An automated tool to recognize Japanese text within manga from an entire image files or folder and export the results to a text file.
Speech Bubble Detection: Fine-tuned YOLOv8m model trained to detect speech bubbles across Manga, Webtoon, Manhua, and Western Comic style images.
Model Source: ogkalu/comic-speech-bubble-detector-yolov8m
Manga OCR: Deep learning model to convert Japanese text images inside speech bubbles into digital text using the Vision Encoder Decoder framework.
Model Source: kha-white/manga-ocr-base


If you just want to use the GUI app to extract text, simply install uv and execute the run.bat file.
To install uv, open PowerShell and run the following command:
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
2 commits
Jupyter Notebook
95.0%
Python
5.0%
An automated tool to recognize Japanese text within manga from an entire image files or folder and export the results to a text file.
Speech Bubble Detection: Fine-tuned YOLOv8m model trained to detect speech bubbles across Manga, Webtoon, Manhua, and Western Comic style images.
Model Source: ogkalu/comic-speech-bubble-detector-yolov8m
Manga OCR: Deep learning model to convert Japanese text images inside speech bubbles into digital text using the Vision Encoder Decoder framework.
Model Source: kha-white/manga-ocr-base


If you just want to use the GUI app to extract text, simply install uv and execute the run.bat file.
To install uv, open PowerShell and run the following command:
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
2 commits
Jupyter Notebook
95.0%
Python
5.0%