- Newest
- Most votes
- Most comments
Use FFmpeg for Post-Processing If AWS MediaConvert continues to produce files with the "video/x-m4v" MIME type, you can use FFmpeg to re-mux the file and ensure it has the correct MIME type:
ffmpeg -i input.m4v -c copy output.mp4
These settings on MediaConvert:
It will give the following result:
Thanks for your help, although it didn't solve my problem directly (Linux still sees this file as "video/x-m4v"), it did make me wonder about changing the "major brand" field to something else and eventually figuring out the solution which I posted.
I figured out the solution.
With just the default settings, with the AWS Generated MP4 video, the "Major Brand" in the metadata comes back as:
Major Brand : MP4 v2 [ISO 14496-14]
Which Linux's "file" command sees as a mimeType of "video/x-m4v"
In Media Convert's Output Settings -> MPEG-4 container settings, if you change the Major Brand field to "isom", the resulting AWS generated video has "Major Brand" of:
Major Brand : MP4 Base Media v1 [IS0 14496-12:2003]
Which Linux's "file" command sees as a mimeType of "video/mp4", which will be accepted by the program I'm trying to use them on.
The default settings for AWS MediaConvert jobs may streamline transcode configuration, though may not provide specific file types and formats. You can change output file types and extensions by forcing in the job. See the previous configuration screen capture for guidance.
There should be no need to reprocess an output from MediaConvert with FFMPEG to achieve what the desired results. Documentation for MediaConvert details the output file formats. https://docs.aws.amazon.com/mediaconvert/latest/ug/reference-codecs-containers.html
Yes, this does work, but defeats the whole point of using AWS to convert media. If I'm going to have to run it through ffmpeg, I might as well do everything I need to do with ffmpeg locally and bypass AWS entirely.