MLT Framework

Our newest version 7.36.0 makes it possible to do 10- and 12-bit video processing end-to-end on the CPU with many (not all) filters & transitions supported. Plus now it can do all processing in linear color on the CPU. (Movit/GPU already had these.) mltframework.org/blog/v7.36.0_r…

Version 7.32.0 released last month, mostly bug fixes but also adding some support for HLG HDR. (Previous version improved support for end-to-end 10-bit Y'CbCr on CPU when using small subset of effects). mltframework.org/blog/v7.32.0_r…

Version 7.26.0 released with more fixes on FFmpeg 7 plus the addition of: subtitles, dropshadow filter, and LV2 & VST2 audio plugins. mltframework.org/blog/v7.26.0_r…

Some new things now in our default git branch for the next version: support for LV2 audio plugins and new filters to read SRT subtitle from file or string property and render them using a Qt or GTK (Pango) text filter.

Mind-blowing reveal: Synaptic Dominion A visual story completely by AI! @OpenAI's ChatGPT: story AND image-prompts Visuals: @NightcafeStudio @elevenlabs lent a voice. Aeneas @acutebit got it all in sync @MLTframework melted it all together.

New version 7.20.0 has a nice set of fixes especially the SDL2 consumer plus new stream index properties for the avformat (@FFmpeg) producer that are relative to the stream type (astream=1 for the second audio stream) mltframework.org/blog/v7.20.0_r…

New version 7.18.0 has some very good fixes plus 2 new audio track/playlist filters to remove clicks and crackles around edit points. mltframework.org/blog/v7.18.0_r…

📼 Frei0r v2.3.0 was just released Major bug fixes to various filters that should make life easier for users by fixing various crashes detectable by using a new ASAN test suite. 🎷🐛 Download this release here: 🔗 github.com/dyne/frei0r/re…. Enjoy!

New version 7.16 improves Movit (OpenGL effects) integration and adds 10-bit video support. It also fixes some serious regressions in previous version while improving both interlace and deinterlace. Plus @FFmpeg 6 and #Glaxnimate 0.5.3 mltframework.org/blog/v7.16.0_r…

Shout-out to @MLTframework devs who've been doing amazing work lately improving GPU support, 10 bit color, de-interlacing and more. 💙

New version 7.14! Some highlights: FFmpeg v6 support, Glaxnimate 0.5.2 - also with Qt 6 - Movit love, audio resampler links (fixes artifacts at cuts) mltframework.org/blog/v7.14.0_r…

$ docker run -it --rm mltframework/melt -version melt 7.12.0 Copyright (C) 2002-2022 Meltytech, LLC <mltframework.org> This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

New version 7.10.0 mltframework.org/blog/v7.10.0_r…

Released new version 7.8.0 with a new glaxnimate module. This requires explicit opt-in via CMake and a git checkout using --recurse-submodules to build with this module. mltframework.org/blog/v7.8.0_re…

We released version 7.6.0 today. This version is mainly about completing things: more plugin metadata/docs, end-to-end full color range option, and greater usage of slice threading for image processing especially with @FFmpeg version 5. mltframework.org/blog/v7.6.0_re…

Frei0r.dyne.org version 1.8.0 is out. I believe that our community project is today the biggest collection of free and open source video effects in the world. It is widely used even beyond entertainment applications, i.e for medical visualizations.

slice threading for image scaling and converting pixel format and colorspace with @FFmpeg v5 just landed in our git master branch for next MLT and @shotcutapp releases

Today, we released a new version 7.4.0 that adds property animation to the avfilter! Thank you to @BMahol from @FFmpeg for the help. mltframework.org/blog/v7.4.0_re…

We have released a new version 6.26.0 with cross platform hardware accelerated decoding. This is the last planned major version 6 released before 7, which breaks API mltframework.org/blog/v6.26.0_r…

MLT version 6.24.0 has been released that includes mostly bug fixes but also some new filters: choppy, pillar_echo, qtcrop, and rich text for qtext. Many improvements to the CMake build. mltframework.org/blog/v6.24.0_r…