import streamlit as st
from transformers import MarianMTModel, MarianTokenizer
def translate_text(text, src_lang, tgt_lang):
model_name = f'Helsinki-NLP/opus-mt-{src_lang}-{tgt_lang}'
model = MarianMTModel.from_pretrained(model_name)
tokenizer = MarianTokenizer.from_pretrained(model_name)
translated = model.generate(**tokenizer(text, return_tensors="pt", padding=True))
translated_text = tokenizer.decode(translated[0], skip_special_tokens=True)
return translated_text
# Main Streamlit app
def main():
# Set the title with HTML
st.markdown(
"""
LangToLang
Language Translator
""", unsafe_allow_html=True)
# Sidebar with additional options
st.sidebar.header("Settings")
st.sidebar.markdown("**Choose Translation Options**")
# Language options
language_dict = {
"English": "en",
"French": "fr",
"German": "de",
"Spanish": "es",
"Italian": "it",
"Russian": "ru",
"Chinese": "zh",
"Japanese": "ja",
"Korean": "ko",
"Arabic": "ar",
"Urdu": "ur",
}
src_lang = st.sidebar.selectbox("Select Source Language", list(language_dict.keys()), index=0)
tgt_lang = st.selectbox("Select Target Language", list(language_dict.keys()), index=1)
# Text input area
text = st.text_area("Enter Text to Translate")
# Translate button
if st.button("Translate"):
if text:
translated_text = translate_text(text, language_dict[src_lang], language_dict[tgt_lang])
st.subheader("Translated Text:")
st.success(translated_text)
else:
st.warning("Please enter text to translate.")
# Add some info in the sidebar
st.sidebar.markdown("**About LangToLang**")
st.sidebar.info("LangToLang is a simple language translator app powered by Hugging Face models, allowing you to translate text between different languages effortlessly.")
# Footer
st.sidebar.markdown(
"""
""", unsafe_allow_html=True)
if __name__ == "__main__":
main()