Skip to content

Commit

Permalink
Add README and LICENSE.
Browse files Browse the repository at this point in the history
  • Loading branch information
MaxPower15 committed Feb 10, 2024
1 parent d3e3e92 commit a004c45
Show file tree
Hide file tree
Showing 2 changed files with 61 additions and 0 deletions.
26 changes: 26 additions & 0 deletions LICENSE.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
MIT License

Copyright (c) 2024 Max Schnur, Wistia, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

Attribution:
This technology was developed by Max Schnur at Wistia, Inc. Wistia
Max Schnur on GitHub: https://github.com/MaxPower15
Wistia Website: https://www.wistia.com

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
35 changes: 35 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
# Seamless AAC Split and Stitch Demo

This repo demonstrates calculations and ffmpeg commands to encode portions of an audio file with the AAC codec and to recombine them without transcoding.

The general rule is that, when choosing your audio segment sizes, they _need_ to be aligned with AAC frame boundaries. With aligned frame boundaries, we can use the concat demuxer to cut out the silence ffmpeg adds, as well as some extra padding we add to account for AAC's dependency on previous frames.

This tech is important because it allows faster and more efficient cloud rendering. It may also be used, for example, to render and mux individual HLS segments (TS files) independently of the full file.

I've add more comments and explanations in the code itself.

## Requirements

This demo assumes ffmpeg is installed and compiled with support for the libfdk_aac codec. It also assumes you have a modern version of ruby installed.

Some versions of ffmpeg (around 5) may not work properly as there was a temporary regression with aac concatenation. ffmpeg 6 seems to work well. The author's build config looks like this:

```
ffmpeg version 6.0 Copyright (c) 2000-2023 the FFmpeg developers
built with Apple clang version 15.0.0 (clang-1500.0.40.1)
configuration: --prefix=/Users/maxschnur/.asdf/installs/ffmpeg/6.0 --enable-gpl --enable-libass --enable-libfdk-aac --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxvid --enable-libzimg --enable-nonfree --enable-openssl --enable-shared
```

## Usage

Split and stitch:

ruby run.rb

You can find all artifacts in the "out" directory:

ls out

On a Mac, you may want to examine out/stitched.mp4 with a visualization program. The author uses [Audacity](https://www.audacityteam.org/) for that.

open -a Audacity out/stitched.mp4

0 comments on commit a004c45

Please sign in to comment.