app.py

import streamlit as st
import pandas as pd
from transformers import AutoTokenizer, AutoModelForCausalLM
import speech_recognition as sr
from pydub import AudioSegment

# Load the Netflix dataset from CSV
@st.cache_data
def load_data():
    return pd.read_csv("netflix_titles.csv")

# Load DialoGPT model and tokenizer
@st.cache_resource
def load_model():
    tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-medium")
    model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-medium")
    return tokenizer, model

# Function to search the dataset for movie details
def search_movie_details(query, data):
    query = query.lower()
    results = data[
        data["title"].str.lower().str.contains(query) |
        data["cast"].str.lower().str.contains(query) |
        data["director"].str.lower().str.contains(query)
    ]
    return results

# Function to convert voice to text
def voice_to_text():
    recognizer = sr.Recognizer()
    with sr.Microphone() as source:
        st.write("Speak now...")
        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 "Sorry, the speech service is down."

# Streamlit App
st.title("Netflix Movie Details Chatbot 🎬")

# Load dataset and model
data = load_data()
tokenizer, model = load_model()

# Input options: Text or Voice
input_option = st.radio("Choose input method:", ("Text", "Voice"))

user_input = ""
if input_option == "Text":
    user_input = st.text_input("Enter the movie name, director, or cast:")
elif input_option == "Voice":
    if st.button("Start Recording"):
        user_input = voice_to_text()
        st.write(f"You said: {user_input}")

# Generate response
if user_input:
    # Search for movie details
    movie_results = search_movie_details(user_input, data)
    
    if not movie_results.empty:
        st.write("Here are the matching results:")
        for _, row in movie_results.iterrows():
            st.write(f"**Title:** {row['title']}")
            st.write(f"**Type:** {row['type']}")
            st.write(f"**Director:** {row['director']}")
            st.write(f"**Cast:** {row['cast']}")
            st.write(f"**Release Year:** {row['release_year']}")
            st.write(f"**Rating:** {row['rating']}")
            st.write(f"**Description:** {row['description']}")
            st.write("---")
    else:
        # Use DialoGPT for general conversation
        inputs = tokenizer.encode(user_input, return_tensors="pt")
        outputs = model.generate(inputs, max_length=100, num_return_sequences=1)
        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
        st.write(f"Chatbot: {response}")