import streamlit as st
import speech_recognition as sr
from transformers import pipeline
from tts import TTS  # Hugging Face TTS model
import requests

# Load the chatbot model
chatbot = pipeline("conversational", model="facebook/blenderbot-400M-distill")

# Function to convert speech to text
def speech_to_text():
    recognizer = sr.Recognizer()
    with sr.Microphone() as source:
        st.info("Listening...")
        audio = recognizer.listen(source)
        try:
            text = recognizer.recognize_google(audio)
            return text
        except sr.UnknownValueError:
            return "Sorry, I could not understand the audio."
        except sr.RequestError:
            return "Speech recognition service is not available."

# Function to generate avatar video
def generate_avatar_video(text_response):
    # Call the API of an avatar service (e.g., D-ID, Synthesia)
    api_url = "https://api.example.com/generate-avatar"
    payload = {"text": text_response}
    response = requests.post(api_url, json=payload)
    video_url = response.json().get("video_url")
    return video_url

st.title("🗣️ Live Video Chatbot")

# Button to start recording
if st.button("Speak"):
    user_input = speech_to_text()
    if user_input:
        st.write(f"**You:** {user_input}")
        # Generate chatbot response
        bot_response = chatbot(user_input)
        response_text = bot_response[0]["generated_text"]
        st.write(f"**Bot:** {response_text}")
        
        # Generate avatar video
        video_url = generate_avatar_video(response_text)
        
        # Display the video response
        st.video(video_url)