The C3VOC creates lecture recordings from German hacker/tech conferences. In the past we have used dvswitch very successfully but it has some serious limitations. Therefore we started looking for a replacement in 2014. We tested snowmix and gst-switch and while both did some things we wanted right, we realised that no existing tool would be able to fulfil all our wishes. Furthermore both are a nightmare to extend. So we decided to build our own implementation of a Live-Video-Mixer.
The Voctomix project consists of three parts:
- Voctocore, the videomixer core-process that does the actual video- and audio crunching
- Voctogui, a GUI implementation in GTK controlling the core's functionality and giving visual feedback of the mixed video
- Voctomix Example Scripts, a collection of tools and examples for talking to the core-process, feeding and receiving video-streams and controlling operations from scripts or command-line.
Voctomix requires a fairly recent Version of GStreamer (at least 1.5, though we recommend 1.6 and later). This is natively present on Debian Stretch or Sid and Ubuntu Wily. On these systems it should run out of the box. We recommend using one of them. A Docker image that uses Ubuntu Wily as a base is bundled with Voctomix. Please refer to the seperate readme how to use the Docker image.
Install the required dependencies:
# Requirements apt-get install gstreamer1.0-plugins-bad gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-ugly gstreamer1.0-tools libgstreamer1.0-0 python3 python3-gi gir1.2-gstreamer-1.0 gir1.2-gst-plugins-base-1.0 # Optional for the Example-Scripts apt-get install python3-pyinotify gstreamer1.0-libav rlwrap fbset
For the GUI you'll -- additionally to a gnome-desktop -- need to install the following dependencies:
apt-get install gstreamer1.0-plugins-bad gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-ugly gstreamer1.0-alsa gstreamer1.0-tools libgstreamer1.0-0 python3 python3-gi python3-gi-cairo gir1.2-gstreamer-1.0 gir1.2-gst-plugins-base-1.0 gir1.2-gtk-3.0
Now you should be able to clone the git-repository and run Voctomix or the GUI like this:
git clone https://github.com/voc/voctomix.git cd voctomix ./voctocore/voctocore.py -vv ./voctogui/voctogui.py -vv
Installation on Jessie
Because we are using Debian Jessie as our production system (which only has 1.4 packaged), we are packaging the required libraries in our own Debian repository. The packages inside this repository are built against deb-multimedia.org, so to use them you should add the following lines to your
deb http://www.deb-multimedia.org jessie main non-free deb http://c3voc.de/voctomix jessie non-free
You'll then need install the GPG keys:
apt-get update apt-get install deb-multimedia-keyring curl https://c3voc.de/voctomix/gpg-key.asc | apt-key add - apt-get update
Now proceed as described under Installation.
Quickstart using Docker
Run the core and two example source streams
docker run -it --rm --name=voctocore c3voc/voctomix core docker run -it --rm --name=cam1 --link=voctocore:corehost c3voc/voctomix gstreamer/source-videotestsrc-as-cam1.sh docker run -it --rm --name=bg --link=voctocore:corehost c3voc/voctomix gstreamer/source-videotestsrc-as-background-loop.sh
Run the GUI
xhost +local:$(id -un) touch /tmp/vocto/configgui.ini docker run -it --rm --name=gui --env=gid=$(id -g) --env=uid=$(id -u) --env=DISPLAY=:0 --link=voctocore:corehost \ -v /tmp/vocto/configgui.ini:/opt/voctomix/voctogui/config.ini -v /tmp/.X11-unix:/tmp/.X11-unix -v /tmp/.docker.xauth:/tmp/.docker.xauth c3voc/voctomix gui
show more commands availabe
docker run -it --rm c3voc/voctomix help docker run -it --rm c3voc/voctomix examples
A word on CPU usage
Voctomix requires a fair amount of CPU time to run in the default configuration of 1920×1080 at 25fps. Our production systems have these CPUs:
Intel Core i7-3770 CPU 4x 3.40GHz but we're also experimenting with newer ones like these:
Intel Core i7-6700K, 4x 4.00GHz.
For testing and development you may want to use a
config.ini that reduces the resolution and also turns off the JPEG preview encoders, which take a huge amount of the required CPU power and are not required, as long as the GUI and the Core run on the same machine (or have a 10GE Link between them, FWIW). Don't forget to modify your source scripts to provide the correct resolution.
Such a config.ini might look like this:
[mix] videocaps=video/x-raw,format=UYVY,width=320,height=180,framerate=25/1,pixel-aspect-ratio=1/1 [previews] enabled=false videocaps=video/x-raw,width=320,height=180,framerate=25/1
A word on running in a VM
While the Core runs fine inside any VM, the GUI uses OpenGL to display the video streams. Don't forget to enable 3D acceleration in your VM to support this.
To get in touch with us we'd ask to join
#voctomix on the hackint IRC network, mail to
voc AT c3voc DOT de or meet us on one of the many conferences we're at.
You may also want to watch MaZderMinds talk about Voctomix and a presentation some of us gave about C3VOCs video infrastructure.