app.py

# -*- coding: utf-8 -*-
"""
Created on Thu Sep 21 22:17:43 2023

@author: Loges
"""

import streamlit as st
import sentencepiece
from gtts import gTTS
import base64
import time
from transformers import pipeline, T5Tokenizer, T5ForConditionalGeneration

model=T5ForConditionalGeneration.from_pretrained("Logeswaransr/T5_MineAI_Prototype").to("cpu")
tokenizer=T5Tokenizer.from_pretrained("Logeswaransr/T5_MineAI_Prototype")

pipe=pipeline('text2text-generation', model=model, tokenizer=tokenizer)

greetings=["Hello! My name is MineAI, A specially trained LLM here to assist you on your Mining Related Queries.","How may I help you?"]

st.set_page_config(page_title='Sample Chatbot', layout='wide')

if 'messages' not in st.session_state:
    st.session_state.messages=[]

st.subheader("Mine AI")

for message in st.session_state.messages:
    with st.chat_message(message['role']):
        st.markdown(message['content'])
        
## messages element format: {'role':'user', 'content':'<user prompt>'}

if st.session_state.messages==[]:
    for gr in greetings:
        with st.chat_message("assistant"):
            st.markdown(gr)

        if gr==greetings[1]:
            tts=gTTS(gr)
            tts.save('greeting_audio.mp3')
            with open('greeting_audio.mp3', 'rb') as file:
                greeting_audio_data=file.read()
            greeting_audio_base64 = base64.b64encode(greeting_audio_data).decode('utf-8')
            greeting_audio_tag = f'<audio autoplay="true" src="data:audio/mp3;base64,{greeting_audio_base64}">'
            st.markdown(greeting_audio_tag, unsafe_allow_html=True)
    
        st.session_state.messages.append({
            'role':'assistant',
            'content': gr})

audio_stream="response_audio.mp3"
tts=gTTS("Here is your answer")
tts.save(audio_stream)
with open(audio_stream, 'rb') as file:
    audio_data=file.read()
audio_base64 = base64.b64encode(audio_data).decode('utf-8')
audio_tag = f'<audio autoplay="true" src="data:audio/mp3;base64,{audio_base64}">'

if prompt:=st.chat_input("Enter your query"):
    with st.chat_message("user"):
        st.markdown(prompt)

    st.session_state.messages.append({
        'role':'user',
        'content': prompt})

    out=pipe(prompt)
    response=out[0]['generated_text']
    
    # response = f"Analysis: {response}"
    
    with st.chat_message("assistant"):
        st.markdown(response)

    st.markdown(audio_tag, unsafe_allow_html=True)
        
    st.session_state.messages.append({
        'role':'assistant',
        'content': response})