Spaces:
Sleeping
Sleeping
File size: 2,373 Bytes
6bd039b e84b59a 6bd039b e84b59a 6bd039b e84b59a 6bd039b e84b59a 6bd039b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 |
import streamlit as st
from transformers import MarianMTModel, MarianTokenizer
import torch
@st.cache_resource
def load_model_and_tokenizer(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)
return model, tokenizer
def translate_text(text, model, tokenizer):
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = model.to(device)
inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True).to(device)
translated = model.generate(**inputs)
return tokenizer.decode(translated[0], skip_special_tokens=True)
st.set_page_config(page_title="Language Translator", layout="wide")
st.title("🌍 Language Translator")
st.markdown("Translate text quickly and easily between multiple languages using open-source models.")
language_codes = {
"English": "en", "Spanish": "es", "French": "fr", "German": "de",
"Chinese": "zh", "Japanese": "ja", "Hindi": "hi", "Arabic": "ar",
"Russian": "ru", "Portuguese": "pt",
}
col1, col2 = st.columns(2)
with col1:
source_language = st.selectbox("Select source language:", options=language_codes.keys(), index=0)
with col2:
target_language = st.selectbox("Select target language:", options=language_codes.keys(), index=1)
st.subheader("Input Text")
input_text = st.text_area("Enter the text you want to translate:", height=150)
if st.button("Translate"):
if not input_text.strip():
st.error("Please enter text to translate.")
elif source_language == target_language:
st.warning("Source and target languages are the same. Please select different languages.")
else:
try:
src_lang = language_codes[source_language]
tgt_lang = language_codes[target_language]
model, tokenizer = load_model_and_tokenizer(src_lang, tgt_lang)
translation = translate_text(input_text, model, tokenizer)
st.subheader("Translated Text")
st.text_area("Translation:", value=translation, height=150, disabled=True)
except Exception as e:
st.error(f"An error occurred: {e}")
st.markdown("---")
st.markdown("Developed using [Helsinki-NLP](https://huggingface.co/Helsinki-NLP) open-source models and Streamlit.")
|