External/SnapOtter
1
0
Fork 0
mirror of https://github.com/snapotter-hq/SnapOtter.git synced 2026-08-03 07:46:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix: correct PaddleOCR language codes for model download and OCR

Browse Source 
PaddleOCR uses its own language codes (ch, japan, korean, latin) not
ISO codes (zh, ja, ko, de, fr, es). The download script and ocr.py
now map API language codes to PaddleOCR codes correctly. German,
French, and Spanish all use the "latin" script model.
This commit is contained in:
Siddharth Kumar Sah
2026-04-10 00:40:45 +08:00
parent 7face19238
commit e1ee57103c
2 changed files with 8 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docker/download_models.py
+3 -1
View File
@@ -23,7 +23,9 @@ REMBG_MODELS = [
"birefnet-general",
]
PADDLEOCR_LANGUAGES = ["en", "de", "fr", "es", "zh", "ja", "ko"]
# PaddleOCR language codes (not ISO). German/French/Spanish use "latin" model.
# Valid keys: ch, en, korean, japan, chinese_cht, ta, te, ka, latin, arabic, cyrillic, devanagari
PADDLEOCR_LANGUAGES = ["en", "ch", "japan", "korean", "latin"]
def download_rembg_models():
packages/ai/python/ocr.py
+5 -1
View File
@@ -36,8 +36,12 @@ def run_paddleocr(input_path, language):
from paddleocr import PaddleOCR
from gpu import gpu_available
# Map API language codes to PaddleOCR codes
paddle_lang_map = {"en": "en", "de": "latin", "fr": "latin", "es": "latin", "zh": "ch", "ja": "japan", "ko": "korean"}
paddle_lang = paddle_lang_map.get(language, "en")
emit_progress(20, "Loading")
ocr = PaddleOCR(lang=language, use_gpu=gpu_available())
ocr = PaddleOCR(lang=paddle_lang, use_gpu=gpu_available())
emit_progress(30, "Scanning")
result = ocr.ocr(input_path)
emit_progress(70, "Extracting text")