Spaces:
Build error
Build error
File size: 2,197 Bytes
3937682 ebf0184 3937682 |
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 58 59 60 61 62 63 64 65 66 67 68 69 70 |
import gradio as gr
import torch
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
import os
import sys
# Add local IndicTransToolkit path
sys.path.append(os.path.abspath("libs/IndicTransToolkit"))
from IndicTransToolkit.processor import IndicProcessor
# Load processor and model
ip = IndicProcessor(inference=True)
tokenizer = AutoTokenizer.from_pretrained("ai4bharat/indictrans2-en-indic-dist-200M", trust_remote_code=True)
model = AutoModelForSeq2SeqLM.from_pretrained("ai4bharat/indictrans2-en-indic-dist-200M", trust_remote_code=True)
LANG_OPTIONS = [
"hin_Deva", # Hindi
"ben_Beng", # Bengali
"pan_Guru", # Punjabi
"guj_Gujr", # Gujarati
"tam_Taml", # Tamil
"tel_Telu", # Telugu
"mal_Mlym", # Malayalam
"mar_Deva", # Marathi
"kan_Knda", # Kannada
"asm_Beng", # Assamese
"kas_Arab", # Kashmiri (Arabic)
"kas_Deva", # Kashmiri (Devanagari)
"san_Deva", # Sanskrit
"brx_Deva", # Bodo
"mai_Deva", # Maithili
"sat_Olck", # Santali
"eng_Latn", # English
"urd_Arab" # Urdu
]
def translate(text, target_lang):
if not text.strip():
return "Please enter some text."
try:
batch = ip.preprocess_batch([text], src_lang="eng_Latn", tgt_lang=target_lang)
batch = tokenizer(batch, padding="longest", truncation=True, max_length=256, return_tensors="pt")
with torch.inference_mode():
outputs = model.generate(**batch, num_beams=5, max_length=256)
with tokenizer.as_target_tokenizer():
decoded = tokenizer.batch_decode(outputs, skip_special_tokens=True, clean_up_tokenization_spaces=True)
translated = ip.postprocess_batch(decoded, lang=target_lang)[0]
return translated
except Exception as e:
return f"Error: {e}"
demo = gr.Interface(
fn=translate,
inputs=[
gr.Textbox(label="Enter text in English", lines=5),
gr.Dropdown(choices=LANG_OPTIONS, label="Select Target Language")
],
outputs="text",
title="IndicTrans Translator",
description="Translate English text into Indian languages using IndicTrans2."
)
if __name__ == "__main__":
demo.launch()
|