Releases: timminator/VideOCR
Releases · timminator/VideOCR
VideOCR v1.3.2
Highlights
- Integrates the latest PaddleOCR 3.2 version which adds two new languages (Greek and Thai) and improved accuracy for the English model
Fixes:
- Fixes incompatibility with Nvidia 10 Series graphics cards
- Fixes crash on videos whose frame count is unknown
- Fixes key error on the "Save SRT in Video Directory" option
Highlight of the v1.3.x release
- The main highlight is the newly added support for PaddleOCR 3.x.x with greatly improved OCR capabilities in more than 40 languages,
including English, Chinese, Japanese, French, Spanish, Portuguese, Russian, Korean, etc.
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- Linux version requires glibc 2.35 or newer (can be checked by running
ldd --versionin a terminal)
Download Tips:
- Windows:
- CPU version:
VideOCR-CPU-v1.3.2-setup-x64.exe - GPU version:
VideOCR-GPU-v1.3.2-CUDA-11.8-setup-x64.exe (Nvidia 10 Series graphics cards)
VideOCR-GPU-v1.3.2-CUDA-12.9-setup-x64.exe (Nvidia 16 - 50 Series graphics cards)
- CPU version:
- Linux:
- CPU version:
VideOCR-CPU-v1.3.2-Linux.7z - GPU version:
VideOCR-GPU-v1.3.2-CUDA-11.8-Linux.7z (Nvidia 10 Series graphics cards)
VideOCR-GPU-v1.3.2-CUDA-12.9-Linux.7z.001 (Nvidia 16 - 50 Series graphics cards)
VideOCR-GPU-v1.3.2-CUDA-12.9-Linux.7z.002 (Nvidia 16 - 50 Series graphics cards)
- CPU version:
VideOCR v1.3.1
Highlights
- Add Blackwell support
- Add support for .ts and .m2ts files
- Add downscaling step before performing OCR (increases performance drastically on high resolution videos)
- Add feature to save crop box selection
- Add multilingual support for the UI and the Installer
- Add hardware support check
- Enhance Subprocess Error Handling and Diagnostic Logging
Fixes:
- Fixes text output for right to left languages
- Fixes taskbar progress bug
Highlight of the v1.3.x release
- The main highlight is the newly added support for PaddleOCR 3.x.x with greatly improved OCR capabilities in more than 40 languages,
including English, Chinese, Japanese, French, Spanish, Portuguese, Russian, Korean, etc.
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- Due to bug with the 10 Series with CUDA 12.8 the GPU version requires an Nvidia graphics card with compute capability between 7.5 and 12.0 (16 Series or newer) and driver version 527.41 (Linux: 525.60.13) or higher!
- If you are using a 10 Series card please check out the GPU version of the v1.3.0 release for now.
- Linux version requires glibc 2.35 or newer (can be checked by running
ldd --versionin a terminal)
Download Tips:
VideOCR v1.3.0
This release adds a lot of new features and improvements.
- The main highlight is the newly added support for PaddleOCR 3.x.x with greatly improved OCR capabilities in more than 40 languages,
including English, Chinese, Japanese, French, Spanish, Portuguese, Russian, Korean, etc.
Further highlights
- Add support for videos with variable frame rate
- Add support for videos with a start time offset
- Add Dual Zone OCR support
- Adds a post processing step to split wrongfully merged words
- Deprecated parameters similar image threshold and similar pixel threshold in favor of SSIM threshold. This new parameter enables speed improvements of over 500% for certain videos. More info about it can be found in the parameters section.
- Adds Taskbar progress indication for Windows
- Adds the ability to send notifications if the OCR process finished
- Add subtitle position parameter
- Add minimum subtitle duration parameter
- Adds an About Tab
Fixes:
- Fix sub sorting behaviour if there are multiple lines
- Fix video box not correctly scaling on Linux
- Disables angle classification by default
- Sets a unique app identifier
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card (10 Series or newer) and driver version 522.25 (Linux: 520.56.06) or higher!
- Nvidia 50 Series is currently not [yet] supported!
- Linux version requires glibc 2.35 or newer (can be checked by running
ldd --versionin a terminal) - This release deprecates the old CLI in favor of CLI SA
- Windows releases were updated two days after the initial v1.3.0 release to fix a missing dependency on systems that did not have the Visual C++ Redistributable Packages for Visual Studio 2013 installed.
Download Tips:
VideOCR v1.2.1
- This release adds a new parameter called "Max Merge Gap". More info about it can be found in the parameters section.
- Several bugfixes are also included.
Fixes:
- Subtitles disappeared one frame too early
- Subtitles which where visible until the end of the video or the specified time had a wrong srt timestamp ending
- Crop box was not applied if it extended to the top left corner
- Crop box was not completely seen when it extended all the way to the right and bottom
- Icon was displayed blurry in the taskbar on Windows
- Update tooltip for brightness threshold
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card (10 Series or newer) and driver version 522.25 (Linux: 520.56.06) or higher!
- Linux version requires glibc 2.35 or newer (can be checked by running
ldd --versionin a terminal)
Download Tips:
VideOCR v1.2.0
This release adds initial support for Linux.
- Requires glibc 2.35 or newer (can be checked by running
ldd --version)
Fixes:
- Subtitle numeration started at zero and not at one
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card and driver version 522.25 (Linux: 520.56.06) or higher!
Download Tips:
VideOCR v1.1.1
This release brings a few handy features and some bugfixes with it!
Changes:
- Subtitles will now by default be saved in the same directory as the selected video file
- When browsing for a file location, the file name of the video is suggested as the name of the subtitle file by default
Fixes:
- Popups appeared in the wrong position when the window was maximized
- Text from settings in the advanced settings tab was truncated on displays with WQHD and DPI-Scaling set to 1.25
Notes:
- No further changes to CLI versions since v1.1.0
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card and driver version 522.25 or higher!
VideOCR v1.1.0
This new release comes with an easy-to-use GUI!
Everything can now be done via a few clicks instead of needing to tinker with the command line!
Notes:
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card and driver version 522.25 or higher!
VideOCR v1.0.0
First release
- Utilizes standalone version of PaddleOCR from https://github.com/timminator/PaddleOCR-Standalone
- The GPU version requires an Nvidia graphics card and driver version 522.25 or higher!