_api_key)

# Function to transcribe audio
def transcribe_audio(audio_path):
    try:
        result = model.transcribe(audio_path)
        return result["text"]
    except Exception as e:
        return f"Error transcribing audio: {str(e)}"

# Function to get response from Groq's API
def get_groq_response(transcribed_text):
    try:
        chat_completion = client.chat.completions.create(
            messages=[{
                "role": "user",
                "content": transcribed_text,
            }],
            model="llama3-8b-8192",
        )
        return chat_completion.choices[0].message.content
    except Exception as e:
        return f"Error getting response from Groq: {str(e)}"

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

# Gradio chatbot function
def chatbot(audio):
    # Step 1: Transcribe the audio
    transcribed_text = transcribe_audio(audio)
    
    # Step 2: Get LLM response from Groq API
    response_text = get_groq_response(transcribed_text)
    
    # Step 3: Convert response text to speech
    response_audio = text_to_speech(response_text)
    
    # Return the response audio
    return response_audio

# Create a Gradio interface
iface = gr.Interface(
    fn=chatbot,
    inputs=gr.Audio(type="filepath"),  # Updated to remove 'source'
    outputs="audio",
    title="Voice-to-Voice Chatbot",
    description="Speak to the chatbot and listen to the response!",
    live=True
)

# Launch the interface
iface.launch()