InterviewAnalyzer / pages /2_Interview_Summarization_πŸ“–_.py
DrDominikDellermann's picture
Rename pages/2_Earnings_Summarization_πŸ“–_.py to pages/2_Interview_Summarization_πŸ“–_.py
1521e2b
import streamlit as st
from functions import *
st.set_page_config(page_title="Interview Summarization", page_icon="πŸ“–")
st.sidebar.header("Summarization")
max_len= st.slider("Maximum length of the summarized text",min_value=70,max_value=200,step=10,value=100)
min_len= st.slider("Minimum length of the summarized text",min_value=20,max_value=200,step=10)
st.markdown("####")
st.subheader("Summarized Interview with matched Entities")
if "earnings_passages" not in st.session_state:
st.session_state["earnings_passages"] = ''
if st.session_state['earnings_passages']:
with st.spinner("Summarizing and matching entities, this takes a few seconds..."):
try:
text_to_summarize = chunk_and_preprocess_text(st.session_state['earnings_passages'])
print(text_to_summarize)
summarized_text = summarize_text(text_to_summarize,max_len=max_len,min_len=min_len)
except IndexError:
try:
text_to_summarize = chunk_and_preprocess_text(st.session_state['earnings_passages'])
summarized_text = summarize_text(text_to_summarize,max_len=max_len,min_len=min_len)
except IndexError:
text_to_summarize = chunk_and_preprocess_text(st.session_state['earnings_passages'])
summarized_text = summarize_text(text_to_summarize,max_len=max_len,min_len=min_len)
entity_match_html = highlight_entities(text_to_summarize,summarized_text)
st.markdown("####")
with st.expander(label='Summarized Interview',expanded=True):
st.write(entity_match_html, unsafe_allow_html=True)
st.markdown("####")
summary_downloader(summarized_text)
else:
st.write("No text to summarize detected, please ensure you have entered the YouTube URL on the Sentiment Analysis page")