A simple tool for recording podcast interviews (or the sound from meetings) on Linux.
Simply specify the microphone device to be used to record your voice,
the playback monitor device to be used to record your subject's voice,
and click the Record button. PulseCaster saves the combined result to
the convenient and lightweight OGG Vorbis format, ready for publication.
You can also record your voice separately from the subject's voice, and optionally use the FLAC format, for your post-production needs.
PulseCaster has been updated to require Python 3. It will no longer run on Python 2. If you must use Python 2, please use a release prior to version 0.9.
If you are looking at the source, 'cd' to the top of this project and then run the following command to try it out:
$ cd pulsecaster
$ ./pulsecaster/pulsecaster
The code currently contains a very hacky function to allow you to record to FLAC (the Free Lossless Audio Codec) instead of Ogg Vorbis, which is the default. To turn that capability on, run this command:
$ gsettings set org.pulsecaster.PulseCaster codec flac
To switch back to Vorbis:
$ gsettings set org.pulsecaster.PulseCaster codec vorbis
There's an additional function for setting audio rate (default is 48000 Hz):
$ gsettings set org.pulsecaster.PulseCaster audiorate 44100
$ gsettings set org.pulsecaster.PulseCaster audiorate 48000
The easiest way to use this application is to simply install it using your platform's preferred tool set. To install it using Fedora, run the folowing command:
dnf install pulsecaster
To install it on another flavor of Linux, check the documentation for your particular distribution.
To install directly from this source code, use the handy "distutils" script that's provided:
$ python setup.py build
$ python setup.py install
Refer to the wiki at http://pulsecaster.org/ for a full list of dependencies and requirements.
Translation is done via Transifex: https://www.transifex.com/stickster/pulsecaster/dashboard/
The pipeline for capturing from a running PulseAudio source:
gst-launch pulsesrc device-name='<NAME>' \
! vorbisenc quality=0.5 \
! oggmux \
! filesink location=foo.ogg
Thanks to Harry Karvonen for his Python ctypes-based bindings for PulseAudio. (These are now removed in favor of the pulsectl module.) Thanks also to Jürgen Geuter for helping me understand distutils and contributing some fixes.