Generate separate lower bitrate ogg proxy audio files for the Blender VSE.

My camera records audio as 16b PCM embedded inside the recorded MOV file. I am using NFS & SFTP and editing videos over a network. Sometimes Wifi. While the proxy video files work great, Blender still streams the full audio from the original source.

So I decided it would be best to generate proxy audio sources. Using ffmpy I can go through all audio clips in the scene and create smaller separate audio files.

By default, these are stored in the BL_proxy project proxy directory under a new audio folder as an ogg. Both the path of the proxies and the output format are configurable in the render panel.

This plugin adds an option to the Strip menu in the VSE for "Audio Proxy". After you have loaded all your sound clips into the scene you can go into the menu and select "Create Audio Proxies". This will freeze Blender but in the background ffmpy is creating the proxy files. After it is done the sound clips in the scene will now have a custom property that stores the location of the original source and the new proxy. You can use the options in the menu "Use Proxy" or "Use Original" to switch between them. When a render begins it will automatically switch back to the originals.

You will need to install ffmpy with pip so blender can use FFMpeg. Blender is uses Python 3 for plugins so you will need to specify that when installing.

1. Install FFMPEG on your system
2. Download get-pip.py
3. Run get-pip.py with the version of python blender is using. Likely sudo python3 ./get-pip.py
4. Use that pip install to install ffmpy. Likely sudo pip3 install ffmpy

*Note: Only tested on Ubuntu