This document defines functions for translating text and documents between languages. It extracts names from text, translates text to a target language, and replaces names in translated documents. It then builds a Streamlit app with these functions to allow users to upload documents, select a target language, replace names, translate the document, and download the translated output.
This document defines functions for translating text and documents between languages. It extracts names from text, translates text to a target language, and replaces names in translated documents. It then builds a Streamlit app with these functions to allow users to upload documents, select a target language, replace names, translate the document, and download the translated output.
This document defines functions for translating text and documents between languages. It extracts names from text, translates text to a target language, and replaces names in translated documents. It then builds a Streamlit app with these functions to allow users to upload documents, select a target language, replace names, translate the document, and download the translated output.
This document defines functions for translating text and documents between languages. It extracts names from text, translates text to a target language, and replaces names in translated documents. It then builds a Streamlit app with these functions to allow users to upload documents, select a target language, replace names, translate the document, and download the translated output.
Download as DOCX, PDF, TXT or read online from Scribd
Download as docx, pdf, or txt
You are on page 1/ 2
import streamlit as st
from transformers import MarianMTModel, MarianTokenizer
from docx import Document import os import string
def extract_names(text): """ Extract potential names from the given text.
Args: text (str): Input text to extract names from.
Returns: set: Set of potential names. """ potential_names = set(word.strip(string.punctuation) for word in text.split() if word.istitle()) return potential_names
def translate_to_language(text, target_language):
""" Translate the given text to the target language using the specified model.
Args: text (str): Input text to translate. target_language (str): Target language code.
Returns: str: Translated text. """ if not text.strip(): return "" else: model_name = f"Helsinki-NLP/opus-mt-en-{target_language}" model = MarianMTModel.from_pretrained(model_name) tokenizer = MarianTokenizer.from_pretrained(model_name)
if uploaded_file: doc = Document(uploaded_file) doc_text = "\n".join([para.text for para in doc.paragraphs])
# Extract and display potential names from input DOCX
potential_names = extract_names(doc_text) st.sidebar.header("Potential Names from Demo Script") selected_names = st.sidebar.multiselect("Select Names to Replace:", potential_names)
# Dictionary to store name replacements
name_replacement_dict = {} for selected_name in selected_names: new_name = st.sidebar.text_input(f"Replace '{selected_name}' with:", "") if new_name: name_replacement_dict[selected_name] = new_name
# Translate and download button
if st.button("Translate and Download"): translated_doc = translate_and_replace_names(uploaded_file, target_language, name_replacement_dict) output_filename = "translated_output.docx" translated_doc.save(output_filename) st.success("Translation and Name Replacement complete!")
# Input field for custom filename
custom_filename = st.text_input("Enter the filename:", "translated_output.docx")
