Av1transcoder

Latest version: v0.4.2

Safety actively analyzes 710445 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 2 of 2

0.1.3

- Fixed broken Two-pass log file moving.
- Fixed skipping the first pass of two-pass encodes, if that is already finished.
- Always overwrite aborted single-pass or second-pass encodes.

0.1.2

- Gracefully handle missing input files
- Enable -notation in the argument parser to load arguments from files
- Move first pass encoder logs to the completed temporary directory, so that completed first passes can be skipped.

0.1.1

- Replaced -2/--two-pass with -1/--single-pass. Two-Pass encoding is the new default.
- Per default, use 8 concurrent encodes instead of 2
- Removed tile usage from the default encoder parameters, due to possible frame curruptions
- Removed threading support from the default encoder parameters,
because multiple encoder instances scale more consistently.

0.1.0

- Initial version.

-Implemented features:
- Command line interface
- Transcode Videos to AV1 using libaom-av1
- Single-Pass and Two-Pass encoding mode (currently defaulting to Single-Pass)
- Scene cut detection with customizable threshold
- Scene length control by providing a lower bound to the individual scene length.
Shorter detected scenes will be merged to improve encoding efficiency
- Scene-parallel encoding using multiple ffmpeg instances
- Concurrency control by a configurable upper limit of parallel instances
- Deinterlacing support using the yadif filter.
- Passing raw, custom encoder parameters to ffmpeg, both global ffmpeg options and encoder-specific options
- Resumable encodes: Finished scenes are collected in a specific directory. If the encoding is aborted and
restarted, already present data is skipped when the encoding process is restarted.
- Customizable data directories
- Customizable ffmpeg binary location and name
- Integration with Cutelog for log display purposes
- Dumping of parsed scenes for inspection purposes
- Optional dumping of generated ffmpeg command lines for inspection purposes

- Limitations and missing features:
- The output video name is automatically generated and currently can’t be changed.
- The output will always be muxed into a matroska container. Raw stream output or other containers are unsupported.
Use mkvextract or the muxer of your choice to remux the output video.
- Due to the output file handling, encoding multiple videos with names differing only in the extension
(e.g. "video.mp4" and "video.avi") will cause a naming conflict and abort the encoding process
in the final step, if central data storage is used (with --output-dir or --temp-dir).
The encoding in this situation should be resumable, if the videos are processed individually.
- If a central data storage is specified, (with --output-dir or --temp-dir),
files with the same name but originating from different source directories will collide in the same way
as in the point above.
- A maximal scene length cannot be specified. This is planned, but the algorithm for this is not yet designed.
- No GUI

Page 2 of 2

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.