app.py

import streamlit as st

# Use a pipeline as a high-level helper
from transformers import pipeline

toxic_model = pipeline("text-classification", model="Matt09Miao/GP5_tweet_toxic")  
    
def text2audio(text):
    pipe = pipeline("text-to-audio", model="Matthijs/mms-tts-eng")
    audio_data = pipe(text)
    return audio_data

st.set_page_config(page_title="Tweet Toxicity Analysis")

st.header("Please input a Tweet for Toxicity Analysis :performing_arts:")
input = st.text_area("Enter a Tweer for analysis")

if st.button("Toxic Analysis"):

   result = toxic_model(input)

   # Display the result
   st.write("Tweet:", input)
   st.write("label:", result[0]['label'])
   st.write("score:", result[0]['score'])

   # Read the result
   audio_data1 = text2audio(input)
   st.audio(audio_data1['audio'],
                     format="audio/wav",
                     start_time=0,
                     sample_rate = audio_data1['sampling_rate'])

   audio_data2 = text2audio(result[0]['label'])
   st.audio(audio_data2['audio'],
                     format="audio/wav",
                     start_time=0,
                     sample_rate = audio_data2['sampling_rate'])