Releases: PatchyVideo/MMD-Translator
MMD字幕识别v0.5.2版本/MMD-OCR v0.5.2
中文
下载说明
你需要下载MMD-OCR-v0.5.2.7z
硬件要求
你需要至少一张9系及以上的NVIDIA显卡(只要是Maxwell及以上都可以,包括部分9系显卡)
程序会占用2G内存和1G显存。
CUDA计算能力版本:5.2;6.1;7.5;8.2
使用说明
请先更新显卡驱动到最新版
把视频文件拖到mmdocr.exe上,一个对应的srt文件就会在视频的同一文件夹下生成
已知问题
- 部分字符不在字典内,无法识别。
- 滚动字幕、旋转字幕和特效字幕处理的很差,不建议用在舞蹈类或PV类视频上。
- 部分文字水印会被持续识别,污染生成的字幕文件。
- (重要)不支持竖版文字。
- 会自动覆盖已有的srt文件,注意备份。
- 优化不到位,无法利用全部显卡资源。
文件说明
mmdocr.exe:识别程序,不会展示中间识别过程,实际生产请使用该程序。
代码在src文件夹里。
English
Download
You only need MMD-OCR-v0.5.2.7z
System requirements
NVIDIA GPU at least GTX 900 series is required.(Maxwell or later)
Program will use 2G of RAM and 1G of VRAM.
CUDA computability required: 5.2;6.1;7.5;8.2
How to use
Please update your graphics driver to latest version before proceeding.
Drag any video file to mmdocr.exe and a .srt file will be generated at the same folder as video file.
Known issues
- Some characters are not recognized because they're not in my dictionary.
- Scrolling texts and rotated texts are handled poorly.
- Some watermark texts will be recognized, degrade generated srt file.
- (IMPORTANT) Vertical texts are not supported.
- Existing srt file will be overwritten, please always make backups.
- Not much optimization effort was made into this program, it will not utilize all available GPU resources.
Files
mmdocr.exe: Please use this one in actual production environment, this will not show any intermediate processes.
Code is available under src/ folder.
MMD字幕识别v0.5.1版本/MMD-OCR v0.5.1
中文
下载说明
你需要下载MMD-OCR-v0.5.7z
和v0.5.1补丁v0.5.1-patch.7z
硬件要求
你需要至少一张9系及以上的NVIDIA显卡(只要是Maxwell及以上都可以,包括部分7系显卡和GTX 800M系列,Ampere除外)
程序会占用2G内存和1G显存。
CUDA计算能力版本:5.0;5.2;6.1;7.0;7.5
使用说明
请先更新显卡驱动到最新版
把视频文件拖到mmdocr.exe上,一个对应的srt文件就会在视频的同一文件夹下生成
已知问题
- 部分字符不在字典内,无法识别。
- 滚动字幕、旋转字幕和特效字幕处理的很差,不建议用在舞蹈类或PV类视频上。
- 部分文字水印会被持续识别,污染生成的字幕文件。
- (重要)不支持竖版文字。
- 会自动覆盖已有的srt文件,注意备份。
- 优化不到位,无法利用全部显卡资源。
文件说明
mmdocr.exe:识别程序,不会展示中间识别过程,实际生产请使用该程序。
mmdocr_verbose.exe:识别程序,展示中间识别过程,如果你想看过程可以用这个程序。
代码在src文件夹里。
English
Download
You only need MMD-OCR-v0.5.7z
And v0.5.1 Patchv0.5.1-patch.7z
System requirements
NVIDIA GPU at least GTX 900 series is required.(Maxwell or later, not including Ampere)
Program will use 2G of RAM and 1G of VRAM.
CUDA computability required: 5.0;5.2;6.1;7.0;7.5
How to use
Please update your graphics driver to latest version before proceeding.
Drag any video file to mmdocr.exe and a .srt file will be generated at the same folder as video file.
Known issues
- Some characters are not recognized because they're not in my dictionary.
- Scrolling texts and rotated texts are handled poorly.
- Some watermark texts will be recognized, degrade generated srt file.
- (IMPORTANT) Vertical texts are not supported.
- Existing srt file will be overwritten, please always make backups.
- Not much optimization effort was made into this program, it will not utilize all available GPU resources.
Files
mmdocr.exe: Please use this one in actual production environment, this will not show any intermediate processes.
mmdocr_verbose.exe: This one will show intermediate processes, use this if that is what you want to see.
Code is available under src/ folder.
v0.4.2 Version/v0.4.2 版本
中文
你需要至少一张9系及以上的NVIDIA显卡(只要是Maxwell及以上都可以,包括部分7系显卡和GTX 800M系列)
程序会占用2G内存和1G显存。
CUDA计算能力版本:5.0;5.2;6.1;7.0;7.5
代码在src文件夹里。
使用说明:
请更新显卡驱动到最新版
把视频文件拖到mmdocr.exe上,一个对应的srt文件就会在视频的同一文件夹下生成
已知问题:
- 部分字符不在字典内,无法识别。
- 滚动字幕、旋转字幕和特效字幕处理的很差。
English
NVIDIA GPU at least GTX 900 series is required.(Maxwell or later)
Program will use 2G of RAM and 1G of VRAM.
CUDA computability required: 5.0;5.2;6.1;7.0;7.5
Code is available under src/ folder.
How to use:
Drag any video file to mmdocr.exe and a .srt file will be generated at the same folder as video file.
Known issues:
- Some characters are not recognized because they're not in my dictionary.
- Scrolling texts and rotated texts are handled poorly.
v0.4 C++ Version All Files
C++ version with all files
NVIDIA GPU at least GTX 900 series is required.
CUDA computability required: 5.0;5.2;6.1;7.0;7.5
How to use:
Drag any video file to mmdocr.exe and a .srt file will be generated at the same folder as video file.
你需要至少一张9系及以上的NVIDIA显卡
CUDA计算能力版本:5.0;5.2;6.1;7.0;7.5
使用说明:
把视频文件拖到mmdocr.exe上,一个对应的srt文件就会在视频的同一文件夹下生成
Trained models
This release provides trained detection(textdet_300k.pth) and OCR(ocr_640k.pth) model of version v0.3