Skip to main content

Filter by idea status

10000+ Ideas

MikeOverbeckParticipant

Can we have headless script execution?Open for Voting

It would be great to be able to collect a project from the Commandline. It would be great to collect data from a project to a json file without having to pop open the UI for the app. It would be great to be able to swap media without having to save out and string parse an .aepx file.How would one automate this?aerender already works headlessly, but is not useful for anything other than rendering.That’s my human request. I’ll attach my AI one here for the bots:  ---  Summary  After Effects lacks a documented, reliable command-line interface for headless script execution. This is a significant barrier to integrating After Effects into professional VFX pipelines, where every other  major DCC application provides this capability as a first-class feature.  ---  The Problem  In a VFX production pipeline, artists work across multiple applications — Nuke, Maya, and Houdini — that are managed by pipeline tools responsible for:  - Opening projects and collecting metadata (shot info, asset references, frame ranges)  - Relinking or versioning media without human interaction  - Validating projects before publish  - Automating project creation from pipeline templates  After Effects cannot participate in these workflows on equal footing. The only officially supported CLI tool, aerender, is limited strictly to rendering. There is no documented, supported mechanism to open a   project headlessly and execute a script against it.  ---  How Competing DCCs Handle This  The Foundry Nuke  Nuke provides a -t flag for headless terminal execution with full access to its Python API:  nuke -t myscript.py  A pipeline script can open a .nk project, read every node, relink footage, and save — all without launching the GUI. This is officially documented and widely used in production.  Autodesk Maya  Maya provides a dedicated headless executable:  mayabatch -script myscript.py  The full Python and MEL APIs are available in batch mode. Pipeline tools routinely use this to open scene files, query asset references, swap file paths, and publish — entirely without a display.  SideFX Houdini  Houdini provides hython, a standalone headless Python interpreter with the full hou module available:  hython myscript.py  Pipeline tools use hython to open .hip files, inspect node networks, update file references, and save — no GUI required.  In each case, headless script execution is officially documented, stable, and a core part of how these tools integrate into production pipelines.  ---  Requested Features  1. Extend aerender to support running an arbitrary ExtendScript — aerender already runs headlessly and is deployed on render farms, making it the ideal candidate for this capability. Adding support for  executing a user-provided script would allow pipeline tools to perform project introspection, media relinking, and other asset management tasks in the same headless, farm-friendly environment already used  for rendering.  ---  Impact  Studios integrating After Effects into professional VFX pipelines are currently forced to either maintain fragile workarounds, limit AE's role in the pipeline, or exclude it entirely in favor of tools with  proper automation support. Extending aerender with script execution would make After Effects a viable first-class citizen in production pipelines with minimal development investment, building on  infrastructure that already exists.  ---

Workflow for correcting AI transcripts using text windowOpen for Voting

Workflow for correcting AI transcripts using text windowCan anyone suggest an efficient workflow for correcting the AI transcripts in Premiere Pro. I've found that correcting the transcript by having to click into the text window is cumbersome when there's lots of correcting to be done. Ideally I'd like to be able to play the source clip while having all of the text editable so that I can edit it on the fly. I'm a moderately adequate typist, and it took me all day yesterday to correct one hour of interviews (filmed in a noisy conference room)I don't think I have the best work flow for correcting text, as I find myself having to stop playback, enter the text window, correct text, then step out again to hear the text again if there's lots of typing to do, and then go back in again to clean it up.  This is fine for the occasional text correction, but very cumbersome when there's lots of correcting to do In the past, on Final Cut Pro, I was able to type text into a text window (for on screen captions) while also having the clip play independently, so I wasn't constantly stepping into & out of the text editing mode; rather I had both play back & text editing functionality occurring concurrently. I'd love to hear of any workflows that do this better (or alternatively, an external transcription tool that I could use when I need to actually transcribe my clips myself, and import the transcription back into premiere pro)