app.py

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

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

# Groq API
client = Groq(api_key=os.getenv("Voice"))

def voice_chat(audio):

    # Step 1: Speech to Text
    result = model.transcribe(audio)
    user_text = result["text"]

    # Step 2: Send to Groq LLM
    completion = client.chat.completions.create(
        model="llama-3.1-8b-instant",
        messages=[
            {"role": "user", "content": user_text}
        ]
    )

    bot_text = completion.choices[0].message.content

    # Step 3: Text to Speech
    tts = gTTS(bot_text)
    tts.save("response.mp3")

    return user_text, bot_text, "response.mp3"


interface = gr.Interface(
    fn=voice_chat,
    inputs=gr.Audio(type="filepath", label="Speak Here"),
    outputs=[
        gr.Textbox(label="Your Speech (Text)"),
        gr.Textbox(label="Bot Response (Text)"),
        gr.Audio(label="Bot Voice Response")
    ],
    title="AI Voice Chatbot",
    description="Speak to the AI. It will reply in text and voice."
)

interface.launch()