Spaces:
Running
Running
| from fastapi import FastAPI, HTTPException | |
| from transformers import pipeline | |
| import langdetect | |
| import logging | |
| import os | |
| import opencc # Untuk konversi Mandarin Tradisional ke Sederhana | |
| os.environ["HF_HOME"] = "/app/cache" | |
| os.environ["TRANSFORMERS_CACHE"] = "/app/cache" | |
| app = FastAPI() | |
| logging.basicConfig(level=logging.INFO) | |
| logger = logging.getLogger(__name__) | |
| MODEL_MAP = { | |
| "id": "Helsinki-NLP/opus-mt-id-en", # Indonesia ke Inggris | |
| "th": "Helsinki-NLP/opus-mt-th-en", # Thailand ke Inggris | |
| "fr": "Helsinki-NLP/opus-mt-fr-en", # Prancis ke Inggris | |
| "es": "Helsinki-NLP/opus-mt-es-en", # Spanyol ke Inggris | |
| "ja": "Helsinki-NLP/opus-mt-ja-en", # Jepang ke Inggris | |
| "zh-CN": "Helsinki-NLP/opus-mt-zh-en", # Mandarin Sederhana ke Inggris | |
| "zh-TW": "Helsinki-NLP/opus-mt-zh-en", # Mandarin Tradisional ke Inggris | |
| "vi": "Helsinki-NLP/opus-mt-vi-en", # Vietnam ke Inggris | |
| } | |
| translators = {} | |
| try: | |
| for lang, model_name in MODEL_MAP.items(): | |
| logger.info(f"Loading model for {lang}...") | |
| translators[lang] = pipeline("translation", model=model_name) | |
| logger.info(f"Model for {lang} loaded successfully") | |
| except Exception as e: | |
| logger.error(f"Model initialization failed: {str(e)}") | |
| raise Exception(f"Model initialization failed: {str(e)}") | |
| # Inisialisasi konverter OpenCC untuk Tradisional ke Sederhana | |
| converter = opencc.OpenCC('t2s') # t2s = Traditional to Simplified | |
| def is_traditional_chinese(text: str) -> bool: | |
| """Cek apakah teks menggunakan karakter Mandarin Tradisional.""" | |
| # Contoh sederhana: deteksi beberapa karakter Tradisional | |
| traditional_chars = set('繁體字') # Bisa diperluas dengan daftar karakter yang lebih lengkap | |
| return any(char in traditional_chars for char in text) | |
| def detect_language(text: str) -> str: | |
| """Deteksi bahasa dari teks menggunakan langdetect.""" | |
| try: | |
| lang = langdetect.detect(text) | |
| # Jika terdeteksi sebagai Mandarin (zh), cek apakah Tradisional atau Sederhana | |
| if lang == "zh": | |
| return "zh-TW" if is_traditional_chinese(text) else "zh-CN" | |
| return lang if lang in MODEL_MAP else "en" # Default ke Inggris jika tidak didukung | |
| except Exception as e: | |
| logger.warning(f"Language detection failed: {str(e)}, defaulting to English") | |
| return "en" | |
| async def translate(text: str): | |
| if not text: | |
| raise HTTPException(status_code=400, detail="Text input is required") | |
| try: | |
| # Deteksi bahasa | |
| source_lang = detect_language(text) | |
| logger.info(f"Detected source language: {source_lang}") | |
| # Jika sudah Inggris, kembalikan teks asli | |
| if source_lang == "en": | |
| return {"translated_text": text} | |
| # Jika Mandarin Tradisional, konversi ke Sederhana | |
| input_text = text | |
| if source_lang == "zh-TW": | |
| input_text = converter.convert(text) | |
| logger.info("Converted Traditional Chinese to Simplified Chinese") | |
| # Terjemahkan ke Inggris | |
| translator = translators.get(source_lang) | |
| if not translator: | |
| raise HTTPException(status_code=400, detail=f"Translation not supported for language: {source_lang}") | |
| result = translator(input_text) | |
| return {"translated_text": result[0]["translation_text"]} | |
| except Exception as e: | |
| logger.error(f"Processing failed: {str(e)}") | |
| raise HTTPException(status_code=500, detail=f"Processing failed: {str(e)}") |